Disclosure of Invention
The application provides a distributed energy storage power station fault multistage diagnosis method and system, which at least solve the technical problem of how to efficiently utilize the large-quantity and large-quantity storage data and accurately diagnose the faults of the energy storage power station.
An embodiment of a first aspect of the present application provides a method for multistage fault diagnosis of a distributed energy storage power station, the method including:
collecting operation data of each battery in the distributed energy storage power station, and preprocessing the operation data;
respectively determining multistage diagnosis data corresponding to the distributed energy storage power station according to the preprocessed operation data;
and predicting the faults of the distributed energy storage power station according to the multi-stage diagnosis data.
Preferably, the determining the multi-stage diagnosis data corresponding to the distributed energy storage power station according to the preprocessed operation data respectively includes
Screening out characteristic values of all batteries in the distributed energy storage power station from the operation data, and determining basic data index values of all batteries based on the characteristic values of all batteries;
screening primary diagnosis data of each battery in the distributed energy storage power station from the data, and determining a primary diagnosis data index value of each battery based on the primary diagnosis data of each battery;
determining voltage standard components, coulomb efficiency and IC curve peak values of each battery according to the operation data, and determining secondary diagnosis data index values of each battery based on the voltage standard components, the coulomb efficiency and the IC curve peak values;
respectively determining weight coefficients of secondary diagnosis data index values of all batteries, and determining three-level diagnosis data index values of all batteries based on the weight coefficients;
wherein the multi-level diagnostic data comprises: basic data index value, primary diagnosis data index value, secondary diagnosis data index value and tertiary diagnosis data index value of each battery.
Further, the calculation formula of the voltage standard score of each battery is as follows:
in the method, in the process of the invention,
for the voltage standard of the ith battery in a distributed energy storage power station, < >>
For the voltage value of the ith battery in the decentralized energy storage power station, is->
For the average value of the voltages of the batteries in the decentralized energy storage power station, < >>
Is the standard deviation of the voltage;
the coulombic efficiency of each cell is calculated as follows:
in the method, in the process of the invention,
coulombic efficiency for the ith cell in a decentralized energy storage power station, < >>
For the discharge capacitance of the ith battery in a decentralized energy storage power station, < >>
Is the charge capacitance of the ith cell in the decentralized energy storage power station.
Further, the calculation formula of the weight coefficient of the index value of the secondary diagnosis data of each battery is as follows:
in the method, in the process of the invention,
for the value of the d index of the secondary diagnostic data index values of the i-th battery in the distributed energy storage power station,/->
Standard deviation of the (d) th index of the (i) th battery in the distributed energy storage power station>
Mean value of d index in index values of secondary diagnosis data of i-th battery in distributed energy storage power station, +.>
The number of the secondary diagnosis data indexes.
Further, the determining the three-level diagnostic data index value of each battery based on the weight coefficient includes:
determining a single factor diagnostic value of each index in the secondary diagnostic data indexes of each battery;
and determining three-level diagnosis data index values of each battery according to the weight coefficient and the single-factor diagnosis value of each index.
Further, the calculation formula of the three-level diagnosis data index value of each battery is as follows:
in the method, in the process of the invention,
three-level diagnostic data index value for the ith battery in a distributed energy storage power station,/for the power station>
For the matrix of single factor diagnostic values of the ith battery in the decentralized energy storage power station,/for the battery>
Is a weight coefficient matrix.
Further, the predicting the fault of the distributed energy storage power station according to the multi-stage diagnosis data includes:
performing standardization processing on the basic data index value, the first-level diagnosis data index value, the second-level diagnosis data index value and the third-level diagnosis data index value of each battery;
substituting the basic data index value, the first-level diagnosis data index value, the second-level diagnosis data index value and the third-level diagnosis data index value of each battery after the standardized processing into a pre-built comprehensive prediction model to obtain a fault prediction result of the distributed energy storage power station.
Further, the method further comprises:
and displaying the fault prediction result of the distributed energy storage power station.
