CN103033752B - Method for predicting and prolonging service life of battery of electric vehicle - Google Patents

Abstract

The invention provides a method for predicting the service life of a lithium battery of an electric vehicle, which predicts the service life of the battery according to the acquired working temperature and charging and discharging frequency of the battery and the ratio of the energy of the battery required by a driver during one-day driving to the energy of the battery after the battery is fully charged just before the delivery of the battery. The invention also provides a method for predicting the service life of the lithium battery of the electric vehicle, which comprises the step of predicting the service life range of the battery according to the acquired character parameters of a driver, the charging strategy, the ambient temperature and the driving distance on the expressway and the urban road on the weekdays and weekends. The invention also provides a method for prolonging the service life of the lithium battery of the electric vehicle according to the formula. The method provides a basis for accurately predicting the service life of the lithium battery; the service life of the lithium battery of the electric vehicle is predicted by considering the behavior characteristics of the driver, so that the prediction result is more real; the service life of the lithium battery can be predicted without hardware, and the cost is very low.

Description

Method for predicting and prolonging service life of battery of electric vehicle

Technical Field

The invention relates to the field of electric vehicles, in particular to a method for predicting the service life of an electric vehicle battery and a method for prolonging the service life of the electric vehicle battery.

Background

Current environmental issues and possible oil production issues have prompted the development of various electric vehicles. Compared with the traditional automobile, the electric vehicle can play an important role in reducing pollutant emission and energy consumption. In an electric vehicle, the life and cost of an on-board battery directly affect the performance, life and cost of the electric vehicle, and predicting the life of the battery has become an important issue today.

First, in the existing research, researchers have only experimentally demonstrated the qualitative relationship between battery life and related factors, and have not provided mathematical quantitative predictions for battery life.

In addition, most of the existing research on battery life focuses on physical and chemical processes of the battery, such as state of charge (SOC), operating temperature, and number of charge and discharge, but neglects the influence of the driver. Since the driver acts as an operator of the electric vehicle and directly affects the battery life, the battery life of the electric vehicle cannot be predicted comprehensively and truly without considering or not considering the behavior characteristics of the driver sufficiently.

In summary, a mathematical method for predicting the battery life of an electric vehicle by comprehensively considering the relevant factors of the battery life is lacked in the prior art, and in addition, a method for predicting the battery life of an electric vehicle by comprehensively considering the behavior characteristics of a driver is lacked in the prior art.

Disclosure of Invention

The invention aims to provide a method for predicting the service life of an electric vehicle battery, which is used for mathematically and quantitatively predicting the service life of a lithium battery of the electric vehicle.

In order to solve the above problems, the present invention provides a method for predicting a lifetime of a lithium battery of an electric vehicle, comprising:

the method comprises the following steps of firstly, obtaining the working temperature and the charging and discharging frequency of a battery and the ratio of the battery energy required by a driver in one-day driving to the energy of the battery after the battery is fully charged when the battery leaves a factory; and

step two, predicting the service life of the battery according to the following modes:

$\mathrm{Capacity}(\%)=\frac{1}{100}\·(167.583-1.264T-0.097\·t\·f)\·(84.1+3.71\·f-0.0788\·t\·f)$

wherein T represents the working temperature of the battery, and the unit is; t represents the elapsed time, namely the elapsed time length from the factory delivery of the battery to the calculation of the number of battery cycles, and the unit is day; capacity (%) represents a ratio of the amount of electricity after the battery is fully charged to the amount of energy after the battery is fully charged immediately after the battery is taken out of the field; f represents a charge-discharge frequency;

if the battery life is over when the electric quantity after the full charge of the battery can not provide the energy required by the automobile for one-day driving, the elapsed time T is calculated according to the charging frequency f at the critical point, the working temperature T of the battery, and the ratio Capacity (%) of the battery energy required by the driver for one-day driving and the energy after the battery is fully charged just before the factory, and the value represents the predicted life of the battery.

Optionally, the operating temperature of the battery is calculated according to the following manner:

$T=T_{\mathrm{initial}}+0.0339+0.004t^{\′}+0.001\·C+2.385(\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}-\mathrm{\Δheat})-0.004\·C_p$

wherein, T_initialRepresents the initial temperature of the battery during operation, and the unit is; t' represents a continuous working time of the battery, and the unit is second; c represents a battery discharge rate;represents the heat generation rate of the battery pack in watts; Δ t represents the time interval in seconds for the relevant instrument to record driving data; n represents the number of cells in one battery pack; Δ heat represents the difference in thermal decay produced by other types of cooling modules relative to the air cooling module, in W/cell; c_pThe heat capacity of the battery is expressed in J/kg/K.

The invention aims to solve another technical problem of providing a method for predicting the service life of a battery of an electric vehicle, which considers the behavior characteristics of a driver and comprehensively and truly predicts the service life of the lithium battery of the electric vehicle.

In order to solve the above problems, the present invention provides a method for predicting a lifetime of a lithium battery of an electric vehicle, comprising:

step one, acquiring personality parameters of a driver, a charging strategy, an environment temperature, and driving distances of a working day and weekends on an expressway and an urban road;

step two, predicting the service life range of the battery according to the following modes:

Lifetime＝pa+pb×Personality+pc×Charging+pd×T_initial

+pe×D_hd+pf×D_ud+pg×D_he+ph×D_ue

wherein, Personality represents the Personality parameter of the driver, and the quantization mode of the Personality parameter of the driver is as follows: non-impulse type, normal and impulse type are represented by-1, 0 and 1 in sequence; charging represents a Charging strategy, namely the minimum amount of electricity left in the battery before the battery is charged, and the value is a percentage; t is_initialRepresents the ambient temperature; d_hdAnd D_udIndicating driving distance on weekdays on highways and urban roads, D_heAnd D_ueRepresenting the driving distance on weekend expressways and urban roads, with the unit of mile; pa is a constant, pb-ph represents a coefficient;

wherein pa ranges from 393.7864 + -147.2502, pb ranges from-35.1552 + -21.9882, pc ranges from-2.85 + -2.145, pd ranges from-1.3822 + -1.1946, pe ranges from-9.716 + -2.7264, pf ranges from-2.5916 + -3.2112, pg ranges from-4.3606 + -2.7264, and ph ranges from-11.533 + -4.9872.

Optionally, the formula of step two is:

Lifetime＝393.768-35.1552Personality-2.85Charging-1.382T_initial。

-9.716D_hd-2.592D_ud-4.361D_he-11.533D_ue

it is yet another object of the present invention to provide a method for extending the life of a lithium battery of an electric vehicle by optimizing battery configuration and optimizing driving and charging behavior.

In order to solve the above problems, the present invention provides a method for prolonging a lifetime of a lithium battery of an electric vehicle, comprising:

step one, the following modifications are made to the formula in the method of claim 2:

$\left\{\begin{array}{c}\frac{1}{100}\·(167.583-1.264T-0.097\·t\·f)\·(84.1+3.71\·f-0.0788\·t\·f)*\mathrm{CAPACITY}-E=0\\ T=T_{\mathrm{initial}}+0.0339+0.004t^{\′}+0.001\·C+2.385(\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}-\mathrm{\Δheat})-0.004\·C_p\end{array}\right.$

wherein CAPACITY is the energy of the battery pack after being fully charged just before the battery pack leaves the factory, and the unit is J; e is the total energy consumption during one day of driving, and the unit is J; t is the elapsed time, here representing the target life of the battery, in days;

step two, acquiring battery energy E consumed by driving every day, deciding a reference value DMR and charging frequency;

step three, solving the formula to obtain the optimal configuration of the battery, namely C_pΔ heat and N.

