CN115952900A - AI analysis-based intelligent energy operation optimization method - Google Patents
AI analysis-based intelligent energy operation optimization method Download PDFInfo
- Publication number
- CN115952900A CN115952900A CN202211687293.6A CN202211687293A CN115952900A CN 115952900 A CN115952900 A CN 115952900A CN 202211687293 A CN202211687293 A CN 202211687293A CN 115952900 A CN115952900 A CN 115952900A
- Authority
- CN
- China
- Prior art keywords
- parameters
- optimal
- vacuum
- working condition
- optimization
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 77
- 238000004458 analytical method Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000004519 manufacturing process Methods 0.000 claims abstract description 22
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 238000007726 management method Methods 0.000 claims abstract description 10
- 238000012360 testing method Methods 0.000 claims abstract description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 12
- 238000010219 correlation analysis Methods 0.000 claims description 10
- 239000013535 sea water Substances 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000010248 power generation Methods 0.000 claims description 8
- 230000006872 improvement Effects 0.000 claims description 7
- 238000010977 unit operation Methods 0.000 claims description 7
- 238000013459 approach Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000004134 energy conservation Methods 0.000 claims description 6
- 238000010845 search algorithm Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000007418 data mining Methods 0.000 claims description 4
- 238000003745 diagnosis Methods 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 239000000779 smoke Substances 0.000 claims description 3
- 238000003971 tillage Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 239000003245 coal Substances 0.000 description 14
- 230000009467 reduction Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004781 supercooling Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000003750 conditioning effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011234 economic evaluation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention provides an AI analysis-based intelligent energy operation optimization method, which comprises the following steps: and establishing an optimization algorithm model based on historical data, setting different evaluation indexes according to different optimization parameters after fully considering external boundary conditions by using the historical operating data of the power plant, reasonably classifying the operating conditions under various operating conditions, searching the optimal operating conditions of the unit, and displaying the optimal operating conditions on line. The invention improves the management level of the operation indexes and improves the operation production efficiency by establishing the operation index standard through the operation management function taking performance calculation and consumption difference analysis as test points.
Description
Technical Field
The invention relates to the technical field of power generation, in particular to an intelligent energy operation optimization method based on AI analysis.
Background
Zhejiang Datang Wusha mountain power generation Limited liability company is located in the town of West week of Xiangshan county, nibo city, zhejiang province, and four 60 thousands of Watt domestic supercritical coal-fired power generation units are built. The current business pain point problem faced:
(1) The fuel cost of the power enterprise is high, the power price of the power network bears pressure, the power is influenced by national policies such as productivity and environmental protection, and the profit margin is greatly compressed. The pressure of survival, energy conservation and consumption reduction of power enterprises is high.
(2) The requirement of the automation level of the power plant is continuously improved, the requirement of personnel is continuously reduced, the contradiction between the personnel and the automation management and control is caused, and the professional technical management personnel of the existing power plant cannot meet the automation management of the height of the power plant.
(3) Economic parameter coupling is difficult to decide, for example, cold end optimization, increase of circulating pump output, vacuum change, reduction of unit coal consumption, increase of circulating pump output, increase of service power and unit coal consumption, and the two have relative contradictions.
(4) The operation adjustment of the unit in the production process depends on the experience of personnel and professional technical level, and is difficult to copy and master; the mediation by different professionals is different; the operation adjustment limitation is large, and no unified standard exists.
In view of the fact that a large amount of production data are accumulated in daily production and operation of a power plant, data optimal production parameter mining by utilizing a big data algorithm technology becomes a research target of universities and colleges, scientific research units and experts, valuable information is extracted from a large amount of historical data by adopting real-time data mining, and data preparation based on performance calculation and consumption difference analysis is carried out, so that a target value of an optimization parameter is determined, real-time matching of working condition change and full working condition response are realized, and fine adjustment and optimization are realized. And (4) deciding production on the economic level, planning and arranging production based on an economic model, and ensuring profit maximization to realize revenue generation at the initial planning stage. Meanwhile, the working condition optimization of historical data is realized by using a big data new technology, energy conservation and consumption reduction are realized in the real-time production process through reasonable adjustment of equipment and a system, and the production cost is reduced.
Disclosure of Invention
The invention provides an AI analysis-based intelligent energy operation optimization method, which aims to solve at least one technical problem.
To solve the above problems, as an aspect of the present invention, there is provided an intelligent energy operation optimization method based on AI analysis, including:
(1) Establishing an optimization algorithm model based on historical data, setting different evaluation indexes according to different optimization parameters after fully considering external boundary conditions (such as atmospheric pressure, ambient temperature, humidity and the like) by utilizing the operation historical data of the power plant, reasonably classifying the operation conditions under various operation working conditions (unit load), searching the optimal operation conditions of the unit (system and equipment), and displaying the optimal operation conditions on line;
(2) Constructing a real-time calculation model of the performance indexes of the equipment (system), realizing real-time minute-level and hour-level data real-time online calculation and analysis of the performance indexes of the equipment (system), displaying the current running condition and the historical optimal condition of the equipment (system) in real time, and guiding the running management in real time by taking the calculation data result as the guide;
(3) Constructing an online self-learning model: the online comparison of real-time operation conditions, equipment operation indexes and historical optimal conditions is realized, the optimal conditions are continuously recorded and optimally compared, the optimal operation conditions are deeply excavated, and the energy-saving and efficiency-increasing level of unit operation is continuously improved;
(4) Based on the correlation analysis of the optimization parameters, the correlation influence parameters of the optimization parameters are sequenced from high to low, the optimization parameters are optimized and adjusted through comparison of current values and historical optimal values and fine tillage, and the optimization parameters are continuously optimized and adjusted towards the final target of the optimal operation condition.
