Non-detachable direct identification method for characteristic parameters of single-tuning filter
Technical Field
The invention relates to the technical field of single-tuned filters, in particular to a non-detachable direct identification method for characteristic parameters of a single-tuned filter.
Background
With the development of modernization of various industries of national economy, a large number of nonlinear loads mainly including various power electronic devices are increasingly and widely applied, and the problem of harmonic suppression is receiving more and more attention. The single-tuning filter formed by connecting a filter reactor and a filter capacitor in series is a basic unit for forming a power system filter device. Along with the increase of the service time of the single-tuned filter and the change of the temperature of the working environment, the internal components of the single-tuned filter can correspondingly have the conditions of temperature rise and incapability of working normally. In this case, the values of characteristic parameters of various kinds of components inside the filter may be changed and aged gradually, thereby impairing the filtering performance of the filter. Therefore, the real-time monitoring of the parameters of the components in the single-tuning filter has very important practical significance for grasping the filtering performance of the single-tuning filter and the reasonable and effective use of the single-tuning filter.
However, single-tuned filter products are often packaged, and if the parameters are easily damaged by disassembly for detection, it is difficult to detect the parameters without disassembly. Thus, research on this aspect has been conducted, and the filtering performance of the single-tuned filter actually depends on the values of three lumped basic element values, i.e., the characteristic parameters R, L, C, i.e., the internal resistance, inductance and capacitance of the filter, and these parameters R, L, C together determine the frequency characteristic of the single-tuned filter, which characterizes the filtering performance of the single-tuned filter. As shown in fig. 1, an equivalent circuit model of a conventional single-tuning filter is mainly formed by connecting a capacitor C, an inductor L and a resistor R in series, and has an input voltage u i (t), a capacitor voltage u c (t), a current i (t), and a capacitor C:
from the circuit knowledge, it can be seen that:
The input voltage u i (t) is a step signal, namely:
before t <0, i.e. before operation, there is no voltage input, i.e. u i (t) =0, the voltage across the capacitor is 0V, and the current flowing through the capacitor is 0A, i.e. the single-tuned filter is in zero state. When t is more than or equal to 0, the input voltage U i (t) of the single-tuned filter is a constant value U 0. The characteristic polynomial of the single tuning filter usually has two distinct negative real poles to ensure good filtering performance, and is connected with (1.1) to (1.3) to solve the voltage u c (t) and the current i (t) in the time domain. For a single tuned filter to be produced, it is difficult to measure u c (t) of the filter.
The existing identification method of the characteristic parameters of the single-tuning filter generally can only play a role after the actual fault of the single-tuning filter has occurred, cannot predict or prevent the occurrence of the fault in advance, and cannot directly acquire the characteristic parameters in the single-tuning filter. Still other identification methods only focus on the aging of the capacitor, and ignore the aging of other components such as inductance, resistance, etc. in the device. In addition, the calculation quality of the existing identification method is greatly affected by the initial point, and a local optimal point rather than an overall optimal point can be obtained, so that a larger error is generated.
Disclosure of Invention
The invention aims to solve the problems that most of existing identification methods for the characteristic parameters of the single-tuned filter cannot predict or prevent faults in advance, the characteristic parameters in the single-tuned filter cannot be directly obtained, ageing conditions of other elements such as inductance and/or resistance in a device can be ignored, and errors are large.
In order to achieve the aim of the invention, the invention is realized by the following technical scheme that the non-detachable direct identification method for the characteristic parameters of the single-tuned filter comprises the following steps:
firstly, preparing a constant direct current voltage source with a voltage amplitude of U 0 and a current detection system in advance, initializing a single-tuned filter to be detected into a zero state, and detecting the current i (t) of the single-tuned filter by using the current detection system until the detected current i (t) is constant to zero;
Step two, connecting the voltage source prepared in the step one to the input end of the single tuning filter to be detected, sampling the current of the single tuning filter to be detected by using the current detection system prepared in the step one and storing samples, wherein the samples are denoted as f (j) =i (jT), the current collected in the j sampling period is represented as j=1, 2,3, the number of the samples is equal to the number of the samples, and the sampling period is equal to T;
Calculating basic coefficients a (j), b (j), c (j), a basic factor u 1(j)、u2 (j) and an average basic factor av 1、av2 according to the sample f (j) stored in the step two, wherein j=1, 2,3, and q, q < N-2;
Calculating an estimated dynamic factor A (j) according to the average basic factor av 1、av2 calculated in the step three, and then calculating an average estimated dynamic factor av a according to the estimated dynamic factor A (j), wherein j=1, 2, 3.