In order to solve the above problems, the present invention further provides a method for prolonging a lifetime of a lithium battery of an electric vehicle, comprising:

step one, the following modifications are made to the formula in the method of claim 2:

$\left\{\begin{array}{c}\frac{1}{100}\·(167.583-1.264T-0.097\·t\·f)\·(84.1+3.71\·f-0.0788\·t\·f)*\mathrm{CAPACITY}-E=0\\ T=T_{\mathrm{initial}}+0.0339+0.004t^{\′}+0.001\·C+2.385(\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}-\mathrm{\Δheat})-0.004\·C_p\end{array}\right.$

wherein CAPACITY is the energy of the battery pack after being fully charged just before the battery pack leaves the factory, and the unit is J; e is the total energy consumption during one day of driving, and the unit is J; t is the elapsed time, here representing the target life of the battery, in days;

step two, acquiring the battery energy E consumed by driving every day and the battery configuration N, C_pΔ heat, initial temperature T_initial；

Step three, solving the formula to obtain the optimal C,And f;

step four, obtaining the optimal I from the optimal C, and obtaining the optimal P from the following formula:

$\stackrel{\·}{q}=I(U_{\mathrm{ocv}}-U_{\mathrm{op}})={\mathrm{IU}}_{\mathrm{ocv}}-P$

wherein P represents the power consumption of the vehicle battery pack in units of W;represents the heat generation rate of the battery pack, and has a unit of W; i represents the current of the battery pack and has the unit of A; u shape_ocvRepresenting the open-loop voltage, U, of the battery_opRepresents the operating voltage of the battery under load conditions, in volt;

step five, obtaining the optimal driving speed v and the optimal acceleration a through the following formulas:

$p=(\mathrm{ma}+\frac{1}{2}{\mathrm{\ρv}}^2{C}_{d}A+C_{\mathrm{rr}}\mathrm{mg})0.4v;$ $p=\frac{(\mathrm{ma}+\frac{1}{2}{\mathrm{\ρv}}^2{C}_{d}A+C_{\mathrm{rr}}\mathrm{mg})v}{0.8}$

wherein, p represents the motion power consumption of the electric automobile and the unit is W; m represents mass, in Kg; a represents the acceleration of the vehicle in m/s²Rho sea level air weight per cubic meter is approximately Kg/m³(ii) a v represents velocity in m/s; c_dRepresenting a drag coefficient of the vehicle; a represents the frontal area of the vehicle in m²，C_rrDimensionless coefficient representing the rolling resistance of the tyre, g representing the gravitational acceleration in m/s²，

And guiding the behavior of a driver according to the obtained charging frequency f, speed v and acceleration a, thereby prolonging the service life of the lithium battery of the electric vehicle.

Compared with the prior art, the invention has the advantages that:

firstly, a human-electric vehicle model is provided, the model comprehensively considers the behavior characteristics of a driver and other factors influencing the service life of a vehicle-mounted battery, and a basis is provided for accurate prediction of the service life of a lithium battery;

secondly, the service life of the lithium battery of the electric vehicle is predicted by considering the behavior characteristics of a driver, and the prediction result is more real;

thirdly, the service life of the lithium battery can be predicted without hardware, and the cost is very low;

fourth, the conventional method for prolonging the battery life mainly focuses on the battery design itself, and the present invention extends the battery life and obtains the optimal design of the battery through the driving speed, acceleration and charging frequency.

Drawings

FIG. 1 is a schematic diagram of a theoretical model derivation process provided in one embodiment of the present invention;

FIG. 2 is a schematic diagram of an analysis of a correlation of human factors with battery life provided in one embodiment of the present invention.

Detailed Description

Defining:

(1) life schedule of driver: driving time and distance on motorways and cities on weekdays, weekends.

(2) And (3) charging strategy: the minimum amount of charge remaining in the battery before charging the battery is a percentage.

(3) Decision reference value: in driving experiments, the driver needs to choose how many miles per hour above the speed limit he wants to travel, and presented with the options are the corresponding "monetary cost if a ticket is received" and "safety and time gain if no ticket is received". The driver determines a new driving speed when a speed limit sign appears by comprehensively considering the two factors, and the difference value of the new driving speed and the speed limit is a 'decision reference value' with the unit of mile/hour.

(4) Human-electric vehicle experience model: a formula that embodies the relationship between driver behavior characteristics, battery configuration, and battery life.

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Establishing a theoretical model

Based on the above findings, in one embodiment of the present invention, a theoretical model is provided for calculating the lifetime of a lithium battery of an electric vehicle. As shown in fig. 1, a change in the value of the factor at the beginning of the concatenated symbol results in a change in the value of the factor at the end of the concatenated symbol. The establishment process of the model is as follows:

1. connection (H), relationship of daily driving data to energy consumption:

the power required for vehicle motion when the acceleration is negative and non-negative may be described by equations 6 and 7, respectively, as set forth in Peterson (Peterson, s.b., j.apt, and j.whitacre, Lithium-ionbateric de gradiation, and from regenerative-to-gridesign, journal of powersource, 2010.195 (8): p.2385-2392), et al, as follows:

$p=(\mathrm{ma}+\frac{1}{2}{\mathrm{\ρv}}^2{C}_{d}A+C_{\mathrm{rr}}\mathrm{mg})0.4v---\left(6\right)$

$p=\frac{(\mathrm{ma}+\frac{1}{2}{\mathrm{\ρv}}^2{C}_{d}A+C_{\mathrm{rr}}\mathrm{mg})v}{0.8}---\left(7\right)$

wherein P represents the power consumption (unit is w) of the electric automobile movement, m represents the mass (unit is Kg), and a represents the acceleration (unit is m/s) of the automobile²) Rho sea level air approximately air weight per cubic meter (in Kg/m)³) V denotes speed, C_dRepresenting the drag coefficient of the vehicle, A represents the frontal area of the vehicle (unit is m)²)，C_rrDimensionless coefficient representing the rolling resistance of a tire, g represents the gravitational acceleration (in m/s)²)。

The lower case p represents the power consumption of the electric vehicle, and when the acceleration of the vehicle is negative, the formula 6 is used, and when the acceleration is not negative, the formula 7 is used. In the following, the capital letter P indicates the output power of the automobile battery pack. Wherein the battery output power includes the vehicle motion consumption p, and the power consumption of the device independent of the vehicle motion.

According to Peterson et al, one can obtain:

m＝1590kg，ρ＝1.23kg/m³，C_d＝0.28，A＝2.67m²，C_rr＝0.01，g＝9.8m/s²。

then the power and energy consumption may be calculated based on formula 6/7 after the speed/acceleration data for the vehicle is known.

2. Connection (E), energy consumption versus SOC relationship:

SOC represents the state of charge, or state of charge, equal to the amount of charge remaining in the battery divided by the amount of charge when the battery is fully charged.

3. Connection (D), SOC versus voltage relationship:

the battery open loop voltage includes the battery output voltage and the voltage dissipated by the battery internal impedance.

The battery open loop voltage depends in large part on the SOC of the battery, and can be calculated from equation 18:

U_ocv′＝-1.031e^-35×SOC+3.658+0.2156×SOC

-0.1178×SOC²+0.321×SOC³(18)

the battery output voltage can be expressed by equation 19, and the operating voltage of the cell under load is:

U_op′＝U_ocv′-I′Z_eq′(19)

wherein, U_op' denotes the operating voltage (volt in U) of the cell under load conditions_ocv' represents the open-loop voltage (volt) of the cell, I ' represents the current of the cell (I ' > 0 at the time of discharge), and Z_eq' denotes the internal impedance of the battery.

Then, the battery output voltage can be expressed by equation 9:

U_op＝U_ocv-IZ_eq(9)

wherein, U_opRepresents the operating voltage (volt, U) of the battery under load conditions_ocvRepresents the open loop voltage (volt) of the battery, I represents the current of the battery (I' > 0 when discharging), and Z represents the open loop voltage_eqRepresenting the internal battery impedance.