(5) Based on the historical data and the operation condition of the unit, the real-time value of the system operation parameter is compared with the historical optimal value by combining the factors of the external environment (such as the environmental temperature, the seawater temperature and the like), and the index deviation depth analysis and the perspective of potential problems are realized;
(6) And a big data mining and expert experience mode, reasonable classification, reasonable parameter adjustment and classification iteration are carried out, an own expert database is constructed, and optimal deviation guidance is realized. Through the optimization suggestion, the adjustment suggestion is given to the power plant operator from the operation parameter adjustment, the direction is indicated for realizing the operation under the optimal working condition, and the purpose of improving the unit efficiency is finally achieved.
Preferably, the working condition optimizing system is based on big data and AI technology, analyzes, processes and excavates massive historical data through an algorithm, combines multiple economic indexes to find a historical optimal value, and matches the current running state of the generator set, so that a running optimizing operation suggestion for improving the running safety and the economy of the generator set is obtained;
the method comprises the following steps of optimizing working condition of a large data platform, developing vacuum by optimizing parameter indexes, slicing the working condition by considering seawater temperature, steam turbine steam admission load and steam admission parameters (calculating enthalpy value of main steam by using main steam pressure and main steam temperature, multiplying the enthalpy value by main steam flow to obtain steam turbine steam admission load) by the vacuum index optimization, searching the optimal vacuum under each working condition slice, simulating the principle of a steam turbine power micro-increase test, and defining the optimal vacuum according to the optimal vacuum: the vacuum of the condenser is improved, so that the power of the steam turbine is increased, and the vacuum value when the difference between the increased power and the power consumed by the circulating water pump reaches the maximum is the optimal vacuum of the condenser. The operation mode of the combination of the auxiliary equipment of the vacuum system is considered as a better operation mode, the operation mode of the combination of the auxiliary equipment of the vacuum system, the operation mode and parameters of the vacuum system and the circulating water system of the condenser are analyzed under the optimal vacuum by taking the vacuum as an evaluation index, a benchmark of the optimal vacuum operation mode and parameters is given, and an operator is guided to adjust the corresponding parameters to approach and reach the working condition of the benchmark, so that the state of the whole system approaches and reaches the state of the optimal vacuum.
Preferably, the modeling process for vacuum indicator optimization comprises:
(1) Dividing the stable load working conditions of the unit according to parameters such as load, ambient temperature and the like, and dividing the working conditions with similar conditions into the same class;
(2) Selecting key influence factors, finding adjustable parameters with large influence on vacuum through correlation analysis, and determining parameters which are mainly considered during optimization;
(3) Excavating and screening excellent historical samples, finding an optimal batch of samples in each type of working condition according to corresponding evaluation indexes, and constructing a case library according to the optimal batch of samples;
(4) Establishing an expert database, namely establishing a specific operation adjustment suggestion according to the historical optimal case database to guide operation adjustment in a large direction;
(5) And matching the current working condition with the working conditions in the case library, and guiding the adjustment of the current working condition according to the matched optimal working condition related operation parameter setting. Iterating again for unreasonable classifications;
(6) The parameter adjustment guidance is that directly adjustable parameters and indirectly adjustable parameters obtained by correlation analysis are ranked according to the importance degree of the influence factors on the vacuum, and the parameter adjustment direction is indicated by comparing the parameters with the optimal values under the same historical working condition;
(7) The online learning function is used for updating the current model by using real-time latest high-frequency data of actual operation of the power plant as training data, and comparing, updating and iterating optimal values of various performance indexes under the same stable operation condition, so that the real-time performance and the accuracy of an algorithm model, the optimal condition and an operation suggestion are greatly improved, and the energy conservation and the efficiency improvement of unit operation are ensured;
(8) And (3) performing off-line updating, namely periodically extracting full and accurate original data of specific equipment of the power plant, and performing off-line updating and deployment from three elements of data, a model and an algorithm according to the latest operating condition of the power plant, so as to ensure the real-time performance and feasibility of an optimization result.
Preferably, the division of the working condition slices is a brute force search algorithm, the granularity of the division of the working condition slices, that is, the interval size of each coordinate influences the search efficiency, and is a key for whether the search is successful, and specifically, the minimum unit (that is, the working condition slice range) of each boundary parameter in operation can be determined by using professional knowledge.
(1) Comprehensively considering steam turbine steam inlet parameters (calculating a main steam enthalpy value by using main steam pressure and a main steam temperature, multiplying the main steam enthalpy value by the main steam flow to obtain steam turbine steam inlet load), and confirming that the steam inlet parameter slice interval is 5000kJ according to the power (not more than 5000 kW) change range of the circulating water pump and the comprehensive consideration that the number of samples in each working condition slice cannot be too small;
(2) Dividing stable working conditions (steam inlet parameters are in the same working condition slice within the past 3 minutes) to eliminate the influence of unstable parameters, and removing samples which do not belong to the stable working conditions from the database without the need of query;
(3) The seawater temperature is expressed by the temperature of the inlet water of a condenser, the number of samples in each working condition slice cannot be too small by comprehensively considering, and 2 ℃ is taken as the minimum unit.
Preferably, the vacuum optimization algorithm comprises:
construction of Pk = f1 (Dw, N, tw 1) function 1
f ([ delta ] Dw, N, tw 1) = [ delta ] NT- [ delta ] Np = ] delta N, and when the difference between output increase [ delta ] NT of the unit corresponding to vacuum improvement and increase [ delta ] Np of service power increase [ delta ] of the vacuum system caused by vacuum system increase is the largest, the vacuum is the best vacuum.
Δ NT, establishing a function 2 of Δ NT = f2 (Δ Dw, pk, tw 1) based on function 1
Δ Np,. DELTA.np = f3 (Δ Dw) function 3
f2 (. DELTA.Dw, pk, tw 1) -f3 (. DELTA.Dw) =. DELTA.N, maximum solution of the difference can be obtained
Wherein: n load, tw1 circulating water inlet temperature, pk vacuum and Dw circulating water quantity.