Step five, calculating estimation factors EP 1 and EP 2 according to the average basic factor av 1、av2 calculated in the step three;
And step six, respectively calculating the capacitance value C, the inductance value L and the resistance value R of the single-tuned filter according to the average estimated dynamic factor av a calculated in the step four and the estimated factors EP 1 and EP 2 calculated in the step five.
In the first step, the input end and the output end of the single-tuned filter to be tested are respectively grounded to initialize the single-tuned filter to be tested to a zero state.
The further improvement is that in the third step, the calculation formula of the basic coefficient a (j) is as follows:
a(j)=f(j+1)f(j+1)-f(j)f(j+2);
the calculation formula of the basic coefficient b (j) is as follows:
b(j)=f(j)f(j+3)-f(j+1)f(j+2);
The calculation formula of the basic coefficient c (j) is as follows:
c(j)=f(j+2)f(j+2)-f(j+1)f(j+3)。
The further improvement is that in the third step, the calculation formula of the basic factor u 1 (j) is as follows:
The calculation formula of the basic factor u 2 (j) is as follows:
The further improvement is that in the third step, the calculation formula of the average basic factor av 1 is as follows:
The calculation formula of the average basic factor av 2 is as follows:
The further improvement is that in the fourth step, the calculation formula of the estimated dynamic factor A (j) is as follows:
The calculation formula of the average estimated dynamic factor av a is as follows:
The further improvement is that in the fifth step, the calculation formula of the estimation factor EP 1 is as follows:
the calculation formula of the estimation factor EP 2 is as follows:
in the sixth step, the calculation formula of the capacitance value C of the single-tuned filter is as follows:
The inductance value L of the single-tuned filter is calculated according to the following formula:
The calculation formula of the resistance value R of the single tuned filter is as follows:
The method has the beneficial effects that the single-tuned filter is taken as a detection object, the characteristic parameters of the internal components are monitored in real time, a simulation model of the single-tuned filter is built based on the MATLAB/Simulink platform, and the calculation method of the characteristic parameters of the internal components of the filter is verified through experiments. Through analysis of experimental results, the recognition method provided by the invention can accurately estimate the characteristic parameter of the single-tuned filter by monitoring the current of the single-tuned filter in real time on the premise of not disassembling and damaging the filter device, has the advantages of strong timeliness, high accuracy and no damage to the filter in data acquisition, and has the advantages of strong anti-interference capability, high recognition precision, stable and reliable recognition, convenient understanding of the whole method and convenient operation.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an equivalent circuit model of a conventional single-tuned filter in the background of the invention;
FIG. 2 is a flow chart of a non-detachable direct identification method of the characteristic parameters of the single-tuned filter according to the present invention;
fig. 3 is a schematic diagram of a single tuned filter circuit model in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 2 and 3, in this embodiment, a circuit model of a single-tuned filter circuit model is built in Simulink under Matlab environment, and an experiment is developed on the circuit model to verify the method of the present invention, which specifically includes the following steps:
Firstly, a single tuning filter circuit model is established, as shown in fig. 3, wherein a capacitor c=1×10 -3 F, an inductance l=2.2× 10 -3 H and a resistance r=20Ω are adopted, the gain of white noise is 0, namely no noise is input, a step signal with the amplitude of 100000 controls a controlled voltage source to generate a step voltage source U i (t) with the amplitude of U o =100000v for a single tuning filter, a current sensor collects a current i (t) of the single tuning filter, the voltage at two ends of the capacitor is initialized to 0V, the inductance current is initialized to 0A, and the single tuning filter is initialized to a zero state;