The equivalent internal impedance of the battery in equation 19 is defined by a series resistance (R)_series) Two RC networks (comprising R)_{Transient_s}，C_{Transient_s}，R_{Transient_L}And C_{Transient_L}) These values depend on the SOC of the battery and can be calculated by empirical formulas derived from experiments, as detailed in formula 8:

$\left\{\begin{array}{c}{R}_{\mathrm{series}}=0.1562e^{-24.37\mathrm{SOC}}+0.07446\\ R_{\mathrm{Transient}\_s}=0.3208e^{-29.14\mathrm{SOC}}+0.0469\\ C_{\mathrm{Transient}\_s}=-752.9e^{-13.51\mathrm{SOC}}+703.6\\ R_{\mathrm{Transient}\_L}=6.603e^{-155.2\mathrm{SOC}}+0.04984\\ C_{\mathrm{Transient}\_L}=-6056e^{-27.12\mathrm{SOC}}+4475\end{array}\right.---\left(8\right)$

from equation 8, Z can be found_eq' depends on the SOC of the battery. In addition, it will be understood from the following description that I' is determined by battery power and operating voltage. To sum up, U_op' i.e., the operating voltage of the cells under load depends on the SOC of the battery.

4. Connection (F), energy consumption, voltage versus current relationship:

we can find from the following equation 9 that the current of the battery is determined by the battery power and the operating voltage:

p′＝U_op′I′(9)

wherein the parameter p' represents the output power of the cell, U_op' represents the operating voltage of the cell under load conditions, and I ' represents the current of the cell (I ' > 0 upon discharge).

If we assume that the number of cells connected in series in the battery pack is n, the current value can be obtained by equation 10:

P＝U_op·I＝n·U_op′·I(10)

wherein P represents the output power of the battery pack, U_opIndicating the operation of the battery under loadReference voltage, I represents the current of the battery pack (I > 0 at the time of discharge), n represents the number of cells connected in series in the battery pack, U'_opRepresenting the operating voltage of the cell under load conditions.

5. Connection (C), relationship of voltage, current and heat generated:

the amount of heat generated by the battery depends on the discharge process of the battery, see equation 5:

${{\stackrel{\·}{q}}_{\mathrm{gen}}}^{\′}=I^{\′}({U_{\mathrm{ocv}}}^{\′}-{U_{\mathrm{op}}}^{\′})-I^{\′}T\frac{d{U_{\mathrm{ocv}}}^{\′}}{\mathrm{dT}}---\left(5\right)$

wherein,the heat generated by the single cell (unit is Kg/m3), I 'represents the current of the single cell (I' > 0 during discharge) (unit is A), U_ocv' denotes the open-loop voltage (volt) of the cell, U_op' operating Voltage of Single cell under load (unit is volt), T tableThe operating temperature of the cell (in degrees C.) is shown.

Therefore, we can obtain the heat generation rate of the battery packWhen the open-loop voltage conversion ratio of the battery pack is difficult to obtain, the differential part of equation 5 can be ignored. Thus, the simplified battery pack heat generation rate can be obtained from equation 11:

$\stackrel{\·}{q}=I(U_{\mathrm{ocv}}-U_{\mathrm{op}})={\mathrm{IU}}_{\mathrm{ocv}}-P---\left(11\right)$

wherein,represents the heat generation rate of the battery pack, I represents the current of the battery pack (I > 0 at the time of discharge), U_ocvRepresenting the open-loop voltage, U, of the battery_opRepresenting the operating voltage of the battery under load conditions.

6. Connection (I), relationship between current and discharge rate:

the discharge rate C is used to indicate the rate, i.e., the rate, of the magnitude of the current when the battery is discharged. For a 1200mAh cell, 0.2C indicates a discharge current of 240mA (0.2 rate of 1200 mAh), and 1C indicates 1200mA (1 rate of 1200 mAh).

7. Connection (G), battery generated heat versus battery operating temperature:

high power output lithium batteries generate a large amount of heat, which causes a severe increase in the temperature of the battery, thereby causing damage to the battery, and therefore include a heat management module therein. The thermal management system functions to provide an optimal average temperature and an evenly distributed temperature to the battery pack. The module in the electric vehicle can also keep the temperature of the battery at an optimum average value so that the battery life is not shortened.

The transient temperature rise of the heat generated continuously in a module, which uses an air cooling module, was investigated by Pesaran (Pesaran, a.a., batterthermal management evaluation current, 2001.43 (5): p.34-49). The increased heat generation, discharge rate, or small heat capacity of the battery can produce a greater temperature rise. These explain the connection (G) from the box "heat capacity" to the "operating temperature" and the connection (G) from the box "discharge rate" and "generated heat" to the box "operating temperature".

Therefore, we can obtain the linear relationship between these relevant factors from the data of Pesaran, i.e. the inventor summarizes equation 12:

$T=T_{\mathrm{initial}}+\mathrm{\ΔT}=T_{\mathrm{initial}}+t_a+t_t\·t^{\′}+t_R\·C+t_H\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}+t_C\·C_p$

(12)

$=T_{\mathrm{initial}}+0.0339+0.004t^{\′}+0.001\·C+2.385\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}-0.004\·C_p$

wherein T represents the operating temperature of the battery, T_initialRepresenting the initial temperature of the battery during operation, which can be simplified to ambient temperature, Δ T representing the temperature rise of the battery during operation, T' representing the one-time continuous operation time of the battery, C representing the discharge rate of the battery, Δ T representing the time interval of the driving data recorded by the relevant instrument, N representing the number of batteries in a battery pack, C_pRepresenting the thermal capacity of the cell.

From the experimental results provided by Pesaran, we estimated the regression coefficient t_a，t_t，t_R，t_HAnd t_CValues of (1), they are divided into0.0339, 0.004, 0.001, 2.385 and-0.004, respectively. The R-square value adjusted in the results of linear regression analysis of SPSS was 0.842.

Since Pesaran was tested using only air cooling modules, we assume that different cooling modules will produce different thermal attenuations. We add a term delta heat in equation 12 to represent the difference in thermal decay that occurs between different types of cooling modules relative to air cooling modules, see equation 13. The unit of Δ heat is W/cell.

$T=T_{\mathrm{initial}}+\mathrm{\ΔT}=T_{\mathrm{initial}}+t_a+t_t\·t^{\′}+t_R\·C+t_H(\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}-\mathrm{\Δheat})+t_C\·C_p$

$=T_{\mathrm{initial}}+0.0339+0.004t^{\′}+0.001\·C+2.385(\frac{\mathrm{\Σ}\stackrel{\·}{q}\mathrm{\Δt}}{t^{\′}\·N}-\mathrm{\Δheat})-0.004\·C_p---\left(13\right)$

8. Connection (J), relationship between operating temperature of battery and battery life:

ramadas (Ramadas, P., et al., CapacityfadeOfSony18650cell cycle energy sources, 2002.112 (2): p.606-613.) (Ramadas, P., et al., CapacityfadeOfSony18650cell cycle energy sources, p.606-613.) et al, (Capacityfadeanalysis, JournaalofPowersources, 2002.112 (2): p.614-620.) et al, have made a complete capacity decay analysis of the charge and discharge cycles of the Sony 50 (mAh) lithium battery when the temperature rises 45-55 ℃, which elucidates that they lose the initial charge and discharge capacities at 25 ℃ and 45% at 1800 ℃ and 36% after 800 cycles in the study. The battery loses more than 60% of initial capacity after being charged and discharged for 600 times at 50 ℃, and loses 70% of initial capacity after being charged and discharged for 500 times at 55 ℃.

We can conclude that longer elapsed times and higher operating temperatures result in more capacity loss from the battery. If we express the life of a battery as its capacity after it is fully charged, the life of the battery is lost more quickly at higher discharge temperatures. This can explain the connection (J) from the box "operating temperature" to "battery life".

Based on the Ramadass et al study, we have established equations 14 and 15.

Capacity(％)＝C_b+C_T·T+C_N2N_cycle(14)

LT(％)＝C_b+C_TT+C_N2N_cycle(15)

Wherein, C_b，C_TAnd C_N2Is an indeterminate coefficient, Capacity (%) represents the ratio of the energy of a single cell after being fully charged to the energy immediately after the battery is discharged, T represents the operating temperature of the battery, N represents the operating temperature of the battery_cycleThe number of charge and discharge cycles is shown, and LT (%) shows the ratio of the battery life to the battery life immediately after shipment.