Preferably, the plant power rate is optimized: optimizing parameter indexes are used for carrying out service power optimization, working condition slices are divided by considering parameters such as environment temperature, load and the like, and the lowest value of the unit production service power and the lowest value of the common load service power are respectively searched in the working condition slices. The combined operation mode of the auxiliary equipment is considered to be a better operation mode, the operation mode and the parameters of each production system are also in a better state, the combined operation mode of the auxiliary equipment of each system and the operation mode and the parameters of each production system are analyzed, the benchmarks of the operation modes and the parameters of each production system and equipment are given, operation optimization adjustment is guided by comparing the difference between the operation modes and the benchmarks, the benchmarks are continuously approached and reached, and accordingly the auxiliary power consumption rate is reduced.
Preferably, the modeling process of plant power rate optimization:
(1) Dividing the working conditions according to parameters such as load, ambient temperature and the like, and dividing the working conditions with similar conditions into the same class;
(2) And analyzing the operation parameters, finding out parameters which have larger influence on the auxiliary power, and determining the parameters which are mainly considered in optimization. Such as: the running collocation condition of auxiliary machines, the load condition of each auxiliary power bus and the like;
(3) Finding an optimal batch of samples in each type of working condition according to corresponding evaluation indexes so as to construct a case library;
(4) Matching the current working condition with the working conditions in the case library, and setting and guiding the adjustment of the current working condition according to the matched optimal working condition related operation parameters;
(5) For the condition that the current state of the equipment is not adjustable, an analysis and diagnosis report can be provided to explain the power consumption condition of the main equipment and give a maintenance reference.
Preferably, the division of the working condition slices is a brute force search algorithm, and the granularity of the division of the working condition slices, namely the interval size of each coordinate, influences the search efficiency and is a key for success or failure of the search. Specifically, the minimum unit (i.e. the range of the working condition slice) of each boundary parameter in operation can be determined by professional knowledge;
(1) The power generation load takes 10MW as a load working condition interval;
(2) And dividing the stable working condition (the load is in the same working condition slice in the last 3 minutes) to eliminate the influence of unstable parameters, and removing the samples which do not belong to the stable working condition from the database without the need of query.
(3) Ambient temperature has no special measurement points and is expressed using blower inlet temperature in minimum units of 1 degree celsius.
Preferably, in the same working condition slice, the service power optimization algorithm:
construction of Pk = sum (p 1, p2, p 3)/N, and when the function value was the smallest, it was considered to be the optimum
Wherein: the system comprises p1, p2 and p3 unit subsystems, a service power load (such as a condensate system and a wind and smoke system) and an N unit load.
Compared with the traditional optimizing mode, the intelligent working condition optimizing system based on the big data has the following advantages
(1) The optimal parameters under each working condition based on actual data analysis, non-single working condition and non-single load divided working condition are divided according to various related parameters.
(2) Based on the analysis result of the actual data, the non-experimental fitting data has the actual application effect.
(3) The optimal state of the unit is mined and optimized based on historical data, the optimal working condition is self-learned based on a big data autonomous supervision algorithm, and the optimal state of the unit is kept on line.
(4) The historical data of the optimal state is used for calculating and evaluating results, the large data sample size is large, and the result effectiveness is guaranteed to the maximum extent.
(5) The deep combination of big data artificial intelligence and electric power application gives full play to trade and big data advantage, effectively combines, and full sample full operating mode is sought the optimum.
The potential of energy conservation and emission reduction of the unit is excavated through optimizing control, operation optimization and other means, the overall coal consumption of the unit is reduced, and the unit is more efficient and energy-saving in operation. The method comprises the steps of matching unit load, atmospheric pressure and ambient temperature, finding out a historical optimal operation sample by considering the mutual influence relation among vacuum degree, plant power consumption indexes and the like, enabling the real-time operation state to be close to or even exceed the optimal coal consumption index by adjusting direct influence factors and indirect influence factors among all relevant parameters, realizing an online learning function, and providing adjustment guidance under the same working condition. The unit can run under the optimal economic working condition, and the standard coal consumption of the unit power supply is effectively reduced.
By the operation management function taking performance calculation and consumption difference analysis as test points and establishing an operation index standard, the operation index management level is improved, the precipitation of operation knowledge is enhanced, the dependence of some technologies on professionals is eliminated, and the operation production efficiency is improved.
Drawings
FIG. 1 schematically illustrates a data communications network architecture diagram;
FIG. 2 schematically illustrates a process diagram for condition-optimized modeling.
Detailed Description
The following detailed description of embodiments of the invention, but the invention can be practiced in many different ways, as defined and covered by the claims.
The project realizes research and development work such as data acquisition, processing, analysis, functional modeling and the like required by the project by expanding data storage and calculation capacity and building a big data platform on the basis of the existing data middle station of the Zhejiang Datang Wusha mountain power plant. The big data technology is utilized to analyze and research the unit operation parameters, so that the operation parameters are reasonably controlled, and the operation safety and economy are improved.
The data communication network structure is a basic supporting structure of a large data platform, and the high efficiency and reliability of the network directly influence the implementation of the whole platform. The big data platform is constructed on the basis of a packet header power generation company network, independent sub-networks which are intercommunicated and interconnected with each other in an internal cluster are constructed, data are collected from an SIS system through a time sequence data collection server, an application server provides application services to the outside through a firewall, and other servers are isolated from the outside to ensure the safety of data in the server. The network ensures that the time sequence data acquisition server, the application server and the existing network core switch are interconnected, and ensures the internal security of the data center by deploying the boundary firewall. The network architecture is schematically shown in figure 1.
