A method and system for evaluating data quality of an energy storage power station

Abstract

The application discloses an energy storage power station data quality evaluation method and system, and relates to the technical field of power station data evaluation.The method comprises the following steps: collecting energy storage power station operation data, and preprocessing the operation data; the operation data comprises battery pack temperature data, battery state data and converter operation data; a multi-source data cross-validation model is constructed, a plurality of control operations are triggered based on a preset control instruction sequence, the time sequence response characteristics of the battery state data and the converter operation data are obtained, and a data consistency index is calculated; the data consistency index comprises time sequence matching degree, numerical deviation degree and response delay degree; and the data quality score of the energy storage power station is calculated according to the data consistency index.The application realizes efficient preprocessing and accurate quality evaluation of the energy storage power station operation data by constructing the multi-source data cross-validation model and calculating the data consistency index, significantly improves the accuracy and reliability of data management, and optimizes the operation and maintenance of the energy storage power station.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power station data evaluation, in particular to a method and system for evaluating data quality of energy storage power stations. BACKGROUND

[0002] With the deepening of the construction of new power systems, energy storage power stations, as an important power regulation means, play a key role in power grid frequency regulation, peak shaving, valley filling and other application scenarios. In order to ensure the safe and stable operation of energy storage power stations, the industry generally uses distributed data acquisition technology to monitor the operating state of energy storage power stations in real time. The current mainstream data acquisition schemes include temperature monitoring systems based on optical fiber sensing, state parameter acquisition based on battery management systems, and electrical quantity measurement based on power conversion systems. These acquisition systems provide basic data support for the operation and maintenance and performance evaluation of energy storage power stations by real-time acquisition of battery temperature, voltage, current, power and other key parameters. However, due to the complex coupling relationship between the subsystems in the energy storage power station, combined with the influence of multiple factors such as the nonlinearity of battery characteristics, power conversion loss, communication link delay, the accuracy and reliability of the collected data are facing serious challenges.

[0003] The data quality evaluation methods in the prior art mainly have the following deficiencies: first, the traditional evaluation methods are often based on a single data source for analysis, and it is difficult to establish the correlation characteristics between the data of different subsystems, resulting in that the evaluation results cannot fully reflect the actual operating state of the energy storage power station; second, the existing evaluation index system is too simple, usually only through numerical comparison or statistical analysis for quality evaluation, lacking in-depth analysis of dynamic characteristics such as data time sequence characteristics and response characteristics, making it difficult to guarantee the accuracy and reliability of the evaluation results; third, the existing evaluation methods generally use fixed scoring standards, which cannot dynamically adjust the evaluation strategy according to the actual application scenarios of the energy storage power station, resulting in a disconnection between the evaluation results and the actual application needs; finally, there is a lack of systematic abnormality identification and standardization processing mechanism in the data preprocessing stage, affecting the accuracy of subsequent quality evaluation. SUMMARY

[0004] In view of the above problems, the present application provides a method and system for evaluating data quality of energy storage power stations, which can solve the problems mentioned in the background art.

[0005] To solve the above technical problems, the present application provides the following technical solutions: a kind of energy storage power station data quality evaluation method, comprising: collecting energy storage power station operating data, the operating data is preprocessed;Wherein, the operating data includes battery pack temperature data, battery state data and converter operating data;Cross-validation model of multi-source data is constructed, based on the control instruction sequence of pre-set triggers multiple control operations, the time sequence response characteristics of the battery state data and converter operating data are obtained, and data consistency index is calculated;Wherein, the data consistency index includes time sequence matching degree, numerical deviation degree and response delay degree;According to the data consistency index, the data quality score of energy storage power station is calculated.

[0006] As a preferred scheme of the energy storage power station data quality evaluation method described in the application, wherein: the construction process of the multi-source data cross-validation model includes the following steps: extracting voltage, current and power from the battery state data as battery management system standard feature data, extracting converter input and output voltage, current and power from the converter operating data as converter standard feature data;The numerical correspondence of the battery management system standard feature data and the converter standard feature data at the same sampling time point is calculated, and a feature mapping matrix is generated;According to the feature mapping matrix, a multi-source data cross-validation model based on time sequence correlation is established.

[0007] As a preferred scheme of the energy storage power station data quality evaluation method described in the application, wherein: according to the feature mapping matrix, a multi-source data cross-validation model based on time sequence correlation is established, including the following steps: the feature mapping matrix is divided into N time windows, the correlation coefficient of the standard feature data in each time window is calculated, and a time sequence correlation matrix is generated;The time transfer function of battery management system standard feature data and converter standard feature data is calculated based on the time sequence correlation matrix, and a data response model is constructed;The time sequence correlation matrix and the data response model are combined to form a multi-source data cross-validation model.