From the data provided by Ramadass we estimate C_b，C_TAnd C_N2And with t · f in the following equations 16 and 17_chargingTo replace N_cycleEquations 16 and 17 are obtained. In the results of the linear regression analysis of SPSS, the adjusted R-square was 0.805 and the P-value was 0.000.

capacity(％)＝167.583-1.264T-0.097·t·f(16)

LT(％)＝167.583-1.264T-0.097·t·f(17)

9. Connection (a), relationship of energy consumption and charge-discharge frequency:

the battery charge will be charged when it decreases to a certain predetermined threshold (related to the nature of the driver, different thresholds for different drivers). Because the energy consumption during driving can be calculated based on the speed and acceleration data of the driver during driving, the residual capacity of the battery after every day of driving can be obtained. Once the remaining charge is less than the predetermined threshold, the driver may choose to charge the battery. Thus, the charging and discharging frequency in a specific time period can be obtained.

The number of charge and discharge cycles can be calculated by the length of time elapsed and the charge and discharge frequency, i.e. the length of time multiplied by the charge and discharge frequency is equal to the number of charge and discharge cycles. The elapsed time is the length of time elapsed from the battery shipment to the battery cycle count calculation.

10. Connection (B), relationship of charge and discharge frequency to battery life:

the decay of the circulating energy and the decay of the stored energy of lithium batteries in mobile phones have been studied by Takeno (Takeno, K., et al, Imperial of circulating and purifying efficiencies and lithium ion batteries. They found that as the charge-discharge cycle length of the battery decreased, the rate of decrease in the battery energy increased. That is, the battery power decreases proportionally with the number of charge and discharge cycles over the same period of time. See equation 1:

capacity(％)＝C_a+C_f·f+C_N1·N_cycle(1)

wherein capacity represents the ratio (%) of the energy after the single cell is fully charged to the energy after the battery is fully charged immediately after the battery is taken out of the field, f represents the charge-discharge frequency, and N represents the charge-discharge frequency_cycleIndicating the number of charge and discharge cycles.

Through the experimental data of Takeno, we estimate the parameter C in the formula 1_a，C_f，C_N1The value of (c). In the results of the linear regression analysis of SPSS, adjusted R²The value of (A) is 0.920 and the P value is 0.000. Wherein R is²For the determination of the coefficients or the determination of the coefficients, defined as the quotient of the regression sum of squares and the total sum of squares, the degree of fit of the regression independent variable to the dependent variable is indicated. The P-value is the probability of the appearance of the sample observation or more extreme result obtained when the original hypothesis was true, with a smaller P-value indicating a more significant result. P < (R) >, a process for preparing the same0.05, the regression model was statistically significant. Since the linear model can well describe the relationship between the battery energy, the charge and discharge time and the number of charge and discharge cycles, the linear model is applied to the battery of the electric vehicle.

If we replace N by t.f_cycleAssuming that the amount of electricity after the battery is fully charged can represent the remaining life of the battery, equation 1 can be rewritten as equation 2:

capacity(％)＝C_a+C_f·f+C_N1·N_cycle(2)

＝84.1+3.71·f-0.0788·t·f

where capacity (%) represents a ratio (%) of a battery life to a battery life immediately after shipment, f represents a charge/discharge frequency, and t represents an elapsed time (i.e., a length of time elapsed from shipment of the battery to calculation of the number of battery cycles).

11. Connection (K) (not shown), current to overcharge/discharge protection, voltage to overcharge/discharge protection, SOC to battery life, overcharge/discharge protection to battery life:

with the popularization of electric vehicles, avoiding overcharge and discharge of a battery and ensuring sufficient SOC of the battery become major factors for ensuring effective use of the battery, and SOC estimation of the battery plays an important role in prolonging the life of the battery and displaying available electric power before a user recharges the battery. Mills (Mills, A.and S.Al-Hallaj, immunology of passive thermal analysis system for lithium-ion batteries, journal of Power sources, 2005.141 (2): p.307-315.) et al investigated the overcharge/discharge of lithium batteries by "soft" overcharge techniques. They found that lithium batteries undergo drastic, irreversible changes after overcharging to 150% capacity. Studies have shown that thermal runaway occurs when the cells are operated under overcharge/discharge conditions and when the heat generation and dissipation are not equal, high overcharge/discharge rates can result in thermal runaway and the generation of large amounts of heat within the cells in short reaction times.

12. Theoretical model for predicting battery life:

as can be seen from the graph, the direct influence factors on the battery life are the charge and discharge frequency, the operating temperature, and the elapsed time. The initial temperature of the battery, the open loop voltage, the operating current, the thermal management module, and the thermal capacitance affect the operating temperature of the battery. The open loop voltage, operating voltage and operating current are in turn influenced by the SOC.

In summary, the charge of the battery can be described by equation 3 obtained by combining equation 2 and equation 16, where T can be obtained from equation 13. The remaining capacity of the battery can be described by the following equation 3.

For formula 3, under the condition that the charging frequency is f and the battery operating temperature is T, after the time T elapses, the ratio of the energy after the battery is fully charged to the energy after the battery is fully charged immediately before the factory shipment can be calculated. We consider that the battery life ends when the full charge of the battery is not enough to provide the energy required by the test (i.e. the test subject in the experiment) for one-day driving; the ratio of the battery energy required by a certain driver for driving a day to the energy of the battery after the battery is fully charged when the battery leaves a factory becomes a critical point, and the critical point can be used as a known quantity to reversely calculate other parameters. Therefore, if the charging frequency is f and the battery operating temperature is T, the ratio Capacity (%) of the battery energy required by a driver for driving a day to the energy after the battery is fully charged just before the shipment of the vehicle can be obtained.

$\left\{\begin{array}{c}\mathrm{Capacity}(\%)=\frac{1}{100}\&CenterDot;(167.583-1.264T-0.097\&CenterDot;t\&CenterDot;f)\&CenterDot;(84.1+3.71\&CenterDot;f-0.0788\&CenterDot;t\&CenterDot;f)\\ T=T_{\mathrm{initial}}+0.0339+0.004t^{\&prime;}+0.001\&CenterDot;C+2.385(\frac{\mathrm{\&Sigma;}\stackrel{\&CenterDot;}{q}\mathrm{\&Delta;t}}{t^{\&prime;}\&CenterDot;N}-\mathrm{\&Delta;heat})-0.004\&CenterDot;C_p\end{array}\right.---\left(3\right)$

In formula 3, T represents the operating temperature of the battery, T represents the elapsed time (i.e., the time period from the shipment of the battery to the calculation of the number of battery cycles), f represents the charging and discharging frequency, and T represents the number of battery cycles_initialRepresents the initial temperature of the battery during the operation process (simplified to the ambient temperature), t' represents the continuous operation time of the battery once, C represents the discharge rate of the battery,indicating the heat generation rate of the battery pack, at indicating the time interval during which the relevant instrument records driving data, N indicating the number of batteries in a battery pack, Δ heat indicating the difference in heat decay with respect to the air cooling module, other types of cooling modules, C_pRepresenting the thermal capacity of the cell.

By this time, the theoretical model (i.e., equation 3) is completed.

The derivation shown in the steps 1 to 11 is specific to the lithium battery, and the objective rules of the parameters related to the service life of the lithium battery are summarized, so that the method has universality, and formula 3 theoretically derived according to the steps 1 to 11 reflects the objective rules of the service life of the lithium battery, and has universality for application of the lithium battery. In the application, the formula is applied to the lithium battery electric vehicle.