The overall task is divided into three subtasks, namely, a large data platform is built, a data link is opened based on an advanced information technology, business data of each part of the power plant are fused, the sis real-time data are obtained, a data center platform is formed, and a data base analysis platform is laid for operation optimization. 2. And (4) performance calculation and consumption difference analysis are carried out, the efficiency of each system and equipment of the power plant is calculated, index analysis is used as guidance to analyze the unit performance state so as to conveniently dig the unit potential and realize economic evaluation. 3. The working condition is optimized, the energy-saving and emission-reducing potential of the unit is excavated through optimizing control, operation optimization and other means, the overall coal consumption of the unit is reduced, and the unit is enabled to operate more efficiently and energy-saving.
As the data basis for the project application. The distributed storage application was performed in time series, and the study was performed on a time series basis. And (3) carrying out data preprocessing on the total data, wherein the data preprocessing comprises data missing value condition analysis, abnormal value condition analysis, distribution condition analysis (such as quartiles and the like) of all variables, cross influence and correlation analysis (such as Spearman correlation coefficient principle) of the variables, linear regression abnormal data processing and the like. And performing index comprehensive calculation according to the calculation mode of each power plant. And reasonably classifying the mass historical data by using a K-means algorithm, and continuously classifying and adjusting parameters in the classification process until the classification is reasonable, so that a plurality of economic indexes are combined to find out a historical optimal value, the current operation condition of the generator set of the power generation enterprise is matched in real time, and an optimal adjustment guidance suggestion of the generator set operation is given. And a better state appears in the running process after the unit for self-learning, the running state of the unit is continuously optimized, and the running potential of the unit is excavated.
1. Optimizing operating conditions
Working condition optimization modeling process
(1) An optimization algorithm model based on historical data is constructed, after external boundary conditions (such as atmospheric pressure, ambient temperature, humidity and the like) are fully considered by using the historical operating data of the power plant, different evaluation indexes are set according to different optimization parameters, the operating conditions under various operating conditions (unit loads) are reasonably classified, and the optimal conditions of the operation of the unit (system and equipment) are searched. And the optimal operation condition is displayed on line.
(2) And constructing a real-time calculation model of the performance indexes of the equipment (system), realizing real-time minute-level and hour-level data real-time online calculation and analysis of the performance indexes of the equipment (system), and displaying the current running condition and the historical optimal condition of the equipment (system) in real time. And (5) guiding the operation management in real time by taking the result of the calculated data as a guide.
(3) Constructing an online self-learning model: the online comparison of the real-time operation condition, the equipment operation index and the historical optimal condition is realized, the optimal condition is continuously recorded and optimally compared, the optimal operation condition is deeply excavated, and the energy-saving and efficiency-increasing level of the unit operation is continuously improved.
(4) Based on the correlation analysis of the optimization parameters, the correlation influence parameters of the optimization parameters are sequenced from high to low, the optimization parameters are optimized and adjusted through comparison of current values and historical optimal values and fine tillage, and the optimization parameters are continuously optimized and adjusted towards the final target of the optimal operation condition.
(5) Based on historical data and operating conditions of the unit, the real-time values of the operating parameters of the system are compared with the historical optimal values by combining external environment (such as ambient temperature, seawater temperature and the like), index deviation deep analysis is carried out, and potential problems are seen through.
(7) And a big data mining and expert experience mode, reasonable classification, reasonable parameter adjustment and classification iteration are carried out, an own expert database is constructed, and optimal deviation guidance is realized. Through the optimization suggestion, the adjustment suggestion is given to the power plant operator from the operation parameter adjustment, the direction is indicated for realizing the operation under the optimal working condition, and the purpose of improving the unit efficiency is finally achieved.
1.1 vacuum optimization
The working condition optimizing system is based on big data and AI technology, mass historical data are analyzed, processed and mined through an algorithm, multiple economic indexes are combined to find a historical optimal value, and the current running state of the generator set is matched, so that a running optimizing operation suggestion for improving the running safety and the economy of the generator set is obtained.
The method comprises the following steps of optimizing working condition of a large data platform, developing vacuum by optimizing parameter indexes, slicing the working condition by considering seawater temperature, steam turbine steam admission load and steam admission parameters (calculating enthalpy value of main steam by using main steam pressure and main steam temperature, multiplying the enthalpy value by main steam flow to obtain steam turbine steam admission load) by the vacuum index optimization, searching the optimal vacuum under each working condition slice, simulating the principle of a steam turbine power micro-increase test, and defining the optimal vacuum according to the optimal vacuum: the vacuum of the condenser is improved, so that the power of the steam turbine is increased, and the vacuum value when the difference between the increased power and the power consumed by the circulating water pump reaches the maximum is the optimal vacuum of the condenser. The operation mode of the vacuum system service equipment combination is considered to be a better operation mode, the vacuum is taken as an evaluation index, the operation mode of the vacuum system service equipment combination, the operation mode and parameters of the condenser vacuum system and the circulating water system are analyzed under the optimal vacuum, a benchmark of the optimal vacuum operation mode and parameters is given, and an operator is guided to adjust the corresponding parameters to approach and reach the working condition of the benchmark, so that the state of the whole system approaches and reaches the state of the optimal vacuum.
Modeling process:
(1) And dividing the stable load working conditions of the unit according to parameters such as load, ambient temperature and the like, and dividing the working conditions with similar conditions into the same class.
(2) Selecting key influence factors, finding adjustable parameters with large influence on vacuum through correlation analysis, and determining the parameters which are mainly considered during optimization.
(3) And (4) excavating and screening excellent historical samples, and finding an optimal batch of samples in each type of working condition according to corresponding evaluation indexes so as to construct a case library.
(4) And (4) establishing an expert base, establishing a specific operation adjustment suggestion according to the historical optimal case base, and realizing guidance on operation adjustment in a large direction.