[0008] As a preferred scheme of the energy storage power station data quality evaluation method, the preset control instruction sequence triggers multiple control operations to obtain the time sequence response characteristics of the battery state data and the converter operation data, including the following steps: the remote control instruction sequence is generated by the power distribution master station system; the remote control instruction sequence includes the charging power curve and the discharging power curve; the battery temperature change rate of the energy storage power station is obtained, and if the battery temperature change rate exceeds the preset threshold, the charging power curve and the discharging power curve are adjusted based on the battery temperature change rate to generate a step power curve, so that the step power curve and the battery temperature change rate form a mapping relationship; the sampling time stamp of the battery state data and the sampling time stamp of the converter operation data are obtained, the time offset between the two is calculated, and if the time offset is not zero, a pulse remote control instruction is generated based on the time offset; the pulse remote control instruction sequence is sent to the control system of the energy storage power station, and the battery state data and the converter operation data are synchronously collected; and the battery state data and the converter operation data are combined to form a time sequence response data set.

[0009] As a preferred scheme of the energy storage power station data quality evaluation method, the preset control instruction sequence triggers multiple control operations to obtain the time sequence response characteristics of the battery state data and the converter operation data, including the following steps: the remote control instruction sequence is generated by the power distribution master station system; the remote control instruction sequence includes the charging power curve and the discharging power curve; the battery temperature change rate of the energy storage power station is obtained, and if the battery temperature change rate exceeds the preset threshold, the charging power curve and the discharging power curve are adjusted based on the battery temperature change rate to generate a step power curve, so that the step power curve and the battery temperature change rate form a mapping relationship; the sampling time stamp of the battery state data and the sampling time stamp of the converter operation data are obtained, the time offset between the two is calculated, and if the time offset is not zero, a pulse remote control instruction is generated based on the time offset; the pulse remote control instruction sequence is sent to the control system of the energy storage power station, and the battery state data and the converter operation data are synchronously collected; and the battery state data and the converter operation data are combined to form a time sequence response data set.

[0009] As a preferred scheme of the energy storage power station data quality evaluation method, the preset control instruction sequence triggers multiple control operations to obtain the time sequence response characteristics of the battery state data and the converter operation data, including the following steps: the remote control instruction sequence is generated by the power distribution master station system; the remote control instruction sequence includes the charging power curve and the discharging power curve; the battery temperature change rate of the energy storage power station is obtained, and if the battery temperature change rate exceeds the preset threshold, the charging power curve and the discharging power curve are adjusted based on the battery temperature change rate to generate a step power curve, so that the step power curve and the battery temperature change rate form a mapping relationship; the sampling time stamp of the battery state data and the sampling time stamp of the converter operation data are obtained, the time offset between the two is calculated, and if the time offset is not zero, a pulse remote control instruction is generated based on the time offset; the pulse remote control instruction sequence is sent to the control system of the energy storage power station, and the battery state data and the converter operation data are synchronously collected; and the battery state data and the converter operation data are combined to form a time sequence response data set.

[0010] As a preferred scheme of the energy storage power station data quality evaluation method, wherein: according to the data consistency index, the quality score of the energy storage power station data is calculated, including the following steps: the modified time sequence matching degree, the modified numerical deviation degree and the modified response delay degree are substituted into the score mapping model; wherein, the score mapping model adopts the form of piecewise continuous function, and the function slope of the score mapping model is increased in the critical interval of the data consistency index; based on the actual application scene of the energy storage power station, the importance characteristics of the modified time sequence matching degree, the modified numerical deviation degree and the modified response delay degree are determined; wherein, if the energy storage power station performs frequency modulation task, the modified response delay degree is set as the dominant feature; if the energy storage power station performs peak clipping and valley filling task, the modified numerical deviation degree is set as the dominant feature; the feature vector of the modified time sequence matching degree, the modified numerical deviation degree and the modified response delay degree is extracted, the feature vector is orthogonally transformed based on the importance characteristics, and the feature space of the data consistency index is constructed; the comprehensive characteristic value of the data consistency index is calculated in the feature space, and the comprehensive characteristic value is mapped to the quality score of the energy storage power station data based on the score mapping model.

[0011] As a preferred scheme of the energy storage power station data quality evaluation method, wherein: the battery temperature data is collected by distributed optical fiber winding technology; the battery state data is collected by the battery management system; the converter operation data is collected by the power conversion system; the preprocessing includes outlier identification and data standardization.

[0012] The application further provides an energy storage power station data quality evaluation system for the above-mentioned method, comprising: a collection and processing module for collecting energy storage power station operation data and preprocessing the operation data; a cross-validation module for constructing a multi-source data cross-validation model, triggering multiple control operations based on a preset control instruction sequence, obtaining the time sequence response characteristics of the battery state data and the converter operation data, and calculating the data consistency index; and a quality evaluation module for calculating the quality score of the energy storage power station data according to the data consistency index.