In the experiment, the driving speed, the acceleration and the driving time can be acquired. From these values, the energy consumption of the battery of each electric vehicle under the condition of the given distance between the expressway and the urban road can be obtained according to equations 6 and 7. Thus, we can obtain the average power (unit: W/mile) of each tested driving on the expressway and the urban road. The subject will answer his own real driving distance per day in the following questionnaire, including highway distance and urban road distance. From the previously calculated driving average power, it is possibleThe battery energy consumed by each tested driver every day is calculated. After the end of the drive, all the subjects answered "you will charge the battery after the battery level is less than what percent". By the answer of the question, we can calculate how much energy has been consumed before each charge of each tested bit, and the charging and discharging frequency f of each tested bit can be obtained by dividing the energy value by the energy consumed by each test day. We assume that the life of the battery is over when the full charge of the battery has not been able to provide the energy required for the one-day driving trial. The energy value required by driving in one day after the battery is fully charged after leaving the factory is taken as the capacity (%) in formula 3, and it is assumed that the ratio can represent the ratio of the remaining available life of the battery to the battery life of the battery immediately after leaving the factory. Thus, when the battery configuration C, N, Δ heat, C are known_pAmbient temperature T_initialWhen the driving simulator records the time interval delta t of the driving data in real time, the continuous driving time t' each time, the battery life t can be obtained through a formula 3 or a formula 4.

Establishing a human-electric vehicle experience model

Analyzing an empirical model:

the effects of human personality on leg muscle (brake and throttle control) speed control were well established in previous studies, and in various personality libraries, the relationship between muscle speed and personality was essentially identical to the E and N scales in the exxon personality system in different behavioral paradigms. Bachorowski and Newman (Bachorowski, J.A. and J.P.Newman, Impulse motorbeivisor: Effect of muscle performance. journal of muscle performance and of muscle performance, 1990.58 (3): p.512-518.) have reported that persons who are impulsive (characterized by E + and N +) have faster muscle movement speeds in a mission requiring slow control movements than persons who are non-impulsive (scored lower in both dimensions). And this phenomenon is more pronounced when the results of the behavior are revealed. In the study by Wu (C.Wu. and ZHao, G, materialmodelModelingof AverageDriverSpeedControlandVividalDriverDferationWiththe integration of QueuingNetwork-ModelHumanprocessionRule-based determination Field) (2010), etc., they used a personality trait variable to represent the degree of human impulsivity, or the tendency to act rapidly without a deep thought. That is, the character of the driver (impulsive type, normal type, non-impulsive type) affects the driving behavior of the driver.

In the experiment, all subjects completed a series of questionnaires. The first questionnaire was used to obtain the demographic baseline (e.g., age, gender, etc.) and driving history (e.g., approximate cumulative driving range, year in which the driver's license was obtained, etc.) of the subject. They will then build a subjective matrix of value. The value of each element in the matrix represents the value of each driving speed to the test. Presented with each speed selection is the monetary cost of receiving the ticket, and the security and time benefits that would be obtained if the ticket was not received. Finally, all subjects were classified into three categories according to the revised Exxon personality questionnaire (EPQR-S): normal (test results are E + and N-, or E-and N +, for a total of 6); impulsive type (test results are E + and N +, and total 3 persons); non-impulse type (test results are E-and N-, three people in total) (note: Exxon personality test questionnaire has been used throughout the world for more than 35 years, is a standard questionnaire/scale for measuring human personality, has high reliability (high stability of the first and the last tests) and effectiveness (effectively reflects different personality types) for measuring human personality, and can simultaneously obtain the measured results of a plurality of proofs of experimental psychological research, and has been revised in China as early as 1981, for details: HansJ ü rgen Eysenck & SybilB.G.Eysenck (1975).

In real life, the driving speed decided by the driver is often not equal to the speed limit of the road. Thus, in driving experiments, the driver needs to choose how many miles per hour above the speed limit he or she wants to drive, and presented with the corresponding options of "monetary cost if a ticket is received" and "safety and time gain if no ticket is received". The driver determines the driving speed (such as 100 mph) of the driver when a speed limit sign appears by comprehensively considering the two factors, and the difference between the driving speed determined by the driver and the speed limit (such as 85 mph) is the 'decision reference value' (DMR) which is miles per hour (which can be converted into kilometers per hour). In addition, drivers need to provide personal life schedules, i.e., driving times and distances on highways and cities on weekdays, weekends, in experiments. That is to say the decision reference of the driver also influences the driving behavior of the driver.

In addition to driving behavior affecting battery life, charging strategies, battery configurations, and living schedules will affect battery life. Suppose that the driving distance of the driver on the expressway and the urban road on the working day is D_hdAnd D_udThe driving distance of weekends on expressways and urban roads is D_heAnd D_ueIn summary, the following empirical model can be proposed:

Lifetime＝pa×Personality+pb×DMR+pc×Charging+pd×C_p+pd×Δheat

+pf×N+ph×T_initial+pi×D_hd+pj×D_ud+pk×D_he+pl×D_ue

wherein, personality represents character, DMR represents decision reference value, Charging represents Charging strategy, C_pRepresents the heat capacity of the battery, delta heat represents the difference of heat attenuation, N represents the number of single batteries in the battery pack, T_initialDenotes the ambient temperature, D_hdAnd D_udIndicating the driving distance of the driver on the highway and on the urban road on a working day, D_heAnd D_ueThe driving distance of the driver on the expressway and the urban road on the weekend is represented, and pa to pl represent coefficients.

Experimental environment and experimental procedure:

used in experimentsDriving simulator (STISIMDRIVEM100K, Syst)ems technology inc, Hawthorne, CA). It includes a Logitech with powerful feedbackThe accelerator pedal is at rest 38.2 (the angle between the pedal surface and the ground), the maximum accelerator input angle is 15.2, the brake pedal is at rest 60.1, and the maximum brake input angle is 28.6. furthermore, the driving scene is displayed on a 27 inch 1920 × 1200 pixel resolution LCD display.

The experimental scenario is a simulated one-lane (per direction) environment including urban roads and highways, with no other vehicles, pedestrians, or road signs (e.g., stop signs). The average one-way commute time in the united states is reported to be 26 minutes (average distance is 16 miles). The subject was required to drive 16 miles in the experiment. It is also reported that 55% of gasoline is consumed on urban roads and 45% on expressways in the united states. To simulate a general commute case, the scene sequence was 30% urban roads, 45% freeways, and finally 25% urban roads. The speed limit is 30mph for urban roads and 45mph for expressways, and the speed limit can appear 200 feet in front of the driver. The subject is required to adjust the speed when seeing the speed limit sign as in the case of real driving. In addition, the scene density of the whole experimental scene is kept consistent all the time.

To calculate how much energy the battery is consumed during driving, we assume that the vehicle has the physical characteristics of toyota cameri in 2008: the mass is 1588kg (3500lbs), the drag coefficient is 0.28, and the frontal area of the vehicle is 2.7m². The dimensionless rolling resistance coefficient of the tire is 0.01, which relates the resistance to motion as a normal force. The energy transferred to the battery by regenerative braking is assumed to be 40%, and the transfer efficiency from the battery to the steering wheel is80 percent. The battery power is assumed to be 16kWh (i.e. Volt manufactured by chevrolet). Air density was taken from the standard atmospheric density at sea level in the united states. We load an 800W constant load on the electric vehicle to represent activities unrelated to vehicle motion, such as warm air, air conditioning, lighting and other accessories. The energy required for the vehicle to move when the acceleration is negative and non-negative can be described by equations 6 and 7 above.

After that, the testees will answer to the extent to which the electric vehicle has been reduced in power, they will charge the battery. Since we can calculate the energy consumption of each tested person during commute based on the speed and acceleration data of the tested person during driving, the remaining battery capacity of the battery after commute every day can be obtained. The residual electric quantity is less than the electric quantity value which is charged by the testee, and the testee can select to charge the battery.

Each subject will provide their life schedule including the commute distance and time of the weekdays, the driving activities on weekdays and weekends other than commute, and monthly and yearly life schedules, such as whether or not a particular day will be driven for travel. We collected the driving data of the test as detailed as possible in the experiment, so as to establish a rough framework of their annual driving data.

Based on the calculated energy consumption in the case study, and the subject's charging strategy and schedule of life, we can obtain a simulated battery life.

Determining an empirical formula:

(1) and (3) verifying a theoretical model:

in one embodiment of the present invention, the battery life t is calculated according to equation 3 in combination with the above experimental environment, and the simulated battery life is shown in table I.