(5) And matching the current working condition with the working condition in the case library, and setting and guiding the adjustment of the current working condition according to the matched related operation parameters of the optimal working condition. Re-iterating for unreasonable classifications.
(6) And (3) parameter adjustment guidance, namely arranging the importance degrees of the directly adjustable parameters and the indirectly adjustable parameters obtained by correlation analysis according to the influence factors on the vacuum, comparing the importance degrees with the optimal values under the same historical working condition, and indicating the parameter adjustment direction.
(7) The online learning function is used for updating the current model by using the real-time latest high-frequency data of the actual operation of the power plant as training data, and comparing, updating and iterating the optimal values of all performance indexes under the same stable operation condition, so that the real-time performance and the accuracy of an algorithm model, the optimal condition and an operation suggestion are greatly improved, and the energy conservation and the efficiency improvement of the unit operation are ensured.
(8) And (3) performing off-line updating, namely periodically extracting full and accurate original data of specific equipment of the power plant, and performing off-line updating and deployment from three elements of data, a model and an algorithm according to the latest operating condition of the power plant, so that the real-time performance and the feasibility of an optimization result are ensured.
The division of the working condition slices is a brute force search algorithm, and the granularity of the division of the working condition slices, namely the interval size of each coordinate influences the search efficiency and is the key to whether the search is successful or not. In particular, expert knowledge can be used to determine the minimum unit of each boundary parameter (i.e., the range of the conditioning slice) in operation.
(1) Comprehensively considering steam turbine steam inlet parameters (calculating a main steam enthalpy value by using main steam pressure and a main steam temperature, multiplying the main steam enthalpy value by the main steam flow to obtain steam turbine steam inlet load), and confirming that the steam inlet parameter slice interval is 5000kJ according to the power (not more than 5000 kW) change range of the circulating water pump and comprehensively considering that the number of samples in each working condition slice cannot be too small.
(2) And dividing stable working conditions (the steam inlet parameters are in the same working condition slice in the past 3 minutes) to eliminate the influence of unstable parameters, and removing samples which do not belong to the stable working conditions from the database without the need of query.
(3) The seawater temperature is expressed by the temperature of the inlet water of a condenser, the number of samples in each working condition slice cannot be too small by comprehensively considering, and 2 ℃ is taken as the minimum unit.
The vacuum optimization algorithm comprises the following steps: construction of Pk = f1 (Dw, N, tw 1) function 1
f ([ delta ] Dw, N, tw 1) = [ delta ] NT- [ delta ] Np = ] delta N, and when the difference between output increase [ delta ] NT of the unit corresponding to vacuum improvement and increase [ delta ] Np of service power increase [ delta ] of the vacuum system caused by vacuum system increase is the largest, the vacuum is the best vacuum.
Δ NT, establishing Δ NT = f2 (Δ Dw, pk, tw 1) function 2 based on function 1
Δ Np, [ Δ Np = f3 ([ Δ Dw) ] function 3
f2 (. DELTA.Dw, pk, tw 1) -f3 (. DELTA.Dw) =. DELTA.N, maximum solution of the difference can be obtained
Wherein: n load
tW1 circulating water inlet temperature
Pk vacuum
Dw circulating water volume
1.1 plant rate optimization
Optimizing parameter indexes are used for carrying out station service optimization, working condition slices are divided by considering parameters such as environment temperature and load, and the lowest value of station service power of unit production and the lowest value of station service power of public load are respectively searched in the working condition slices. The combined operation mode of the auxiliary equipment is considered to be a better operation mode, the operation mode and the parameters of each production system are also in a better state, the combined operation mode of the auxiliary equipment of each system and the operation mode and the parameters of each production system are analyzed, the benchmarks of the operation modes and the parameters of each production system and equipment are given, operation optimization adjustment is guided by comparing the difference between the operation modes and the benchmarks, the benchmarks are continuously approached and reached, and accordingly the auxiliary power consumption rate is reduced.
Modeling process:
(1) And dividing the working conditions according to parameters such as load, ambient temperature and the like, and dividing the working conditions with similar conditions into the same class.
(2) And analyzing the operation parameters, finding out parameters which have large influence on the auxiliary power, and determining the parameters which are mainly considered in optimization. Such as: auxiliary machine operation collocation condition, the load condition of each auxiliary power bus, and the like.
(3) And finding an optimal batch of samples in each type of working condition according to the corresponding evaluation indexes so as to construct a case library.
(4) And matching the current working condition with the working conditions in the case library, and guiding the adjustment of the current working condition according to the matched optimal working condition related operation parameter setting.
(5) For the condition that the current state of the equipment is not adjustable, an analysis and diagnosis report can be provided to explain the power consumption condition of the main equipment and give a maintenance reference.
The division of the working condition slices is a brute force search algorithm, and the granularity of the division of the working condition slices, namely the interval size of each coordinate influences the search efficiency and is the key to whether the search is successful or not. In particular, expert knowledge can be used to determine the minimum unit of each boundary parameter (i.e., the range of the conditioning slice) in operation.
(1) The power generation load takes 10MW as a load working condition interval;
(2) And dividing the stable working condition (the load is in the same working condition slice in the last 3 minutes) to eliminate the influence of unstable parameters, and removing the samples which do not belong to the stable working condition from the database without the need of query.
(3) Ambient temperature has no special measurement points and is expressed using blower inlet temperature in minimum units of 1 degree celsius.
In the same working condition slice, the service power optimization algorithm: construction of Pk = sum (p 1, p2, p 3)/N, and when the function value was the smallest, it was considered to be the optimum
Wherein: the service electric loads (such as a condensed water system and a wind and smoke system) of the subsystems of the p1, p2 and p3 units and the loads of the N units
The invention automatically optimizes the important parameters according to the economic condition, displays the optimal working conditions (including directly adjusted parameters and indirectly adjustable parameters) of the corresponding working conditions in the history, and prompts direct operation suggestions, so that the core indexes can tend to the target values of the historical optimal working conditions, and the economic benefit level of the unit is improved.