setting a sampling period T to be T=7.0x10 -7 seconds, starting simulation, enabling u i (T) to be connected to the input end of a single-tone filter, collecting and storing i (T) by a current sensor, recording f (j) =i (jT) to represent a j-th sampling value, j=1, 2,3, and the first to third degrees, and N, wherein when N=2.1x10 5, f (j) changes little, the experiment is ended, and the sampling is ended;
step three, calculating basic coefficients a (j), b (j) and c (j), wherein the basic coefficient a (j) has a calculation formula as follows:
a(j)=f(j+1)f(j+1)-f(j)f(j+2)
The calculation formula of the basic coefficient b (j) is as follows:
b(j)=f(j)f(j+3)-f(j+1)f(j+2)
the calculation formula of the basic coefficient c (j) is as follows:
c(j)=f(j+2)f(j+2)-f(j+1)f(j+3)
Calculating a basic factor u 1(j)、u2 (j), wherein the calculation formula of the basic factor u 1 (j) is as follows:
The calculation formula of the basic factor u 2 (j) is:
j=1,2,3,......,98;
the average basic factor av 1、av2 is calculated, wherein the calculation formula of the average basic factor av 1 is as follows:
the calculation formula of the average base factor av 2 is:
step four, calculating an estimated dynamic factor A (j), wherein the calculation formula is as follows:
j=1,2,3,......,178;
and then calculating an average estimated dynamic factor av a according to the estimated dynamic factor A (j), wherein the calculation formula is as follows:
Step five, calculating estimation factors EP 1 and EP 2, wherein the calculation formula of the estimation factor EP 1 is as follows:
the calculation formula of the estimation factor EP 2 is:
Step six, calculating a capacitance value C, an inductance value L and a resistance value R of the single-tuned filter, wherein the calculation formula of the capacitance value C is as follows:
The inductance value L is calculated as:
the calculation formula of the resistance value R is as follows:
The characteristic parameters of the single-tuned filter are identified by the step six, namely C= 0.000999999861F, L = 0.002200000000357H, R = 19.9999999896 Ω, and the characteristic parameters of the single-tuned filter in the experiment are true values of C=0.001F, L =0.0022H, R =20Ω, so that the characteristic parameter identification method of the single-tuned filter provided by the invention can be obtained, and for specific cases in the experiment, the identification relative errors of C, L, R are respectively-1.39×10 -5%、1.62×10-8%、-5.2×10-8%, so that the characteristic parameter identification method of the single-tuned filter provided by the invention can effectively identify a non-interference system and has high identification precision.
In actual operation, the input signal and the acquisition signal are always mixed with interference (the interference is usually represented as white noise). In order to check how much the identification method provided by the invention is affected by interference, a white noise signal module is added in the model shown in fig. 3 so as to simulate the interference signal suffered by the system. The white noise module sets the power spectral density of white noise to be 0.1, the seed number to be 2341, the interference intensity in the actual working environment is simulated by changing the amplitude of the noise, and the set interference noise-signal ratios are respectively 3.1%, 6.2%, 15.5%, 1.55%, 3.09% and 15%. Similarly, each time the interference amplitude or noise ratio is changed, the above steps one to six are repeated, and an identification experiment is performed, and table 1 records the identification experiment conditions of the method of the present invention for typical six times of the interference.
TABLE 1 identification of single-tone Filter characteristic parameters for white noise interference Signal
As can be seen from table 1, the identification method provided by the invention can effectively identify the characteristic parameters of the single tuning filter, and the identification accuracy of each characteristic parameter is almost the same. With the increase of the white noise amplitude or the increase of the noise-signal ratio, the identification error of the method is larger and larger. In general, the identification method provided by the invention has stronger anti-interference capability, for example, when the noise-signal ratio reaches 15%, the identification error of each characteristic parameter is about 1.8%. The conditions show that the identification method provided by the invention has strong anti-interference capability, high identification precision and stable and reliable identification.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.