Because of T_initialAnd Δ Heat will vary continuously, the operating temperature of the battery is difficult to calculate, and we simply assume that the thermal management module of the battery can maintain the operating temperature of the battery at a constant level. T is_initialAssumed to be 30 ℃, C_p1000.4J/kg/K,. DELTA.heat ═ 1.33W/cell, and N ═ 200. In the experiment, the driving distance of a highway is assumed to be 7.2 miles, the driving distance of an urban road is assumed to be 8.8 miles, and the driving distance of each tested road is the same. Assuming that the battery is fully charged and cannot sustain the energy consumption of a day's commute, the battery must be replaced, i.e., end of life. The Capacity (%) in equation 3 is obtained by dividing the total battery energy E consumed during the day by the amount of electricity when the battery leaves the factory and is fully charged. Substituting the battery configuration and temperature into equation 3 can calculate the battery life t. During the analysis, the personality factors were quantified, specifically represented by-1, 0 and 1 for non-impulsive, normal and impulsive types in order.

TABLE I.T_initialCharacteristics tested at 10 ℃ and Battery Life

The experimental result of the battery life is calculated by equation 3. We compared the simulated battery life with the expected battery life of 96 months provided by general electric, with experimental results for battery life yielding a significant level value of 0.184 (> 0.05) in the independent sample T test, and thus the experimental results were not significantly different from the expected battery life of general electric.

(2) Regression analysis and determination of empirical model:

next, we consider the different living schedules of each tested person, and assume that the driving distances of the tested persons on the expressway and the urban road in the experiment are d respectively_hAnd d_uThe energy of the battery consumed on the expressway and the urban road is e_hAnd e_uAlso, there were tested questionnaires giving daily resultsThe driving distances on the expressway and the urban road are respectively D_hAnd D_uThen the battery energy E actually consumed on the expressway and the urban road every day is tried_hAnd E_uAre respectively as

$\left\{\begin{array}{c}{E}_h=\frac{D_h}{d_h}\&CenterDot;e_h\\ E_u=\frac{D_u}{d_u}\&CenterDot;e_u\end{array}\right.---\left(20\right)$

Then there is a daily actual total power consumption:

$E=E_h+E_u=\frac{D_h}{d_h}\&CenterDot;e_h+\frac{D_u}{d_u}\&CenterDot;e_u---\left(21\right)$

assuming that the battery is fully charged and cannot sustain the energy consumption of a day's commute, the battery must be replaced, i.e., end of life. From equation 21, the value of Capacity (%) in equation 3 is obtained by dividing E by the amount of electricity at the time of factory full charge of the battery. Arranging the battery C_p707J/kg/K,. delta.heat ═ 1.33W/cell, N ═ 200, ambient temperature T_initialSubstituting equation 3 at 10 c, the value of variable t in equation 3, i.e., the life of the battery of the electric vehicle in actual life under test, can be calculated. The results of the characteristics, life schedule and battery life of the test are shown in table II.

Table ii. t_initialThe battery life calculation result in consideration of the tested life schedule at 10 DEG C

The calculation of the battery life for most of the tested was higher than the GM expected battery life because in previous experiments, the daily commute distance (highway and urban road) was set at the us per-capita commute distance of 16 miles (highway 7.2 miles, urban road 8.8 miles). As can be seen in table II, most of the tested commutes are far less than 16 miles away, which results in a significant reduction in the charging frequency, which is why the battery life GM is higher than the GM expected battery life.

Converting the battery configuration and temperature of Table IIThe parameters, 16 sets of battery life and 192 calculations were performed to obtain table III. Wherein, C_p707J/kg/K or 1019J/kg/K, Δ heat 1.33W/cell or 4.45W/cell, N200 or 228, T_initialThe results of the linear regression analysis (the regression analysis results are shown in table III, where x or x indicates that the variable has a significant effect on the battery life) show the factors that have a significant effect on the battery life by the software SPSS, and we can establish a linear equation for the battery life with respect to these factors from this analysis result, see equation 8.

Lifetime＝393.768-35.1552Personality-2.85Charging-1.382T_initial(8)

-9.716D_hd-2.592D_ud-4.361D_he-11.533D_ue

Wherein the parameters are Personality personalitity, Charging strategy Charging, ambient temperature T_initialThe driving distance on the working day highway and the urban road is D_hdAnd D_udThe driving distance on the weekend expressway and the urban road is D_heAnd D_ue. Wherein, the non-normalized coefficients in table III are the coefficients of the variables in equation 8.

The coefficients of equation 8 are fitted based on known characteristics of the test (character, decision reference, charging strategy, life schedule), and battery configuration (heat capacity, number of cells, initial temperature, thermal decay difference) and battery life derived from equation 3. Since q in equation 3 is derived from the driving behaviors (speed, acceleration) under test, which were not predicted in previous studies, previous studies could not predict battery life without driving the electric vehicle. Here we can get the battery life directly from the tested features and the ambient temperature by regression analysis. That is, with the formula 8, we can obtain the life of the battery of the electric vehicle without trying to drive the electric vehicle, and the cost for predicting the life of the battery can be greatly reduced.

Adjusted R in the results of the Linear regression analysis of SPSS²The values are decision coefficients or measurement coefficients, which indicate how well the regression independent variable fits to the dependent variable, where R²The value is 0.8 and the P value is 0.000, i.e. equation 8 can fit well to the battery life calculated by equation 3 and predict the battery life. The significance of the character, charging strategy, ambient temperature, driving distance on motorways and urban roads on weekdays and weekends are all less than 0.05, which indicates that their coefficients are statistically significant other than 0, and that every change in magnitude does change battery life.

Since the physical characteristics of Toyota Camry in 2008 are adopted in the calculation process, however, some of the physical characteristics, such as m and p, change with the change of the driver and the external environment. When the formula 8 is applied to other vehicle types, these physical characteristics will also change, and we should adjust the coefficients. Therefore, we present coefficient ranges here so that equation 8 can be applied more widely when there is a variation in the physical properties of toyota cameri or when equation 8 is applied to other vehicle models. In table III, the values obtained for the non-normalized coefficient (B) ± the coefficient standard deviation (std. development) are the range of the coefficient, from which we can obtain the range of battery life. Therefore, the empirical model (equation 8) can be further written as:

Lifetime＝pa+pb×Personality+pc×Charging+pd×T_initial

+pe×D_hd+pf×D_ud+pg×D_he+ph×D_ue

statistically, 99.7% of samples in a normal distribution fall within the range of "mean. + -. 3 standard deviation", so according to Table III, the coefficient range of equation 8 is "non-normalized coefficient (B) + -3 × coefficient standard deviation (Std. Deviation)" where pa ranges between 393.7864. + -. 147.2502, pb ranges between-35.1552. + -. 21.9882, pc ranges between-2.85. + -. 2.145, pd ranges between-1.3822. + -. 1.1946, pe ranges between-9.716. + -. 2.7264, pf ranges between-2.5916. + -. 3.2112, pg ranges between-4.3606. + -. 2.7264, and ph ranges between-11.533. + -. 4.9872.

Method for predicting service life of lithium battery of electric vehicle

In one embodiment of the present invention, a method for predicting a lifetime of a lithium battery of an electric vehicle is provided, which includes:

firstly, setting personal character parameters, a charging strategy, an initial temperature and a living schedule;

and step two, calculating the service life of the battery according to a formula 8.