Embodiments of a second aspect of the present application provide a distributed energy storage power station fault multistage diagnostic system, the system comprising:
the collecting module is used for collecting the operation data of each battery in the distributed energy storage power station and preprocessing the operation data;
the determining module is used for respectively determining multistage diagnosis data corresponding to the distributed energy storage power station according to the preprocessed operation data;
and the diagnosis module is used for predicting the faults of the distributed energy storage power station according to the multi-stage diagnosis data.
Preferably, the determining module comprises
The first determining unit is used for screening out the characteristic values of the batteries in the distributed energy storage power station from the operation data and determining basic data index values of the batteries based on the characteristic values of the batteries;
the second determining unit is used for screening out primary diagnosis data of each battery in the distributed energy storage power station from the data and determining a primary diagnosis data index value of each battery based on the primary diagnosis data of each battery;
a third determining unit for determining a voltage standard score, coulomb efficiency, and an IC curve peak value of each battery according to the operation data, and determining a second-level diagnostic data index value of each battery based on the voltage standard score, the coulomb efficiency, and the IC curve peak value;
a fourth determining unit for determining weight coefficients of the secondary diagnosis data index values of the respective batteries, respectively, and determining the tertiary diagnosis data index values of the respective batteries based on the weight coefficients;
wherein the multi-level diagnostic data comprises: basic data index value, primary diagnosis data index value, secondary diagnosis data index value and tertiary diagnosis data index value of each battery.
The technical scheme provided by the embodiment of the application at least brings the following beneficial effects:
the application provides a distributed energy storage power station fault multistage diagnosis method and system, wherein the method comprises the following steps: collecting operation data of each battery in the distributed energy storage power station, and preprocessing the operation data; respectively determining multistage diagnosis data corresponding to the distributed energy storage power station according to the preprocessed operation data; and predicting the faults of the distributed energy storage power station according to the multi-stage diagnosis data. According to the technical scheme, the compatibility is high, and meanwhile, faults of the distributed energy storage power station can be accurately diagnosed.
Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present application and are not to be construed as limiting the present application.
The method and the system for multistage fault diagnosis of the distributed energy storage power station provided by the application comprise the following steps: collecting operation data of each battery in the distributed energy storage power station, and preprocessing the operation data; respectively determining multistage diagnosis data corresponding to the distributed energy storage power station according to the preprocessed operation data; and predicting the faults of the distributed energy storage power station according to the multi-stage diagnosis data. According to the technical scheme, the compatibility is high, and meanwhile, faults of the distributed energy storage power station can be accurately diagnosed.
The following describes a fault multistage diagnosis method and system for a distributed energy storage power station according to an embodiment of the present application with reference to the accompanying drawings.
Example 1
Fig. 1 is a flowchart of a fault multistage diagnosis method for a distributed energy storage power station according to an embodiment of the present application, as shown in fig. 1, the method includes:
step 1: collecting operation data of each battery in the distributed energy storage power station, and preprocessing the operation data;
in an embodiment of the disclosure, operational data of each battery in a decentralized energy storage power station is collected in real time from a multi-stage battery data collection system (Battery data acquisition system, BDAS);
it should be noted that, as shown in fig. 2, the BDAS are connected with the battery cells, the modules and the clusters one by one, the BDAS connected with the battery cells are primary BDAS, the BDAS connected with the battery modules are secondary BDAS, the BDAS connected with the battery clusters are tertiary BDAS, and the primary BDAS are summarized at the secondary and summarized at the tertiary to form a multi-stage BDAS.
Further, preprocessing the operation data includes:
and (3) cleaning operation is carried out on the operation data of each battery: processing missing value data, noise data and format arrangement of the data;
and carrying out replacement assignment on the missing value or the abnormal value which is obviously unreasonable by adopting a time number average value before and after the value.
Step 2: respectively determining multistage diagnosis data corresponding to the distributed energy storage power station according to the preprocessed operation data;
wherein the multi-level diagnostic data comprises: basic data index value, primary diagnosis data index value, secondary diagnosis data index value and tertiary diagnosis data index value of each battery.