[0013] A computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the energy storage power station data quality evaluation method as described above when executing the computer program.

[0014] A computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of the energy storage power station data quality evaluation method as described above.

[0015] The beneficial effects of the present application: compared with the single data source or simple comparison method commonly used in the prior art, the present application calculates the data consistency index including time sequence matching degree, numerical deviation degree and response delay degree, which can not only more comprehensively reflect the actual operation condition of the energy storage power station, but also effectively identify data anomalies and improve the accuracy and reliability of data quality evaluation. In addition, the systematic evaluation process proposed by the present application makes the data quality management of the energy storage power station more standardized and intelligent, which has significant progress significance for optimizing the operation and maintenance strategy of the energy storage power station, prolonging the service life of the equipment and reducing the operation cost. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The overall flowchart of the energy storage power station data quality evaluation method proposed by the present application;

[0018] Figure 2 The system structure diagram of the energy storage power station data quality evaluation system proposed by the present application;

[0019] Figure 3 The computer device diagram in the energy storage power station data quality evaluation method proposed by the present application. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings in the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0021] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0022] Embodiment 1, refer to Figure 1 An energy storage power station data quality evaluation method is provided for an embodiment of the present application.

[0023] In the related art, the lack of systematic data collection and preprocessing mechanism leads to unstable data quality foundation; the time sequence correlation and dynamic response characteristics of the data are not fully considered in the evaluation process, making it difficult for the evaluation result to truly reflect the running state of the energy storage system; at the same time, due to the oversimplified scoring method and the lack of scene adaptability, it is difficult to meet the evaluation needs of the energy storage power station in different application scenarios, ultimately affecting the accuracy and reliability of the data quality evaluation result.

[0024] The present application provides a method for evaluating the data quality of an energy storage power station, which can effectively solve the above-mentioned problems. The following will be described in detail in combination with multiple embodiments.

[0025] Figure 1 A flowchart of a method for evaluating the data quality of an energy storage power station is shown, which includes:

[0026] S1: Collecting energy storage power station running data and preprocessing the running data;

[0027] S2: Building a multi-source data cross-validation model, triggering multiple control operations based on a pre-set control instruction sequence, obtaining the time sequence response characteristics of the running data, and calculating data consistency evaluation parameters;

[0028] S3: Calculating the data quality score of the energy storage power station according to the data consistency evaluation parameters.

[0029] Next, this embodiment will be described in detail one by one for S1-S3:

[0030] S1: Collecting energy storage power station running data and preprocessing the running data.

[0031] Specifically, the running data includes battery pack temperature data collected through distributed fiber winding technology, battery state data collected through a battery management system, and converter running data collected through a power conversion system. The collected running data is preprocessed through data cleaning and protocol operation, including outlier identification and data standardization.

[0032] Further, in this embodiment, in order to meet the demand for energy storage power station operation data reliability and quality evaluation, distributed sensing technology (this is prior art, which can be searched on the first page of Baidu) is adopted to collect energy storage power station operation data. Distributed fiber winding technology (this is prior art, which can be seen on the first page of Baidu) is adopted to realize high-precision real-time distributed monitoring of battery pack temperature by winding optical fibers on the surface of the battery pack, solving the problems of low precision and poor stability of traditional point-type temperature measurement systems. The battery management system collects key operating parameters such as voltage, current, and state of charge of the battery in real time through the built-in data acquisition module, providing a data basis for evaluating the operating state of the energy storage system. The power conversion system monitors parameters such as power, voltage, and current of the energy storage converter in real time through the built-in data acquisition interface of the system, providing data support for evaluating the energy conversion efficiency and response characteristics of the energy storage system.

[0033] The collected operation data is preprocessed through data cleaning and protocol operation. Please read this paragraph in detail to ensure the accuracy of subsequent data quality evaluation. The data cleaning operation identifies and processes abnormal conditions such as interference, missing, repetition, and errors in the collected data by setting the valid value range of different types of data, eliminates unreasonable data, and fills in data gaps, thereby ensuring the integrity and effectiveness of the data. The protocol operation converts the collected data from different systems and different dimensions into a standard format, including time stamp alignment, value normalization, and unit unification, to realize data standardization and standardization; through preprocessing, the collected data is purified and standardized, laying a foundation for subsequent data cross-validation and quality evaluation.

[0034] S2: Construct a multi-source data cross-validation model, trigger multiple control operations based on a pre-set control instruction sequence, obtain the time sequence response characteristics of the operation data, and calculate a data consistency index.

[0035] Specifically, in this embodiment, a multi-source data cross-validation model based on time sequence correlation is constructed, multiple remote control operations are triggered by the remote control curve generated by the power distribution master station, the time sequence response characteristics of the battery state data and the converter operation data are obtained, and the data consistency index is calculated based on the time sequence response characteristics. The data consistency index includes time sequence matching degree, value deviation degree, and response delay degree.

[0036] S2.1: Construct a multi-source data cross-validation model, including the following steps:

[0037] First, extract voltage, current, and power from the battery state data as battery management system standard feature data, and extract converter input and output voltage, current, and power from the converter operation data as converter standard feature data;

[0038] Secondly, the numerical correspondence relationship between the battery management system standard feature data and the converter standard feature data at the same sampling time point is calculated. First, time synchronization, confirm the timestamp format and reference time of the two systems; unify the timestamp to the same time reference; handle the case of different sampling frequencies, usually need to resample to the same frequency. Secondly, the data matching method: nearest neighbor matching: find the closest time point linear interpolation: interpolation between two sampling points zero order hold: use the nearest known data point to generate a feature mapping matrix;

[0039] Finally, a multi-source data cross-validation model based on time series correlation is established according to the feature mapping matrix, specifically:

[0040] The feature mapping matrix is divided into N time windows, and the correlation coefficient of the standard feature data in each time window is calculated to generate a time series correlation matrix. Specifically, the feature mapping matrix is divided into N time windows according to the preset time interval T, and the time interval T is 2-10 minutes; the Pearson correlation coefficient of the standard feature data in each time window is calculated, which represents the correlation strength between the battery management system standard feature data and the converter standard feature data; the Pearson correlation coefficients of the N time windows form a time series correlation matrix, which reflects the data correlation characteristics of the energy storage station in different operating stages.

[0041] Based on the time series correlation matrix, the time transfer function of the battery management system standard feature data and the converter standard feature data is calculated, and a data response model is constructed. Using the correlation strength information of the battery management system converter standard feature data and the converter standard feature data in each time window recorded in the time series correlation matrix, a time transfer function model describing the dynamic response relationship between the two groups of data is established by system identification method. Specifically, first, the battery management system standard feature data is taken as the input signal, and the converter standard feature data is taken as the output signal, and the order and coefficient of the system transfer function are obtained based on the least square method; secondly, the input signal is simulated by using the transfer function to obtain the theoretical output response; finally, the theoretical output response is compared with the actual converter data, and the model parameters are optimized by calculating the root mean square error, and finally a data response model is established which can accurately describe the conversion relationship of the battery management system data to the converter data, the model can be used to evaluate the data transmission characteristics and response delay between the two systems.

[0042] The time series correlation matrix is combined with the data response model to form a multi-source data cross-validation model. Specifically, the time series correlation matrix representing data correlation is fused with the data response model describing the dynamic response of the data to construct a comprehensive data validation framework. First, based on the correlation features extracted from the time series correlation matrix, a data consistency evaluation criterion is established to judge the matching degree of the battery management system and the converter data in the static characteristics; secondly, the theoretical response characteristics calculated by the data response model are used to establish a data dynamic characteristic evaluation criterion to evaluate the response matching degree and time delay of the two systems in the dynamic process; finally, the static characteristic evaluation and dynamic characteristic evaluation results are combined by weighting to form a multi-source data cross-validation model that can measure the data static consistency and dynamic response characteristics, which can fully reflect the data quality correlation characteristics between different subsystems in the energy storage power station, and provide reliable technical support for subsequent data quality scoring.

[0043] S2.2: Trigger multiple control operations based on the preset control instruction sequence to obtain the time series response characteristics of the battery state data and the converter operation data.

[0044] Specifically, the power distribution master station system generates a remote control instruction sequence containing a charging power curve and a discharging power curve. If the current battery temperature change rate of the energy storage power station exceeds the preset threshold, the charging power curve and the discharging power curve are adjusted based on the battery temperature change rate to generate a step charging and discharging power curve, so that the step power curve forms a mapping relationship with the battery temperature change rate; the sampling time stamps of the battery state data and the converter operation data are obtained, and the time offset between the two sampling time stamps is calculated. If the time offset is not zero, generate a pulse remote control instruction based on the time offset to verify the time delay characteristics in the data transmission link. Then, the generated pulse remote control instruction sequence is issued to the energy storage power station control system in chronological order. In the process of executing the control instruction, the battery state data of the battery management system and the converter operation data of the power conversion system are synchronously collected, the dynamic change characteristics of the key parameters such as voltage, current and power in the charging and discharging process are recorded, and a time series response data set containing temperature influence characteristics and time delay characteristics is formed. It should be noted that the battery state data and the converter operation data both include voltage parameters, current parameters and power parameters.

[0045] Preferably, through the adaptive remote control instruction generation mechanism based on the battery temperature change rate and the timestamp offset in S2.2, the dynamic matching of the energy storage power station control instruction and the actual operating state is realized, and the technical bottleneck that the instruction sequence is disconnected with the actual working condition in the traditional energy storage power station control scheme is broken through. In particular, by establishing a mapping relationship between the temperature change rate and the power curve, the control instruction can respond to the rapid change of the battery temperature in time, effectively preventing the risk of battery overheating, and significantly improving the safety performance of the energy storage system. At the same time, the pulse type remote control instruction generated based on the timestamp offset provides a reliable test means for evaluating the time delay characteristics of the data transmission link, solving the technical difficulty that the data transmission time delay is difficult to accurately identify and quantify in the traditional scheme. This adaptive control strategy not only improves the operation safety of the energy storage power station through temperature and time delay double constraints, but also lays a solid foundation for subsequent data quality evaluation by collecting complete time sequence response characteristic data, realizes the deep integration of the energy storage power station control process and data quality evaluation, and further guarantees the reliable operation of the energy storage power station under complex working conditions.

[0046] S2.3: Based on the time sequence response characteristics, calculate the data consistency index, the specific process is as follows:

[0047] First, according to the time sequence response characteristics, the initial time sequence matching degree of the battery state data and the converter operating data is calculated. The initial time sequence matching degree is measured by calculating the correlation of the battery management system data and the converter data in the time dimension, and statistical methods such as Pearson correlation coefficient and dynamic time warping DTW can be used to quantitatively evaluate the change trend of the two groups of data sequences in the same time window, and a matching degree value between 0 and 1 is obtained. The initial time sequence matching degree is corrected, specifically: the state of charge of the battery is obtained from the battery state data, and it is judged whether the state of charge is in the nonlinear change interval. If yes, the change rate of the state of charge is calculated, and a functional mapping relationship between the initial time sequence matching degree and the change rate of the state of charge is established to obtain the corrected time sequence matching degree.

[0048] Second, according to the time sequence response characteristics, the initial numerical deviation degree of the battery state data and the converter operating data is calculated, and the initial numerical deviation degree is corrected, specifically: the power conversion efficiency of the energy storage power station is obtained from the converter operating data, and if the power conversion efficiency is lower than the preset nominal value, the theoretical power loss value is calculated based on the converter topology structure of the energy storage power station, and the theoretical power loss value is used as a correction term to calculate the corrected numerical deviation degree.

[0049] Then, according to the time sequence response characteristic, the initial response delay degree of the energy storage power station to the remote control instruction sequence is calculated, and the initial response delay degree is corrected, specifically: the communication message between the battery management system and the power conversion system is obtained, if the transmission congestion of the communication message is detected, the link time delay is calculated based on the time stamp of the communication message, the link time delay is deducted from the initial response delay degree, and the corrected response delay degree is obtained;

[0050] Finally, the corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree are normalized to obtain the data consistency index.

[0051] Preferably, through the three-stage data consistency evaluation method based on the time sequence response characteristic designed in S2.3, the accurate quantification of the data quality of the energy storage power station is realized. The method first establishes a calculation model of the initial evaluation index based on the time sequence response characteristic: the initial time sequence matching degree is calculated by the cross-correlation function, the initial numerical deviation degree is determined by the normalized root mean square error, and the initial response delay degree is obtained based on the sliding time window statistics; Secondly, the targeted correction mechanism is innovatively introduced: by establishing the mapping relationship between the state of charge change rate and the initial time sequence matching degree, the time sequence matching error caused by the nonlinear characteristics of the battery is solved; by taking the theoretical power loss of the converter as the correction term, the numerical deviation caused by the power conversion process is overcome; by identifying and eliminating the communication link time delay, the interference of the communication fluctuation on the response delay evaluation is eliminated; Finally, the unified quantification of the three correction indexes is realized by using the normalization processing. This three-stage evaluation method not only establishes a reasonable evaluation benchmark in the initial calculation stage, but also fully considers the physical characteristics and operation constraints of the energy storage system in the correction stage, effectively improves the accuracy and reliability of the evaluation results through the systematic correction means, and lays a solid technical foundation for the subsequent data quality scoring.

[0052] It should be noted that S2 realizes the comprehensive evaluation of the data reliability of the energy storage power station by constructing a data interaction verification framework of the battery management system and the power conversion system, combining an adaptive control instruction generation strategy and a multi-dimensional data quality evaluation mechanism. S2 first designs an adaptive remote control instruction sequence based on the temperature change rate and the timestamp offset, breaking through the technical limitations of the mutual separation of control strategies and data acquisition in traditional energy storage systems. Second, through the coordinated cooperation of the power distribution master station system, the battery management system and the power conversion system, the data linkage collection among the subsystems of the energy storage power station is realized. Finally, by introducing a data consistency evaluation method considering the nonlinearity of state of charge, power loss and communication delay, the technical difficulty of accurately quantifying data quality in traditional evaluation schemes is solved. This cross-validation mechanism based on multi-source data provides a reliable basis for the safe operation and performance optimization of the energy storage power station, has important engineering application value for improving the operation reliability of the energy storage power station, prolonging the service life of the equipment and reducing the operation and maintenance cost, and promotes the development of the energy storage power station to a more efficient and reliable direction.

[0053] S3: According to the data consistency index, calculate the data quality score of the energy storage power station.

[0054] S3.1: Substitute the corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree in S2.3 into the score mapping model,

[0055] wherein the score mapping model adopts the form of a piecewise continuous function, and the function slope of the score mapping model is increased in the critical interval of the data consistency index.

[0056] S3.2: Determine the importance features of the corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree based on the actual application scenario of the energy storage power station.

[0057] wherein if the energy storage power station performs frequency modulation tasks, the corrected response delay degree is set as the dominant feature; if the energy storage power station performs peak clipping and valley filling tasks, the corrected numerical deviation degree is set as the dominant feature.

[0058] S3.3: Extract the feature vectors of the corrected three indexes by using the principal component analysis method, and perform orthogonal transformation on the feature vectors based on the importance features, to construct the feature space of the data consistency index.

[0059] S3.4: Calculate the comprehensive characteristic value of the data consistency index in the feature space, and map the comprehensive characteristic value to the data quality score of the energy storage power station based on the score mapping model.

[0060] Preferably, S3 first establishes a nonlinear correspondence between the correction index and the score result by designing a score mapping model in the form of a segmented continuous function, wherein the identification ability of the data quality difference close to the critical state is improved through the innovative design of increasing the function slope in the critical interval; secondly, through the introduction of an importance feature allocation mechanism based on the application scenario, the technical difficulty that the traditional scoring method cannot adapt to different application scenarios is solved, especially in typical application scenarios such as frequency modulation and peak clipping, by setting different correction indexes as dominant features, the precise matching of the scoring result and the application demand is realized; finally, the mathematical tools of principal component analysis and orthogonal transformation are used to construct the feature space, breaking through the limitations of the traditional simple weighting method, by extracting the feature vector of the correction index and combining the importance feature for orthogonal transformation, the decoupling between indexes in the scoring process is realized, so that the scoring result can more comprehensively reflect the essential features of data quality. This scoring method which combines nonlinear mapping, scene adaptation and feature decoupling not only provides a new data quality evaluation paradigm, but also provides reliable data support for the operation optimization and maintenance decision of energy storage power station through the fine differentiation of the scoring result, which has important practical value for improving the operation efficiency of energy storage power station in different application scenarios.

[0061] In summary, compared with the single data source or simple comparison method commonly used in the prior art, the present application calculates the data consistency index including time sequence matching degree, numerical deviation degree and response delay degree, which can not only more comprehensively reflect the actual operation status of the energy storage power station, but also effectively identify data anomalies and improve the accuracy and reliability of data quality evaluation. In addition, the systematic evaluation process proposed by the present application makes the data quality management of the energy storage power station more standardized and intelligent, which has significant progress significance for optimizing the operation and maintenance strategy of the energy storage power station, prolonging the service life of the equipment and reducing the operating cost.

[0062] Embodiment 2, refer to Figure 2 For an embodiment of the present application, a data quality evaluation system for an energy storage power station is provided, comprising: a collection and processing module for collecting and preprocessing the operation data of the energy storage power station; a cross-validation module for constructing a multi-source data cross-validation model, triggering multiple control operations based on a pre-set control instruction sequence, obtaining the time sequence response characteristics of the battery state data and the converter operation data, and calculating the data consistency index; a quality evaluation module for calculating the data quality score of the energy storage power station according to the data consistency index.

[0063] Embodiment 3, refer to Figure 3For one embodiment of the present application, different from the previous embodiment, the function, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or the part of the technical solution that essentially contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0064] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a list of executable instructions for implementing logic functions, which can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or in conjunction with these instructions execution systems, apparatuses, or devices. For the purpose of this specification, the "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by an instruction execution system, apparatus, or device, or in conjunction with these instruction execution systems, apparatuses, or devices.

[0065] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical devices), a portable computer diskette (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by electronic editing, interpretation, or necessary processing, and then stored in a computer memory if necessary.

[0066] It should be understood that various parts of the present application can be implemented in hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and as in another embodiment, it can be implemented using any one or a combination of the following technologies known in the art: discrete logic circuit with logic gate circuit for implementing logical functions on data signals, application specific integrated circuit with suitable combination logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) and the like.

[0067] Embodiment 4, which is an embodiment of the present application, provides a method for evaluating data quality of an energy storage power station. In order to verify the beneficial effects of the present application, economic benefit calculation and simulation experiments are carried out for scientific demonstration.

[0068] In order to verify the effectiveness of the technical scheme of the present application in data quality evaluation of the energy storage power station, a large energy storage power station is selected as the experimental object in this embodiment, and a 30-day comparative experiment is carried out under two typical application scenarios of frequency modulation and peak clipping and filling. During the experiment, the operation data of the energy storage power station are evaluated by the present application scheme (denoted as scheme A) and the prior art scheme (denoted as scheme B) respectively. Scheme B uses the traditional single data source evaluation method, and only calculates the data reliability through simple numerical comparison; while scheme A uses the multi-source data cross-validation model proposed in the present application, and combines the adaptive control strategy and the three-stage data consistency evaluation method for comprehensive evaluation.

[0069] During the experimental implementation, this embodiment focuses on key performance indicators such as data acquisition accuracy, evaluation index stability and scene adaptability. Specifically, by deploying a distributed optical fiber sensing system on the surface of the battery pack, temperature distribution data are collected; battery state parameters and inverter operation parameters are collected through the battery management system and the power conversion system respectively. At the same time, based on the adaptive control instruction sequence generated by the power distribution master station system, multiple control operations are triggered, and system response data are collected. Through pre-processing, cross-validation and quality scoring of the collected data, evaluation results under different schemes are obtained.

[0070] Table 1: Comparison of data quality evaluation indicators under different application scenarios

[0071]

[0072] As shown in Table 1, the application has a significant advantage in the accuracy and stability of evaluation indicators. In the frequency modulation scenario, the timing matching degree of scheme A reaches 0.95, which is 0.13 higher than that of scheme B, and the standard deviation is only 0.03, indicating that the evaluation results of the application have higher timing consistency; in terms of numerical deviation degree, the average of scheme A is 0.93, which is 9.4% higher than that of scheme B, and the standard deviation is reduced by 55.6%, which confirms that the correction mechanism based on power loss compensation proposed by the application can effectively improve the accuracy of numerical evaluation; in the response delay degree index, scheme A performs more outstandingly, with an average of 0.96, which is 21.5% higher than that of scheme B, and the standard deviation is only 0.02, fully verifying the effectiveness of the communication delay identification and elimination mechanism designed by the application. In the peak clipping and valley filling scenario, the three indicators also show good performance, especially the numerical deviation degree reaches a high level of 0.95, which matches the high requirement for power accuracy in this scenario, reflecting the scene adaptability of the application. Through comparative analysis of the standard deviation, it can be found that the evaluation index fluctuation of scheme A is significantly smaller than that of scheme B in both scenarios, which shows that the three-stage evaluation method proposed by the application can effectively suppress external interference and maintain the stability of the evaluation results.

[0073] Table 2 Evaluation performance comparison under different temperature conditions

[0074]

[0075] Table 2 reveals the robustness advantage of the application under temperature change conditions. In the standard temperature range (20-30℃), the data acquisition accuracy of scheme A reaches 97.5%, the evaluation index stability reaches 0.95, and the fault recognition rate reaches 96.8%, all of which are significantly higher than those of scheme B; as the temperature rises, the performance indicators of both schemes decrease, but scheme A shows stronger anti-interference ability. Specifically, in the medium temperature range (30-40℃), the data acquisition accuracy of scheme A only decreases by 0.7 percentage points, while that of scheme B decreases by 2.7 percentage points; in the high temperature range (40-50℃), the evaluation index stability of scheme A remains at a high level of 0.91, while that of scheme B decreases to 0.73. By calculating the performance decay rate, it can be known that from the low temperature to the high temperature range, the average decay rate of each index of scheme A is 2.4%, which is much lower than that of scheme B, which is 8.7%. It is worth noting that although the average time consumption of scheme A is about 0.4-0.5 seconds longer than that of scheme B, this is mainly due to the introduction of multi-source data cross-validation and three-stage evaluation mechanism, and considering that the real-time requirement of energy storage power station data evaluation is not strict, this time consumption is acceptable. Overall, the experimental data powerfully confirms the comprehensive advantages of the application in evaluation accuracy, stability and environmental adaptability, etc.

[0076] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for evaluating data quality of an energy storage power plant, characterized in that, The method comprises the following steps: Collecting energy storage power station operation data, and preprocessing the operation data; wherein the operation data includes battery pack temperature data, battery state data and converter operation data; Constructing a multi-source data cross-validation model, triggering multiple control operations based on a preset control instruction sequence, obtaining the time sequence response characteristics of the battery state data and the converter operation data, and calculating a data consistency index; wherein the data consistency index includes time sequence matching degree, numerical deviation degree and response delay degree; According to the data consistency index, calculating the data quality score of the energy storage power station; Based on the time sequence response characteristics, the data consistency index is calculated, specifically: According to the time sequence response characteristics, the initial time sequence matching degree of the battery state data and the converter operation data is calculated, and the initial time sequence matching degree is corrected, specifically: the state of charge of the battery is obtained from the battery state data, it is judged whether the state of charge is in a nonlinear change interval, if yes, the change rate of the state of charge is calculated, the function mapping relationship between the initial time sequence matching degree and the change rate of the state of charge is established, and the corrected time sequence matching degree is obtained; According to the time sequence response characteristics, the initial numerical deviation degree of the battery state data and the converter operation data is calculated, and the initial numerical deviation degree is corrected, specifically: the power conversion efficiency of the energy storage power station is obtained from the converter operation data, if the power conversion efficiency is lower than the preset nominal value, the theoretical power loss value is calculated based on the converter topology structure of the energy storage power station, and the theoretical power loss value is taken as a correction term to calculate the corrected numerical deviation degree; According to the time sequence response characteristics, the initial response delay degree of the energy storage power station to the remote control instruction sequence is calculated, and the initial response delay degree is corrected, specifically: the communication message between the battery management system and the power conversion system is obtained, if the transmission congestion of the communication message is detected, the link delay is calculated based on the timestamp of the communication message, the link delay is deducted from the initial response delay degree, and the corrected response delay degree is obtained; The corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree are normalized to obtain the data consistency index; According to the data consistency index, the data quality score of the energy storage power station is calculated, including the following steps: The corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree are substituted into a scoring mapping model; wherein the scoring mapping model adopts the form of piecewise continuous function, and the function slope of the scoring mapping model is increased in the critical interval of the data consistency index; Based on the actual application scene of the energy storage power station, the importance features of the corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree are determined; wherein if the energy storage power station performs frequency modulation task, the corrected response delay degree is set as the dominant feature; if the energy storage power station performs peak clipping and valley filling task, the corrected numerical deviation degree is set as the dominant feature. extract a feature vector of the corrected time sequence matching degree, the corrected numerical deviation degree and the corrected response delay degree, perform orthogonal transformation on the feature vector based on the importance feature, and construct a feature space of the data consistency index; calculate a comprehensive feature value of the data consistency index in the feature space, and map the comprehensive feature value to the energy storage power station data quality score based on the scoring mapping model.

2. The energy storage plant data quality assessment method of claim 1, wherein: The construction process of the multi-source data cross-validation model includes the following steps: extracting voltage, current and power from the battery state data as battery management system standard feature data, and extracting transformer input and output voltage, current and power from the converter operation data as converter standard feature data; calculating the numerical correspondence of the battery management system standard feature data and the converter standard feature data at the same sampling time point to generate a feature mapping matrix; establishing a multi-source data cross-validation model based on time sequence correlation according to the feature mapping matrix.

3. The energy storage plant data quality assessment method of claim 2, wherein: According to the feature mapping matrix, a multi-source data cross-validation model based on time sequence correlation is established, including the following steps: divide the feature mapping matrix into N time windows, calculate the correlation coefficient of the standard feature data in each time window, and generate a time sequence correlation matrix; based on the time sequence correlation matrix, calculate the time transfer function of the battery management system standard feature data and the converter standard feature data to construct a data response model; combine the time sequence correlation matrix and the data response model to form a multi-source data cross-validation model.

4. The energy storage plant data quality assessment method of claim 1, wherein: Based on the preset control instruction sequence, trigger multiple control operations to obtain the time sequence response characteristics of the battery state data and the converter operation data, including the following steps: generate the remote control instruction sequence through the power distribution master station system; wherein the remote control instruction sequence includes a charging power curve and a discharging power curve; obtain the battery temperature change rate of the energy storage power station, if the battery temperature change rate exceeds a preset threshold, adjust the charging power curve and the discharging power curve based on the battery temperature change rate to generate a step power curve, so that the step power curve and the battery temperature change rate form a mapping relationship; obtain the sampling time stamp of the battery state data and the sampling time stamp of the converter operation data, calculate the time offset between them, if the time offset is not zero, generate a pulse type remote control instruction based on the time offset; issue the pulse type remote control instruction sequence to the control system of the energy storage power station, and synchronously collect the battery state data and the converter operation data; combine the battery state data and the converter operation data to form a time sequence response data set.

5. The energy storage plant data quality assessment method of claim 1, wherein: The battery pack temperature data is collected by distributed optical fiber winding technology; the battery state data is collected by the battery management system; the converter operation data is collected by the power conversion system; The preprocessing includes outlier identification and data standardization.

6. A system for the data quality assessment method of the energy storage power plant of any of claims 1 to 5, characterized in that, It includes: a collection and processing module for collecting energy storage power station operation data and preprocessing the operation data; The cross-validation module is configured to construct a multi-source data cross-validation model, trigger multiple control operations based on a preset control instruction sequence, obtain time sequence response characteristics of the battery state data and the converter operation data, and calculate a data consistency index. The quality evaluation module is configured to calculate a data quality score of the energy storage power station according to the data consistency index. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. The processor implements the steps of the energy storage power station data quality evaluation method in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the energy storage power station data quality evaluation method in any one of claims 1 to 5.

Images