TABLE III regression analysis of Battery Life

	Life (moon)
		Character lattice
Non-normalized coefficient (B)	-35.1552**
		Normalization coefficient (Beta)	-.294
Coefficient standard deviation (Std. development)	7.3294
		Decision reference value (DMR) (mph)
Non-normalized coefficient (B)	3.3056
		Normalization coefficient (Beta)	.104
Coefficient standard deviation (Std. development)	3.1682
		Charging strategy (%)
Non-normalized coefficient (B)	-2.85**
		Normalization coefficient (Beta)	-.380
Coefficient standard deviation (Std. development)	0.715
		Heat capacity (J/kg/K)
Non-normalized coefficient (B)	-0.0142
		Normalization coefficient (Beta)	-.022
Coefficient standard deviation (Std. development)	0.0256
		Thermal attenuation (W/cell)

Non-normalized coefficient (B)	-0.268
		Normalization coefficient (Beta)	-.004
Coefficient standard deviation (Std. development)	2.553
		Number of batteries
Non-normalized coefficient (B)	-0.1484
		Normalization coefficient (Beta)	-.034
Coefficient standard deviation (Std. development)	0.181
		Initial temperature (. degree. C.)
Non-normalized coefficient (B)	-1.3822**
		Normalization coefficient (Beta)	-.140
Coefficient standard deviation (Std. development)	0.3982
		Working day highway driving distance (mile)
Non-normalized coefficient (B)	-9.716**
		Non-normalized coefficient (Beta)	-.726
Coefficient standard deviation (Std. development)	0.7804
		Working day city road driving distance (mile)
Non-normalized coefficient (B)	-2.5916*
		Normalization coefficient (Beta)	-.200
Coefficient standard deviation (Std. development)	1.0704
		Weekend expressway driving distance (mile)
Non-normalized coefficient (B)	-4.3606**
		Normalization coefficient (Beta)	-.243
Coefficient standard deviation (Std. development)	0.9088
		Weekend city road driving distance (mile)
Non-normalized coefficient (B)	-11.533**
		Normalization coefficient (Beta)	-.437
Coefficient standard deviation (Std. development)	1.6624
		Constant number
Non-normalized coefficient (B)	393.7864**
		Coefficient standard deviation (Std. development)	49.0834
Modified square of R	0.800

Significant level (P value)	0.000
		Standard error estimation	54.376
Number of samples	192

*p＜.01；**p＜.001.

Predicting optimal battery configuration given driver characteristics and battery target life according to equation 3

Based on the foregoing analysis, in one embodiment of the present invention, a method is provided for obtaining an optimal configuration of a battery given driver characteristics and a target battery life.

First, it is assumed that the battery must be replaced if the full charge of the battery is not sufficient to maintain the energy consumption required for the commute of the vehicle during the day. From the case study, we can calculate the battery energy E consumed by each driver driving each day, and thus obtain the Capacity (%) value in equation 3. Furthermore, the 'fmisearch' command in Matlab software can be used for inputting a formula 3, and when the characteristics of a driver, namely a decision reference value DMR and a Charging strategy changing are known, the optimal configuration of the battery, namely C, is obtained_pΔ heat and N such that the Capacity (%) value in equation 3 is as close as possible to the energy consumption of the battery and the just-out of the battery on a given day of the commute being testedThe ratio of the plant energies. In other words, the optimal configuration of the battery for a given driver characteristic and battery target life is solved for the approximate zero of equation 22.

$\left\{\begin{array}{c}\frac{1}{100}\&CenterDot;(167.583-1.264T-0.097\&CenterDot;t\&CenterDot;f)\&CenterDot;(84.1+3.71\&CenterDot;f-0.0788\&CenterDot;t\&CenterDot;f)*\mathrm{CAPACITY}-E=0\\ T=T_{\mathrm{initial}}+0.0339+0.004t^{\&prime;}+0.001\&CenterDot;C+2.385(\frac{\mathrm{\&Sigma;}\stackrel{\&CenterDot;}{q}\mathrm{\&Delta;t}}{t^{\&prime;}\&CenterDot;N}-\mathrm{\&Delta;heat})-0.004\&CenterDot;C_p\end{array}\right.---\left(22\right)$

Where CAPACITY is the energy of the battery pack just after it is fully charged when it leaves the factory, E is the total energy consumption during one day of driving, and t is the target life of the battery.

When calculating the value of T in equation 22, T_initialIt may be set to different constants according to different seasons or simply to a constant that is not changed. The value of Δ t was determined by the simulator employed in the case study.

For 96 months of target life for each tested and given battery, we predicted the ambient temperature T_initialThe results are shown in table IV for the optimum configuration of the cell at 10 ℃.

Table iv. t_initialOptimum battery configuration given driver characteristics and target battery life at 10 ℃

We can see in table IV that in order to get the electrical life close to the target life, the simplest way is to change the number of cells in the battery module, with the accompanying change in the battery module volume and cost. The size of the cells is 5 inches x7 inches and the thickness is less than a quarter inch, from which the volume change of the battery pack can be obtained. The thermal conductivity of oil is 1.5 to 3 times that of air. The thermal conductivity of water when used in indirect cooling is 15 times that of air. It can be seen that different types of thermal management modules will produce distinct thermal attenuations and thus thermal management modules are also a way to extend battery life. However, developing a convenient and efficient thermal management module also requires a certain capital investment. Typical heat capacity from the cell is 795J/kg/K, which is determined by the chemical and physical materials of the cell itself. The last method to extend battery life is to develop the battery itself. Many researchers in battery materials and performance want to improve the battery, which requires a lot of capital investment and is therefore the most difficult way to optimize battery life.

To extend battery life by changing battery configuration, the target battery life is set to a value higher than the battery life originally provided by the battery manufacturer, and the optimal battery configuration for achieving the target battery life is determined. The optimal configuration of the battery pack is used to extend the battery life from the original life provided by the battery manufacturer to the target life.

According to equation 3, optimal driver behavior (i.e., driving) is predicted given the battery configuration and battery target life Speed, acceleration, and charging frequency of time) and life schedule

It is also assumed that the battery must be replaced if the full charge of the battery is not sufficient to maintain the energy consumption required for the commute of the vehicle during the day. From the case study, we can calculate the battery energy E consumed by each driver driving each day, and thus obtain the Capacity (%) value in equation 3.

Equation 3 may then be entered using the "fmisearch" command in Matlab software, at the known battery configuration N, C_pΔ heat, initial temperature T_initialThen the optimum C,And f, enabling the Capacity (%) in formula 3 to be as close as possible to the ratio of the energy consumption of the battery to the just-shipped energy of the battery in a given test daily commute. In other words, the optimal C, for a given battery configuration and battery target life is solved,And f is the approximate zero of solving equation 22. The optimal I can be obtained from the optimal C, and the optimal P can be obtained from the formula 11. The optimal driving speed v and the acceleration a can be obtained through formulas 6 and 7, and the value of v minus the actual speed limit of the road is the optimal DMR.

For each test and given a battery target life of 96 months, we fixed the battery configuration to C_p1000.4J/kg/K, Δ heat 43.2W/cell, N200, and ambient temperature T_initialThe driver's optimal driving behavior and life schedule were predicted at 10 ℃, and the results are shown in table IV.

TABLE V.T_initialOptimum driving behavior and life schedule given battery configuration and target life-time, 10 ℃

An optimal decision reference value of 0 means that the driver should not make the vehicle speed exceed the speed limit, and an optimal driving acceleration of 0m/s2 means that the driver should keep the vehicle speed constant when the desired speed is reached during driving. The final driving distance per day is 5 miles for urban roads and 4 miles for highways. The last charge frequency is 0.2 times per day, i.e. the battery should be charged every 5 days.

Using the optimum charging frequency is the simplest way to achieve the target life of the battery. The second method is to change the decision reference value so that the driving speed of the driver is not higher than the speed limit. Personality and life schedule are two other factors that affect battery life. The life schedule is closely related to the daily life of the driver and the distance between destinations, and is difficult to change. Therefore, we consider that changing the driver's charging strategy, decision reference, and acceleration are the three main means to achieve the target battery life function given a battery configuration.

When considering driving distance today, drivers reduce driving distance on an expressway or on an urban road if they drive longer on the expressway or on the urban road. An increase in driving distance may result in an increase in charging frequency, thereby reducing battery life. An increase in the charging frequency also reduces battery life if the driving distance is optimized. Otherwise, the battery life may be extended.

Because the life schedule is difficult to change, we regard it as an input value as the battery configuration, the target life of the battery. Next, we take into account different living schedules to again predict the driver's optimal driving behaviour and charging strategy, the results are shown in table VI. From table VI we can see that the optimal decision reference and the acceleration after the target speed is reached are both 0, regardless of the life schedule.

Table vi.t_initialOptimum driving and charging behavior given battery configuration, target life, and life schedule at 10 ℃

That is, to optimize battery life, the driver should minimize the jerk, keep the vehicle speed constant, and drive at a speed no higher than the speed limit. General car companies report that a 240V charging station can charge the battery in about four hours (10 hours for 120V household power). Because the number of charging stations is small at present, people can only charge after the end of every day driving. Therefore, the number of days between charging in real life must be an integer, and people should try to make the average charging interval close to the optimal value. For example, if the optimal charging interval is 2.5 days, the charging mode should be ". 2 days to 3 days to 2 days to 3 days.

The battery life is prolonged by changing the behavior and the life schedule of a driver, and the target life of the battery is set to be higher than the battery life value originally provided by a battery manufacturer, and the optimal behavior and the life schedule of the driver, which enable the battery to reach the target life, are obtained. By using the optimal set of driver behavior and life schedule, the battery life can be extended from the original life provided by the battery manufacturer to the target life.

It is to be noted and understood that various modifications and improvements can be made to the invention described in detail above without departing from the spirit and scope of the invention as claimed in the appended claims. Accordingly, the scope of the claimed subject matter is not limited by any of the specific exemplary teachings provided.

In summary, according to equation 8 (i.e. empirical model), when the target battery life is given, the relevant parameters such as the optimal configuration of the battery, the charging strategy, and the driving schedule can be obtained. It is understood from another aspect that the battery life may be brought to a target value by optimizing battery configuration, driving behavior, and charging strategy.

Term parameter table

Claims (6)

1. A method for predicting the service life of a lithium battery of an electric vehicle comprises the following steps:

the method comprises the following steps of firstly, acquiring the working temperature and charging frequency of a battery and the ratio of the battery energy required by a driver in one-day driving to the energy of the battery after the battery is fully charged when the battery leaves a factory; and

step two, predicting the service life of the battery according to the following modes:

wherein T represents the working temperature of the battery, and the unit is; f represents the charging frequency in units of 1/day;

if 0 represents the end of the service life of the battery when the electric quantity of the fully charged battery can not provide the energy required by the automobile for one-day driving, the elapsed time T is obtained according to the charging frequency f, the working temperature T of the battery, and the ratio Capacity (%) of the battery energy required by the driver for one-day driving and the energy fully charged battery just before the factory shipment, and the value represents the predicted service life of the battery, and the unit is day.

2. The method of claim 1, wherein the operating temperature of the battery is calculated according to:

wherein, T_initialRepresents the initial temperature of the battery during operation, and the unit is; t' represents a continuous working time of the battery, and the unit is second; c represents a battery discharge rate;represents the heat generation rate of the battery pack in watts; Δ t represents the time interval in seconds for the relevant instrument to record driving data; n represents the number of cells in one battery pack; delta heat represents the difference in thermal decay produced by different types of cooling modules relative to the air cooling module, in watts per single cell; c_pThe heat capacity of the battery is expressed in J/kg/K.

3. A method for predicting the service life of a lithium battery of an electric vehicle comprises the following steps:

step one, acquiring personality parameters of a driver, a charging strategy, an environment temperature, and driving distances of a working day and weekends on an expressway and an urban road;

step two, predicting the service life range of the battery according to the following modes:

Lifetime＝pa+pb×Personality+pc×Charging+pd×T_initial

+pe×D_hd+pf×D_ud+pg×D_he+ph×D_ue

the life represents the predicted service life of the battery, the unit is month, the Personality represents the Personality parameters of the driver, and the quantization mode of the Personality parameters of the driver is as follows: non-impulse type, normal and impulse type are represented by-1, 0 and 1 in sequence; charging represents a Charging strategy, namely the minimum amount of electricity left in the battery before the battery is charged, and the value is a percentage; t is_initialRepresents the ambient temperature; d_hdAnd D_udIndicating driving distance on weekdays on highways and urban roads, D_heAnd D_ueRepresenting driving distance in miles on weekend highways and urban roads; pa is a constant, pb-ph represents a coefficient;

wherein pa ranges from 393.7864 + -147.2502, pb ranges from-35.1552 + -21.9882, pc ranges from-2.85 + -2.145, pd ranges from-1.3822 + -1.1946, pe ranges from-9.716 + -2.7264, pf ranges from-2.5916 + -3.2112, pg ranges from-4.3606 + -2.7264, and ph ranges from-11.533 + -4.9872.

4. The method of claim 3, wherein the formula of step two is:

Lifetime＝393.768-35.1552Personality-2.85Charging-1.382T_initial

-9.716D_hd-2.592D_ud-4.361D_he-11.533D_ue。

5. a method for prolonging the service life of a lithium battery of an electric vehicle comprises the following steps:

step one, the following modifications are made to the formula in the method of claim 2:

wherein CAPACITY is the energy of the battery pack after being fully charged just before the battery pack leaves the factory, and the unit is J; e is the total energy consumption during one day of driving, and the unit is J; t is t_mIs prepared byThe time, here representing the target life of the battery, in days; t represents the working temperature of the battery, and the unit is; t is_initialRepresents the initial temperature of the battery during operation, and the unit is; t' represents a continuous working time of the battery, and the unit is second; c represents a battery discharge rate;represents the heat generation rate of the battery pack in watts; Δ t represents the time interval in seconds for the relevant instrument to record driving data; n represents the number of cells in one battery pack; delta heat represents the difference in thermal decay produced by different types of cooling modules relative to the air cooling module, in watts per single cell; c_pRepresents the heat capacity of the battery and has the unit of J/kg/K;

step two, acquiring the battery energy E consumed by driving every day, wherein the unit of charging frequency f is 1/day;

step three, obtaining the optimal configuration C of the battery by iteratively solving the formula_pΔ heat and N.

6. A method for prolonging the service life of a lithium battery of an electric vehicle comprises the following steps:

step one, the following modifications are made to the formula in the method of claim 2:

wherein CAPACITY is the energy of the battery pack after being fully charged just before the battery pack leaves the factory, and the unit is J; e is the total energy consumption during one day of driving, and the unit is J; t is t_mIs elapsed time, here representing the target life of the battery, in days; t represents the working temperature of the battery, and the unit is; t is_initialRepresents the initial temperature of the battery during operation, and the unit is; t' represents a continuous working time of the battery, and the unit is second; c represents a battery discharge rate;indicating the batteryHeat generation rate in watts; Δ t represents the time interval in seconds for the relevant instrument to record driving data; n represents the number of cells in one battery pack; delta heat represents the difference in thermal decay produced by different types of cooling modules relative to the air cooling module, in watts per single cell; c_pRepresents the heat capacity of the battery and has the unit of J/kg/K; f represents the charging frequency in units of 1/day;

step two, acquiring the battery energy E consumed by driving every day and the battery configuration N, C_pΔ heat, initial temperature T_initial；

Step three, obtaining the optimal C,And f;

step four, obtaining the optimal I from the optimal C, and obtaining the optimal P from the following formula:

wherein P represents the output power of the vehicle battery pack in units of W; i represents the current of the battery pack and has the unit of A; u shape_ocvRepresenting the open-loop voltage, U, of the battery_opRepresents the operating voltage of the battery under load conditions, in volt;

step five, obtaining the optimal driving speed v and the optimal acceleration a through the following formulas:

wherein, p represents the motion power consumption of the electric automobile and the unit is W; m represents mass, in Kg; a represents the acceleration of the vehicle in m/s²Rho sea level air weight per cubic meter is approximately Kg/m³(ii) a v represents velocity in m/s; c_dRepresenting a drag coefficient of the vehicle; a represents the frontal area of the vehicle in m²，C_rrDimensionless coefficient representing rolling resistance of tire, g represents weightAcceleration of force in m/s²，

And guiding the behavior of a driver according to the obtained charging frequency f, speed v and acceleration a, thereby prolonging the service life of the lithium battery of the electric vehicle.