In an embodiment of the present disclosure, the step 2 specifically includes:
step 2-1: screening out characteristic values of all batteries in the distributed energy storage power station from the operation data, and determining basic data index values of all batteries based on the characteristic values of all batteries;
the feature values include: battery charge-discharge capacity, voltage, current, charge-discharge time, power, battery internal resistance and temperature.
One or more of battery charge-discharge capacity, voltage, current, charge-discharge time, power, battery internal resistance and temperature are used as basic data index values of each battery.
Step 2-2: screening primary diagnosis data of each battery in the distributed energy storage power station from the data, and determining a primary diagnosis data index value of each battery based on the primary diagnosis data of each battery;
the threshold value of the first-level diagnosis data index is obtained through the correlation standard and statistics according to the index definition, the threshold value is detected, whether the first-level diagnosis data exceeds the threshold value is judged, and if the first-level diagnosis data exceeds the threshold value, the first-level diagnosis data is regarded as abnormal.
Step 2-3: and determining voltage standard components, coulomb efficiency and IC curve peak values of the batteries according to the operation data, and determining secondary diagnosis data index values of the batteries based on the voltage standard components, the coulomb efficiency and the IC curve peak values, wherein the voltage standard components, the coulomb efficiency and the IC curve peak values are used as the secondary diagnosis data index values of the batteries.
Further, the calculation formula of the voltage standard score of each battery is as follows:
in the method, in the process of the invention,
for the voltage standard of the ith battery in a distributed energy storage power station, < >>
For the voltage value of the ith battery in the decentralized energy storage power station, is->
For the average value of the voltages of the batteries in the decentralized energy storage power station, < >>
Is the standard deviation of the voltage;
the coulombic efficiency of each cell is calculated as follows:
in the method, in the process of the invention,
coulombic efficiency for the ith cell in a decentralized energy storage power station, < >>
For the discharge capacitance of the ith battery in a decentralized energy storage power station, < >>
Is the charge capacitance of the ith cell in the decentralized energy storage power station.
The determination of the peak value of the IC curve of each battery comprises the following steps: obtaining a battery OCV curve according to a first-order equivalent circuit model to obtain dV, obtaining dQ according to combined ampere-hour integration and Kalman filtering calculation, finally obtaining a relation curve of which the IC curve is dQ/dV-SOC, and determining the IC curve peak value of the battery in the relation curve of which the IC curve is dQ/dV-SOC.
The consistency analysis is carried out on the voltage standard components, the coulomb efficiency and the IC curve peak value of each battery, whether the battery with poor consistency exists is judged, and if so, the battery is regarded as abnormal;
illustratively, determining a voltage standard deviation median value according to the voltage standard deviation, if the |voltage standard deviation median value| <1, the battery is healthy; if the median value of the voltage standard score of 1< | <2, the battery is inconsistent; if the median value of the voltage standard division of 2< | <3, the battery fault phenomenon starts to worsen; if the median value of the voltage standard is 3< |, the battery needs to intervene in the fault site.
Step 2-4: and respectively determining the weight coefficient of the secondary diagnosis data index value of each battery, and determining the tertiary diagnosis data index value of each battery based on the weight coefficient.
Wherein, the calculation formula of the weight coefficient of the index value of the secondary diagnosis data of each battery is as follows:
in the method, in the process of the invention,
for the value of the d index of the secondary diagnostic data index values of the i-th battery in the distributed energy storage power station,/->
The d-th finger in the index value of the secondary diagnostic data for the i-th battery in a decentralized energy storage power stationStandard deviation of the target->
Mean value of d index in index values of secondary diagnosis data of i-th battery in distributed energy storage power station, +.>
The number of the secondary diagnosis data indexes.
Further, the determining the three-level diagnostic data index value of each battery based on the weight coefficient includes:
determining a single factor diagnostic value of each index in the secondary diagnostic data indexes of each battery;
and determining three-level diagnosis data index values of each battery according to the weight coefficient and the single-factor diagnosis value of each index.
The determining the single-factor diagnosis value of each of the two-level diagnosis data indexes of each battery includes:
determining a membership function of each index in the secondary diagnosis data indexes, and calculating a membership function value of each index;
and taking the membership function value of each index as a single factor diagnosis value of each index.