By adopting the working condition optimizing system, the economic operation conditions of important indexes such as plant power consumption rate, condensate supercooling degree, unit vacuum and the like can be effectively optimized, the comprehensive power supply coal consumption is optimized (0.1-0.5 g/KW.h), various operation characteristics are monitored, and the system operation safety is improved.
In the case of optimizing the working condition of a certain project, the supercooling degree and the end difference of a condenser are very large, the water level of the condenser is kept about high water level operation (1120 mm) when a unit normally operates due to the problems of measuring point drift, unreasonable setting of a water level alarm value and the like, the water level of the condenser is integrally reduced through optimizing analysis, the large optimization of the indexes of the vacuum degree and the supercooling degree is realized, and the coal consumption is influenced by more than 0.5g/kWh
The invention can realize user-defined configuration aiming at different project conditions and has stronger popularization.
And performance calculation and consumption difference analysis are carried out to evaluate the energy-saving level of the unit in real time, when the coal consumption of the unit is high, working conditions are optimized to fully excavate the energy-saving potential of the unit, each index is compared to help the operation control economic index, and meanwhile, the current operation regulation guidance suggestion of the unit is given. The operation condition of each auxiliary machine and the operation combination of the auxiliary machines can be adjusted through service power optimization, the optimized combined operation of electrical equipment is realized, the service power consumption is effectively reduced, and the service power consumption rate can be generally reduced by about 0.1%. The real-time running state is close to or even exceeds the optimal coal consumption index through optimization of the cold end and the coal consumption, the real-time running state can be generally reduced by 0.5-1g/Kwh, according to a single 600MW unit, when 60 ten thousand kilowatts are generated in one hour and 1440 thousand kilowatts are generated in one day, 1g/Kwh is saved, namely 14.4 tons of coal, and 14400 yuan is saved according to 1000 yuan per ton of standard coal. Therefore, one unit can save 0.5g of coal in full-load operation one day, namely, 7200 yuan, and the unit can save about 0.4 ten thousand yuan in non-full-load operation (50% -80%). In conclusion, the energy can be saved and the consumption can be reduced for the power plant about 200 ten thousand yuan each year.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. An AI analysis-based intelligent energy operation optimization method is characterized by comprising the following steps:
(1) Establishing an optimization algorithm model based on historical data, setting different evaluation indexes according to different optimization parameters after fully considering external boundary conditions (such as atmospheric pressure, ambient temperature, humidity and the like) by using the historical operating data of the power plant, reasonably classifying the operating conditions under various operating conditions (unit load), searching the optimal operating conditions of the unit (system and equipment), and displaying the optimal operating conditions on line;
(2) Constructing a real-time calculation model of the performance indexes of the equipment (system), realizing real-time minute-level and hour-level data real-time online calculation and analysis of the performance indexes of the equipment (system), displaying the current running condition and the historical optimal condition of the equipment (system) in real time, and guiding the running management in real time by taking the calculation data result as the guide;
(3) Constructing an online self-learning model: the online comparison of real-time operation conditions, equipment operation indexes and historical optimal conditions is realized, the optimal conditions are continuously recorded and optimally compared, the optimal operation conditions are deeply excavated, and the energy-saving and efficiency-increasing level of unit operation is continuously improved;
(4) Based on the correlation analysis of the optimizing parameters, the influence parameters of the correlation of the optimizing parameters are sequenced from high to low, the optimizing parameters are optimized and adjusted by comparing the current values with the historical optimal values and carrying out fine tillage and fine operation, and the optimizing parameters are continuously optimized and adjusted towards the final target of the optimal operation condition.
(5) Based on historical data and operating conditions of the unit, the real-time values of the operating parameters of the system are compared with the historical optimal values by combining external environment (such as ambient temperature, seawater temperature and the like), index deviation deep analysis is carried out, and potential problems are seen through;
(6) And a big data mining and expert experience mode, reasonable classification, reasonable parameter adjustment and classification iteration are carried out, an own expert database is constructed, and optimal deviation guidance is realized. Through the optimization suggestion, the adjustment suggestion is given to the power plant operator from the operation parameter adjustment, the direction is indicated for realizing the operation under the optimal working condition, and the purpose of improving the unit efficiency is finally achieved.
2. The AI analysis-based intelligent energy operation optimization method according to claim 1, wherein the working condition optimization system is based on big data and AI technology, analyzes, processes and mines massive historical data through an algorithm, combines multiple economic indicators to find a historical optimal value, and matches the current operating condition of the generator set, thereby obtaining an operation optimization operation proposal for improving the safety and the economic efficiency of the generator set;
the method comprises the following steps of optimizing working condition of a large data platform, developing vacuum by optimizing parameter indexes, slicing the working condition by considering seawater temperature, steam turbine steam admission load and steam admission parameters (calculating enthalpy value of main steam by using main steam pressure and main steam temperature, multiplying the enthalpy value by main steam flow to obtain steam turbine steam admission load) by the vacuum index optimization, searching the optimal vacuum under each working condition slice, simulating the principle of a steam turbine power micro-increase test, and defining the optimal vacuum according to the optimal vacuum: the vacuum of the condenser is improved, so that the power of the steam turbine is increased, and the vacuum value when the difference between the increased power and the power consumed by the circulating water pump reaches the maximum is the optimal vacuum of the condenser. The operation mode of the vacuum system service equipment combination is considered to be a better operation mode, the vacuum is taken as an evaluation index, the operation mode of the vacuum system service equipment combination, the operation mode and parameters of the condenser vacuum system and the circulating water system are analyzed under the optimal vacuum, a benchmark of the optimal vacuum operation mode and parameters is given, and an operator is guided to adjust the corresponding parameters to approach and reach the working condition of the benchmark, so that the state of the whole system approaches and reaches the state of the optimal vacuum.