The membership function of the voltage standard index is parabolic, the membership function of the coulomb efficiency index is S-shaped, and the membership function of the IC curve peak value index is linear.
The parabolic membership function
The index is the most beneficial to the battery state in a certain range, and the increase or decrease of the index from the range is unfavorable for the battery health, and the calculation formula is as follows:
the S-shaped membership function
Is indicated as a certain indexA range that is positively or negatively correlated with the battery state, beyond which the battery state is not greatly affected, is used for a number of beneficial or deleterious factors, and is calculated as:
the linear membership function
The index has no obvious upper and lower limits, and the linear relation is formed between the change of the index and the state of health of the battery, and the calculation formula is as follows: />
;
Is the index lower limit value; />
Is the optimal numerical lower limit; />
Is the optimal upper numerical limit; />
Is the index upper limit value.
Further, the calculation formula of the three-level diagnosis data index value of each battery is as follows:
in the method, in the process of the invention,
three-level diagnostic data index value for the ith battery in a distributed energy storage power station,/for the power station>
For the matrix of single factor diagnostic values of the ith battery in the decentralized energy storage power station,/for the battery>
Is a weight coefficient matrix.
Step 3: and predicting the faults of the distributed energy storage power station according to the multi-stage diagnosis data.
In an embodiment of the present disclosure, the step 3 specifically includes:
step 3-1: performing standardization processing on the basic data index value, the first-level diagnosis data index value, the second-level diagnosis data index value and the third-level diagnosis data index value of each battery;
step 3-2: substituting the basic data index value, the first-level diagnosis data index value, the second-level diagnosis data index value and the third-level diagnosis data index value of each battery after the standardized processing into a pre-built comprehensive prediction model to obtain a fault prediction result of the distributed energy storage power station.
The pre-built comprehensive prediction model can be obtained by fusing one or more models of a machine learning model (KNN, k-Means, SVM, logistic regression), an integrated learning model (XGBoost, lightGBM, catBoost) and a deep learning model (LSTM, BP, CNN).
The training of the model includes:
obtaining basic data index values, primary diagnosis data index values, secondary diagnosis data index values, tertiary diagnosis data index values and corresponding fault results of each battery in a historical period, and carrying out dimensionless standardization processing to obtain a model construction data set;
and carrying out training by taking the model construction data set into the comprehensive prediction model, and outputting a fault prediction model result after the loss function converges through iterative optimization to obtain a trained comprehensive prediction model.
In an embodiment of the disclosure, the method further comprises:
and sending the fault prediction result of the distributed energy storage power station to an EMS system for display, as shown in figure 2.
It should be noted that, in fig. 2, the PCS is a distributed energy storage power station, the BDAS is used for collecting and analyzing real-time data of battery cells, modules and clusters, the database platform is used for storing data, and the diagnostic center is used for comprehensively summarizing and analyzing data uploaded by the BDAS at each level and feeding back the result to the EMS system.
In summary, according to the multi-stage fault diagnosis method for the distributed energy storage power station, which is provided by the embodiment, the battery is connected through the BDAS, the data acquisition is more accurate, meanwhile, the first-stage diagnosis threshold value judgment speed is high, and the method can be used for quickly judging a relatively serious shallow layer secondary fault exceeding the threshold value as the first-stage abnormality detection; the secondary diagnosis is based on battery consistency, and the health state of the energy storage power station is analyzed from various single factor angles, so that the performance and fault condition of the battery can be comprehensively detected in various aspects and multiple angles; the three-level diagnosis is based on a battery operation experience model, the actual operation condition of the battery is predicted and judged more accurately by combining the chemical characteristics and the data characteristics of the battery, and the method has higher compatibility, can flexibly match different project requirements and is very friendly in interface.
Example two
Fig. 3 is a block diagram of a fault multistage diagnosis system of a distributed energy storage power station according to an embodiment of the present application, as shown in fig. 3, the system includes:
the collecting module 100 is used for collecting the operation data of each battery in the distributed energy storage power station and preprocessing the operation data;
the determining module 200 is configured to determine multistage diagnostic data corresponding to the distributed energy storage power station according to the preprocessed operation data;
the diagnosis module 300 is used for predicting the fault of the distributed energy storage power station according to the multi-stage diagnosis data.
In the disclosed embodiment, as shown in fig. 4, the determining module 200 includes
A first determining unit 201, configured to screen out a characteristic value of each battery in the distributed energy storage power station from the operation data, and determine a basic data index value of each battery based on the characteristic value of each battery;
a second determining unit 202, configured to screen out primary diagnostic data of each battery in the decentralized energy storage power station from the data, and determine a primary diagnostic data index value of each battery based on the primary diagnostic data of each battery;
a third determining unit 203 for determining a voltage standard score, coulomb efficiency, IC curve peak value of each battery according to the operation data, and determining a second-level diagnostic data index value of each battery based on the voltage standard score, coulomb efficiency, IC curve peak value;
a fourth determining unit 204 for determining weight coefficients of the secondary diagnosis data index values of the respective batteries, respectively, and determining the tertiary diagnosis data index values of the respective batteries based on the weight coefficients;
wherein the multi-level diagnostic data comprises: basic data index value, primary diagnosis data index value, secondary diagnosis data index value and tertiary diagnosis data index value of each battery.
Further, the calculation formula of the voltage standard score of each battery is as follows:
in the method, in the process of the invention,
for the voltage standard of the ith battery in a distributed energy storage power station, < >>
For the voltage value of the ith battery in the decentralized energy storage power station, is->
For the average value of the voltages of the batteries in the decentralized energy storage power station, < >>
Is the standard deviation of the voltage;
the coulombic efficiency of each cell is calculated as follows:
in the method, in the process of the invention,
coulombic efficiency for the ith cell in a decentralized energy storage power station, < >>
For the discharge capacitance of the ith battery in a decentralized energy storage power station, < >>
Is the charge capacitance of the ith cell in the decentralized energy storage power station.
Wherein, the calculation formula of the weight coefficient of the index value of the secondary diagnosis data of each battery is as follows:
in the method, in the process of the invention,
for the value of the d index of the secondary diagnostic data index values of the i-th battery in the distributed energy storage power station,/->
Standard deviation of the (d) th index of the (i) th battery in the distributed energy storage power station>
For the average value of the d index in the index values of the secondary diagnosis data of the i th battery in the distributed energy storage power station,
the number of the secondary diagnosis data indexes.
The determining the three-level diagnostic data index value of each battery based on the weight coefficient includes:
determining a single factor diagnostic value of each index in the secondary diagnostic data indexes of each battery;
and determining three-level diagnosis data index values of each battery according to the weight coefficient and the single-factor diagnosis value of each index.
Wherein, the calculation formula of the three-level diagnosis data index value of each battery is as follows:
in the method, in the process of the invention,
three-level diagnostic data index value for the ith battery in a distributed energy storage power station,/for the power station>
For the matrix of single factor diagnostic values of the ith battery in the decentralized energy storage power station,/for the battery>
Is a weight coefficient matrix.
In an embodiment of the disclosure, the diagnostic module is specifically configured to:
performing standardization processing on the basic data index value, the first-level diagnosis data index value, the second-level diagnosis data index value and the third-level diagnosis data index value of each battery;
substituting the basic data index value, the first-level diagnosis data index value, the second-level diagnosis data index value and the third-level diagnosis data index value of each battery after the standardized processing into a pre-built comprehensive prediction model to obtain a fault prediction result of the distributed energy storage power station.
In an embodiment of the present disclosure, as shown in fig. 5, the system further includes: and the display module 400 is used for displaying the fault prediction result of the distributed energy storage power station.
In summary, the fault multistage diagnosis system for the distributed energy storage power station provided by the embodiment has higher compatibility, and meanwhile, the fault of the distributed energy storage power station can be accurately diagnosed.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and additional implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.
Although embodiments of the present application have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the application, and that variations, modifications, alternatives, and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the application.