3. The AI-analysis-based intelligent energy operation optimization method according to claim 2, wherein the modeling process for vacuum indicator optimization includes:
(1) Dividing the stable load working conditions of the unit according to parameters such as load, ambient temperature and the like, and dividing the working conditions with similar conditions into the same class;
(2) Selecting key influence factors, finding adjustable parameters with large influence on vacuum through correlation analysis, and determining parameters which are mainly considered in optimization;
(3) Excavating and screening excellent historical samples, and finding an optimal batch of samples in each type of working condition according to corresponding evaluation indexes to construct a case library;
(4) Establishing an expert base, namely establishing a specific operation adjustment suggestion according to the historical optimal case base to guide operation adjustment in a large direction;
(5) And matching the current working condition with the working condition in the case library, and setting and guiding the adjustment of the current working condition according to the matched related operation parameters of the optimal working condition. Re-iterating the unreasonable classifications;
(6) The parameter adjustment guidance is that directly adjustable parameters and indirectly adjustable parameters obtained by correlation analysis are ranked according to the importance degree of the influence factors on the vacuum, and the parameter adjustment direction is indicated by comparing the parameters with the optimal values under the same historical working condition;
(7) The online learning function is used for updating the current model by using the real-time latest high-frequency data of the actual operation of the power plant as training data, and comparing, updating and iterating the optimal values of all performance indexes under the same stable operation condition, so that the real-time performance and the accuracy of an algorithm model, the optimal condition and an operation suggestion are greatly improved, and the energy conservation and the efficiency improvement of the unit operation are ensured;
(8) And (3) performing off-line updating, namely periodically extracting full and accurate original data of specific equipment of the power plant, and performing off-line updating and deployment from three elements of data, a model and an algorithm according to the latest operating condition of the power plant, so that the real-time performance and the feasibility of an optimization result are ensured.
4. The AI-analysis-based intelligent energy operation optimization method according to claim 1, wherein the operating condition slice division is a brute force search algorithm, and the granularity of the operating condition slice division, i.e. the size of the interval of each coordinate, affects the search efficiency, which is the key to the success of the search, and specifically, the expert knowledge can be used to determine the minimum unit of each boundary parameter in operation (i.e. the operating condition slice range).
(1) Comprehensively considering steam turbine steam inlet parameters (calculating a main steam enthalpy value by using main steam pressure and a main steam temperature, multiplying the main steam enthalpy value by the main steam flow to obtain steam turbine steam inlet load), and confirming that the steam inlet parameter slice interval is 5000kJ according to the power (not more than 5000 kW) change range of the circulating water pump and the comprehensive consideration that the number of samples in each working condition slice cannot be too small;
(2) Dividing stable working conditions (steam inlet parameters are in the same working condition slice within the past 3 minutes) to eliminate the influence of unstable parameters, and removing samples which do not belong to the stable working conditions from the database without the need of query;
(3) The seawater temperature is expressed by the temperature of the inlet water of a condenser, the number of samples in each working condition slice cannot be too small by comprehensively considering, and 2 ℃ is taken as the minimum unit.
5. The AI-analysis-based intelligent energy operation optimization method of claim 2, wherein the vacuum optimization algorithm comprises:
construction of Pk = f1 (Dw, N, tw 1) function 1
f ([ delta ] Dw, N, tw 1) = [ delta ] NT- [ delta ] Np = ] delta N, and when the difference between output increase [ delta ] NT of the unit corresponding to vacuum improvement and increase [ delta ] Np of service power increase [ delta ] of the vacuum system caused by vacuum system increase is the largest, the vacuum is the best vacuum.
Δ NT, establishing a function 2 of Δ NT = f2 (Δ Dw, pk, tw 1) based on function 1
Δ Np,. DELTA.np = f3 (Δ Dw) function 3
f2 (. DELTA.Dw, pk, tw 1) -f3 (. DELTA.Dw) =. DELTA.N, and the maximum solution of the difference value can be obtained
Wherein: n load, tw1 circulating water inlet temperature, pk vacuum and Dw circulating water quantity.
6. The AI analysis-based intelligent energy operation optimization method according to claim 2, wherein plant power usage rate is optimized: optimizing parameter indexes are used for carrying out service power optimization, working condition slices are divided by considering parameters such as environment temperature, load and the like, and the lowest value of the unit production service power and the lowest value of the common load service power are respectively searched in the working condition slices. The combined operation mode of the auxiliary equipment is considered to be a better operation mode, the operation mode and the parameters of each production system are also in a better state, the combined operation mode of the auxiliary equipment of each system and the operation mode and the parameters of each production system are analyzed, the benchmarks of the operation modes and the parameters of each production system and equipment are given, operation optimization adjustment is guided by comparing the difference between the operation modes and the benchmarks, the benchmarks are continuously approached and reached, and accordingly the auxiliary power consumption rate is reduced.
7. The AI analysis-based intelligent energy operation optimization method according to claim 6, wherein the modeling process for plant power rate optimization:
(1) Dividing the working conditions according to parameters such as load, ambient temperature and the like, and dividing the working conditions with similar conditions into the same class;
(2) And analyzing the operation parameters, finding out parameters which have large influence on the auxiliary power, and determining the parameters which are mainly considered in optimization. Such as: the running collocation condition of auxiliary machines, the load condition of each auxiliary power bus and the like;
(3) Finding an optimal batch of samples in each type of working condition according to corresponding evaluation indexes so as to construct a case library;
(4) Matching the current working condition with the working conditions in the case library, and setting and guiding the adjustment of the current working condition according to the matched optimal working condition related operation parameters;
(5) For the condition that the current state of the equipment is not adjustable, an analysis and diagnosis report can be provided to explain the power consumption condition of the main equipment and give a maintenance reference.
8. The AI-analysis-based intelligent energy operation optimization method according to claim 6, wherein the operating condition slice division is a brute force search algorithm, and the granularity of the operating condition slice division, i.e. the interval size of each coordinate, affects the search efficiency and is the key to whether the search is successful or not. Specifically, the minimum unit (i.e. the range of the working condition slice) of each boundary parameter in operation can be determined by professional knowledge;
(1) The power generation load takes 10MW as a load working condition interval;
(2) The stable conditions (the load is in the same condition slice in the last 3 minutes) are divided to eliminate the influence of unstable parameters, and samples which do not belong to the stable conditions are removed from the database without the need of query.
(3) Ambient temperature is not a dedicated measurement point and is expressed using blower inlet temperature in minimum units of 1 degree celsius.
9. The AI-analysis-based intelligent energy operation optimization method according to claim 8, wherein, within the same operating condition slice, the service power optimization algorithm:
construction of Pk = sum (p 1, p2, p 3)/N, and when the function value was the smallest, it was considered to be the optimum
Wherein: the system comprises p1, p2 and p3 unit subsystems, a service power load (such as a condensate system and a wind and smoke system) and an N unit load.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211687293.6A CN115952900A (en) | 2022-12-27 | 2022-12-27 | AI analysis-based intelligent energy operation optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211687293.6A CN115952900A (en) | 2022-12-27 | 2022-12-27 | AI analysis-based intelligent energy operation optimization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115952900A true CN115952900A (en) | 2023-04-11 |
Family
ID=87287261
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211687293.6A Pending CN115952900A (en) | 2022-12-27 | 2022-12-27 | AI analysis-based intelligent energy operation optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115952900A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118195082A (en) * | 2024-04-01 | 2024-06-14 | 芯愿景数字科技(上海)有限公司 | Intelligent energy operation optimization method based on AI analysis and cloud platform |
| CN119357730A (en) * | 2024-09-12 | 2025-01-24 | 上海麦杰科技股份有限公司 | A method and system for optimizing the operating conditions of a thermal power unit |
| CN119397498A (en) * | 2025-01-06 | 2025-02-07 | 朗坤智慧科技股份有限公司 | A method, system, device and storage medium for optimizing working conditions based on historical data analysis |
-
2022
- 2022-12-27 CN CN202211687293.6A patent/CN115952900A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118195082A (en) * | 2024-04-01 | 2024-06-14 | 芯愿景数字科技(上海)有限公司 | Intelligent energy operation optimization method based on AI analysis and cloud platform |
| CN118195082B (en) * | 2024-04-01 | 2024-09-27 | 芯愿景数字科技(上海)有限公司 | Intelligent energy operation optimization method based on AI analysis and cloud platform |
| CN119357730A (en) * | 2024-09-12 | 2025-01-24 | 上海麦杰科技股份有限公司 | A method and system for optimizing the operating conditions of a thermal power unit |
| CN119397498A (en) * | 2025-01-06 | 2025-02-07 | 朗坤智慧科技股份有限公司 | A method, system, device and storage medium for optimizing working conditions based on historical data analysis |
| CN119397498B (en) * | 2025-01-06 | 2025-04-29 | 朗坤智慧科技股份有限公司 | A method, system, device and storage medium for optimizing working conditions based on historical data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115952900A (en) | AI analysis-based intelligent energy operation optimization method | |
| US20230229124A1 (en) | Operation control system and a control method for a gas-steam combined cycle generator unit | |
| CN106054104B (en) | A kind of intelligent electric meter failure real-time predicting method based on decision tree | |
| CN117013527A (en) | Distributed photovoltaic power generation power prediction method | |
| CN106204330A (en) | A kind of power distribution network intelligent diagnosis system | |
| CN118536410B (en) | Big data driven modeling-based energy consumption optimization decision analysis method and system | |
| CN111738462A (en) | Active service early warning method for emergency repair of electric power metering device | |
| CN116976707A (en) | User electricity consumption data anomaly analysis method and system based on electricity consumption data acquisition | |
| CN118246714B (en) | Water conservancy and hydropower engineering construction energy consumption analysis method and system | |
| CN107437135A (en) | A kind of novel energy-storing selection method | |
| CN119518705B (en) | Intelligent electricity load management method based on big data analysis | |
| Ma et al. | Economic operation evaluation of active distribution network based on fuzzy Borda method | |
| CN117610214A (en) | Intelligent power distribution network wiring planning method based on dynamic geographic features | |
| CN118885806A (en) | A method and system for predicting power load based on regional portrait | |
| CN116467658A (en) | Equipment fault tracing method based on Markov chain | |
| CN118504917A (en) | Power grid line planning system based on artificial intelligence | |
| CN119726632A (en) | A method for reducing power transmission and distribution losses in power grid | |
| Abghari et al. | Higher order mining for monitoring district heating substations | |
| CN114595839A (en) | A method and system for analyzing the failure cause of a smart electric energy meter | |
| CN119362609B (en) | A method for coordinated control of multi-energy power systems at spatiotemporal scales | |
| JP2005038098A (en) | Method of monitoring and executing equipment, facility operation status or transactions using data mining | |
| CN115656914B (en) | Intelligent ammeter metering accuracy detection method and device based on big data | |
| Balaji et al. | Enhancing Predictive Accuracy with Machine Learning and IoT Integration in Energy Management | |
| Hao et al. | Demand forecasting for rush repair spare parts of power equipment using fuzzy C-means clustering and the fuzzy decision tree | |
| Zhao et al. | Research on line-transformer-user topological anomaly recognition model based on multi-source data mining |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |