Energy storage management system based on artificial intelligence

Abstract

The invention, which belongs to the technical field of energy storage management, discloses an artificial intelligence-based energy storage management system comprising a data recording module, a calibration analysis module, a strategy analysis module and an energy storage management module. The data recording module is used for carrying out real-time charging and discharging recording on the energy storage battery and generating a charging and discharging recording table; the calibration analysis module is used for checking and analyzing each charging and discharging to obtain a checking result of the corresponding charging and discharging; supplementing the checking result into the charging and discharging record table to obtain a checking record table; the strategy analysis module is used for carrying out optimization analysis on the energy storage management strategy, obtaining a checking record table, calculating an incompatibility rate of the charging and discharging decision reason in a preset unit time period according to the checking record table, and generating an incompatibility curve corresponding to the charging and discharging decision reason; determining a strategy optimization material according to the incongruent curve, and performing optimization adjustment on a corresponding energy storage management strategy according to the strategy optimization material; and the energy storage management module is used for performing energy storage management according to the energy storage management strategy and the monitoring data.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy storage management, and specifically is an energy storage management system based on artificial intelligence. Background Art

[0002] Energy storage systems can effectively balance electricity supply and demand, improve grid stability and reliability, and facilitate the integration and utilization of renewable energy, playing an increasingly important role in the modern energy system. However, current energy storage management systems still face many challenges.

[0003] Traditional energy storage management systems rely primarily on manual settings and fixed control strategies, lacking real-time capabilities and flexibility. This results in low energy storage system operating efficiency and limited economic benefits. For example, in the operation and management of energy storage systems, the flexibility of controlling the number of charge and discharge cycles is crucial. Lack of flexible charge and discharge control may cause batteries to frequently be overcharged or over-discharged, accelerating battery aging and shortening battery life. Frequent charge and discharge cycles will also reduce battery life.

[0004] Based on this, the present invention provides an energy storage management system based on artificial intelligence. Summary of the Invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an energy storage management system based on artificial intelligence.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An artificial intelligence-based energy storage management system, comprising a data recording module, a calibration analysis module, a strategy analysis module, and an energy storage management module; The data recording module is used to record the real-time charge and discharge of the energy storage battery and generate a corresponding charge and discharge record table. The charge and discharge record includes the charge and discharge time, charge and discharge type, and the reason for the charge and discharge decision.

[0007] Furthermore, the generation of the charge and discharge record table includes: Establish a blank charge and discharge record table, identify the charge and discharge records of the energy storage battery in real time, and add the corresponding charge and discharge type and charge and discharge time in the charge and discharge record table according to the charge and discharge records. The charge and discharge type is charge or discharge; According to the charge and discharge record collection, the corresponding charge and discharge decision process data is collected in real time; the feature recognition of the decision process data is performed to obtain the corresponding charge and discharge decision reasons, and the charge and discharge decision reasons are added to the charge and discharge record table.

[0008] The calibration analysis module is used to perform calibration analysis on each charge and discharge in the charge and discharge record table to obtain a calibration result of the corresponding charge and discharge, wherein the calibration result includes a calibration pass and a calibration fail; the calibration result is added to the charge and discharge record table, and the charge and discharge record table with the added calibration result is marked as a calibration record table.

[0009] Furthermore, each charge and discharge in the charge and discharge record table is verified and analyzed, including: Identify the charge and discharge time, charge and discharge type, and charge and discharge decision reasons of the corresponding charge and discharge; define calibration material data, and collect the corresponding charge and discharge calibration material data based on the charge and discharge time and calibration material data definition; Perform simulation analysis based on calibration material data to determine whether energy storage management requirements can be met without corresponding charging and discharging, and obtain corresponding charging and discharging verification results based on the judgment results.

[0010] Furthermore, it is determined whether the energy storage management requirements can be met without corresponding charging and discharging, including: Establish a management judgment model. The expression of the management judgment model is: ; Where: s is the input data, representing the calibration material data. If s meets the energy storage management requirements, it means that the energy storage management requirements can be met without the corresponding charging and discharging. The output data is the management judgment value GB(s), which is 1 or 0. Analyze the corresponding calibration material data through the management judgment model to obtain the management judgment value of the corresponding calibration material data; When the management judgment value is 1, the verification result is verification failure; When the management judgment value is 0, the verification result is verification passed.

[0011] The strategy analysis module is used to optimize and analyze the energy storage management strategy, obtain a verification record table, and identify various charging and discharging decision reasons based on the verification record table; Calculating the non-compliance rate of the charge and discharge decision reason within a preset unit time period according to the calibration record table, and generating a non-compliance curve corresponding to the charge and discharge decision reason according to the non-compliance rate; the horizontal axis of the non-compliance curve represents the independent variable of the corresponding unit time period, and the vertical axis represents the non-compliance rate corresponding to the corresponding unit time period; A strategy optimization material is determined according to the non-conforming curve, and a corresponding energy storage management strategy is optimized and adjusted according to the strategy optimization material.

[0012] Furthermore, the mismatch rate of the corresponding charge and discharge decision reasons within the preset unit period is calculated according to the verification record table, including: Identify the corresponding charge and discharge calibration results and charge and discharge times in the calibration record sheet based on the charge and discharge decision reasons; The verification results are distributed to corresponding unit time periods according to the charge and discharge time; the ratio of the number of verification failures in the unit time period to the total number of verification results is calculated, and the ratio is marked as the failure rate.

[0013] Furthermore, strategies are determined based on the various non-conforming curves to optimize the material, including: Set threshold X1 and threshold X2 according to user needs; Fit the inconsistency curve to obtain the corresponding inconsistency function, which is labeled B(t), where t is the independent variable and represents the corresponding unit time period; A curve evaluation model is established. The expression of the curve evaluation model is: ; Where: kt represents the slope of the discrepancy function in the corresponding unit period; the output data is the curve evaluation value PG[B(t)], and the curve evaluation value is 1, 2 or 0; Analyze the non-conformity function through the curve evaluation model to obtain the curve evaluation value of the non-conformity function in the corresponding unit period; Mark the curve segments whose curve evaluation values are not equal to 0 as optimized segments; Optimize materials based on the optimized segment collection strategy.

[0014] The energy storage management module is used to perform energy storage management according to energy storage management strategies and monitoring data.

[0015] Furthermore, it also includes an abnormality monitoring module, which is used to perform abnormality monitoring.

[0016] Furthermore, the abnormality monitoring method includes: Obtain the calibration record sheet, calculate the calibration failure rate of the corresponding time in real time based on the calibration record sheet, and generate a calibration analysis curve based on the obtained calibration failure rate. The horizontal axis of the calibration analysis curve is time and the vertical axis is the calibration failure rate; issue an abnormal warning based on the calibration analysis curve.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention achieves continuous optimization of energy storage management strategies through the mutual cooperation between various modules. Compared with traditional systems that rely on manual settings and fixed control strategies, the present invention significantly improves the response speed and operating efficiency of the energy storage system; by reducing the number of charge and discharge times, the energy storage system can focus more on charging when electricity prices are low and discharging when electricity prices are peak, or optimize the charge and discharge strategy according to the power generation of renewable energy, thereby improving energy utilization efficiency; frequent charging and discharging will lead to energy loss during the conversion and storage process; reducing the number of charge and discharge times can reduce these losses and improve the overall efficiency of the energy storage system; by increasing the service life of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0020] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] like Figure 1 As shown, an artificial intelligence-based energy storage management system includes a data recording module, a calibration analysis module, a strategy analysis module, an energy storage management module and an abnormality monitoring module; The data recording module is used to record the real-time charge and discharge of the energy storage battery and generate a corresponding charge and discharge record table. The charge and discharge record includes the charge and discharge time, charge and discharge type, and the reason for the charge and discharge decision.

[0022] In one embodiment, generating the charge and discharge record table includes: Establish a blank charge and discharge record table, identify the charge and discharge records of the energy storage battery in real time, and add the corresponding charge and discharge type and charge and discharge time in the charge and discharge record table according to the charge and discharge records. The charge and discharge type is charge or discharge; The corresponding charging and discharging decision-making process data is collected in real time based on the charging and discharging record collection. That is, the decision-making process data of this charging and discharging is carried out according to the energy storage management strategy preset by the energy storage management system. For example, when the SOC is lower than 10%, charging is carried out. The SOC represents the percentage of the current remaining power of the battery to its total capacity. Charging is carried out when the electricity price is lower than a certain price, etc. The decision-making process data is feature-recognized to obtain the corresponding charging and discharging decision reasons, that is, the reason characteristics for why the charging and discharging decision is made are extracted; and the charging and discharging decision reasons are added to the charging and discharging record table.

[0023] In one embodiment, the charge and discharge record table may be intelligently generated based on other existing methods.

[0024] The calibration analysis module is used to perform calibration analysis on each charge and discharge in the charge and discharge record table to obtain the calibration results of the corresponding charge and discharge, and the calibration results include calibration pass and calibration fail; the obtained calibration results are added to the charge and discharge record table, and the charge and discharge record table with the supplemented calibration results is marked as the calibration record table.

[0025] In one embodiment, the verification and analysis of each charge and discharge in the charge and discharge record table includes: Identify the charge and discharge time, charge and discharge type, and charge and discharge decision reasons corresponding to the corresponding charge and discharge, and define calibration material data. The calibration material data is positioned as historical energy storage data for a period of time before and after the corresponding charge and discharge, mainly related data such as battery remaining data before and after charge and discharge, demand satisfaction, etc., that is, data related to verifying whether subsequent normal operation can be carried out, such as whether the power supply demand can be met, whether the SOC exceeds the preset value, etc. The corresponding time is manually preset, such as one hour, half a day, or until the next charge and discharge; the calibration material data of the corresponding charge and discharge is collected according to the charge and discharge time and calibration material data definition; A simulation analysis is performed based on the calibration material data to determine whether the energy storage management requirements can be met without the current charge and discharge. That is, assuming that the current charge and discharge are not performed, the corresponding changes caused by the charge and discharge are deducted to analyze whether the power supply demand can be met and whether the SOC exceeds the preset value. The energy storage management requirements are pre-set and mainly focus on relevant requirements such as power consumption requirements and SOC threshold requirements. The verification result is then obtained, that is, the energy storage management requirements are not met, indicating that the current charge and discharge are necessary and the verification passes.

[0026] In one embodiment, whether the energy storage management requirements can be met without performing the current charge and discharge can be determined and analyzed based on existing methods.

[0027] In one embodiment, determining whether energy storage management requirements can be met without performing corresponding charging and discharging includes: Establish a management judgment model. The expression of the management judgment model is: ; Where: s is the input data, representing the calibration material data. If s meets the energy storage management requirements, it means that the energy storage management requirements can be met without charging and discharging. The training set is set using historical calibration material data for training, and judgment is made based on the calibration material data. The output data is the management judgment value GB(s), which is 1 or 0. Analyze the corresponding calibration material data through the management judgment model to obtain the management judgment value of the corresponding calibration material data; When the management judgment value is 1, the verification result is verification failure; When the management judgment value is 0, the verification result is verification passed.

[0028] The strategy analysis module is used to optimize and analyze the energy storage management strategy, obtain a calibration record table, identify various charge and discharge decision reasons based on the calibration record table, calculate the non-compliance rate of the corresponding charge and discharge decision reason within a preset unit time period based on the calibration record table, and generate a non-compliance curve corresponding to the charge and discharge decision reason based on the non-compliance rate. The horizontal axis of the non-compliance curve represents the independent variable of the corresponding unit time period, such as t=1 represents the first unit time period, t=2 represents the second unit time period, and so on. The unit time period is subsequently determined based on time; the vertical axis represents the non-compliance rate corresponding to the corresponding unit time period. The unit time period is a preset time period, such as one day, three days, or one week. Determine strategy optimization materials based on each non-conforming curve, and optimize and adjust the corresponding energy storage management strategy based on the strategy optimization materials.

[0029] In one embodiment, calculating the mismatch rate of the corresponding charge and discharge decision reasons within a preset unit period according to the verification record table includes: According to the charge and discharge decision reasons, the corresponding charge and discharge calibration results and charge and discharge times are identified in the calibration record table, that is, the charge and discharge calibration results corresponding to the charge and discharge decision reasons; the obtained calibration results are allocated to the corresponding unit time periods according to the charge and discharge times; the ratio of calibration failures within the unit time period is calculated, and the ratio is marked as the non-compliance rate.

[0030] In one embodiment, determining a strategy to optimize the material based on each non-conforming curve includes: Threshold X1 and threshold X2 are set according to user requirements. Threshold X1 is the allowable critical value of the non-conformity rate. If the threshold X1 is exceeded, it is considered to not meet the requirements, such as 10%, 8%, etc. Threshold X2 is the allowable slope change. If the slope of the curve exceeds the threshold X2, it is considered to deviate from the normal non-conformity rate. The specific setting of threshold X1 and threshold X2 by those skilled in the art can be determined based on actual conditions or by simulating a large amount of data until the simulation is adjusted to meet the requirements. Fit the non-conformity curve to obtain the corresponding non-conformity function, and mark the non-conformity function as B(t), where t is the independent variable, i.e., the independent variable of the non-conformity curve, and represents the corresponding unit time period; A curve evaluation model is established. The expression of the curve evaluation model is: ; Where: kt represents the slope of the discrepancy function in the corresponding unit period; the output data is the curve evaluation value PG[B(t)], and the curve evaluation value is 1, 2 or 0; Analyze the non-conformity function through the curve evaluation model to obtain the curve evaluation value of the non-conformity function in the corresponding unit period; Mark the curve segments whose curve evaluation values are not equal to 0 as optimized segments; According to the optimization stage, we collect strategy optimization materials. This means collecting relevant data such as the monitoring data that generated the charge and discharge decision, the reasons for the charge and discharge decision, and the process data of the above simulations that determined the verification failed. These data are then integrated into strategy optimization materials. This facilitates the subsequent determination of the reasons and background for the decision failure based on the strategy optimization materials, and then makes targeted strategy adjustments.

[0031] In one embodiment, the optimization segment can also be determined based on other existing technologies, such as establishing an analysis model based on machine learning, deep learning algorithms, etc., and manually establishing a corresponding training set for training. The training set includes input data and output data, the input data is the non-conforming curve, and the output data is the optimization segment.

[0032] In one embodiment, the corresponding energy storage management strategy is optimized and adjusted according to the strategy optimization material. The adjustment can be made according to the strategy optimization material in an existing manner, such as manually, by the user or the platform. Alternatively, an intelligent strategy adjustment model can be established based on intelligent algorithms such as CNN networks or DNN networks, and the platform can establish a corresponding training set for training, and then subsequently indicate the data and format required for the strategy optimization material. After successful training, the energy storage management strategy is dynamically adjusted by the intelligent strategy adjustment model.

[0033] The energy storage management module is used to perform energy storage management according to energy storage management strategies and monitoring data.

[0034] The abnormality monitoring module is used to perform abnormality monitoring, obtain a calibration record sheet, calculate the calibration failure rate of the corresponding time in real time based on the calibration record sheet, generate a calibration analysis curve based on the obtained calibration failure rate, the horizontal axis of the calibration analysis curve is time, and the vertical axis is the calibration failure rate; and perform abnormality warning based on the calibration analysis curve.

[0035] In one embodiment, an abnormality warning is performed based on the calibration analysis curve. The calibration failure rate can be compared with a preset value, and an abnormality warning is performed when the preset value is reached. An abnormality warning can also be performed based on the cumulative deviation within a preset time period. Similarly, an abnormality warning model can be established based on deep learning algorithms for analysis.

[0036] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0037] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An energy storage management system based on artificial intelligence, characterized in that: Includes data recording module, calibration analysis module, strategy analysis module and energy storage management module; The data recording module is used to record the real-time charge and discharge of the energy storage battery and generate a corresponding charge and discharge record table. The charge and discharge record includes the charge and discharge time, charge and discharge type, and the reason for the charge and discharge decision; The calibration analysis module is used to perform calibration analysis on each charge and discharge in the charge and discharge record table to obtain a calibration result of the corresponding charge and discharge, wherein the calibration result includes a pass or fail calibration; the calibration result is added to the charge and discharge record table, and the charge and discharge record table with the added calibration result is marked as a calibration record table; The strategy analysis module is used to optimize and analyze the energy storage management strategy, obtain a verification record table, and identify various charging and discharging decision reasons based on the verification record table; Calculating the non-compliance rate of the charge and discharge decision reason within a preset unit time period according to the calibration record table, and generating a non-compliance curve corresponding to the charge and discharge decision reason according to the non-compliance rate; the horizontal axis of the non-compliance curve represents the independent variable of the corresponding unit time period, and the vertical axis represents the non-compliance rate corresponding to the corresponding unit time period; Determining strategy optimization materials based on the non-conforming curves, and optimizing and adjusting the corresponding energy storage management strategy based on the strategy optimization materials; The energy storage management module is used to perform energy storage management according to energy storage management strategies and monitoring data.

2. The artificial intelligence-based energy storage management system according to claim 1, characterized in that: Generation of charge and discharge record table, including: Establish a blank charge and discharge record table, identify the charge and discharge records of the energy storage battery in real time, and add the corresponding charge and discharge type and charge and discharge time in the charge and discharge record table according to the charge and discharge records. The charge and discharge type is charge or discharge; According to the charge and discharge record collection, the corresponding charge and discharge decision process data is collected in real time; the feature recognition of the decision process data is performed to obtain the corresponding charge and discharge decision reasons, and the charge and discharge decision reasons are added to the charge and discharge record table.

3. The artificial intelligence-based energy storage management system according to claim 1, characterized in that: Verify and analyze each charge and discharge in the charge and discharge record table, including: Identify the charge and discharge time, charge and discharge type, and charge and discharge decision reasons of the corresponding charge and discharge; define calibration material data, and collect the corresponding charge and discharge calibration material data based on the charge and discharge time and calibration material data definition; Perform simulation analysis based on calibration material data to determine whether energy storage management requirements can be met without corresponding charging and discharging, and obtain corresponding charging and discharging verification results based on the judgment results.

4. The artificial intelligence-based energy storage management system according to claim 3, characterized in that: Determine whether energy storage management requirements can be met without corresponding charging and discharging, including: Establish a management judgment model. The expression of the management judgment model is: ; Where: s is the input data, representing the calibration material data. If s meets the energy storage management requirements, it means that the energy storage management requirements can be met without the corresponding charging and discharging. The output data is the management judgment value GB(s), which is 1 or 0. Analyze the corresponding calibration material data through the management judgment model to obtain the management judgment value of the corresponding calibration material data; When the management judgment value is 1, the verification result is verification failure; When the management judgment value is 0, the verification result is verification passed.

5. The artificial intelligence-based energy storage management system according to claim 1, characterized in that: Calculate the mismatch rate of the corresponding charge and discharge decision reasons within the preset unit period according to the calibration record table, including: Identify the corresponding charge and discharge calibration results and charge and discharge times in the calibration record sheet based on the charge and discharge decision reasons; The verification results are distributed to corresponding unit time periods according to the charge and discharge time; the ratio of the number of verification failures in the unit time period to the total number of verification results is calculated, and the ratio is marked as the failure rate.

6. The artificial intelligence-based energy storage management system according to claim 1, characterized in that: Determine strategies to optimize materials based on different curves, including: Set threshold X1 and threshold X2 according to user needs; Fit the inconsistency curve to obtain the corresponding inconsistency function, which is labeled B(t), where t is the independent variable and represents the corresponding unit time period; A curve evaluation model is established. The expression of the curve evaluation model is: ; Where: kt represents the slope of the discrepancy function in the corresponding unit period; the output data is the curve evaluation value PG[B(t)], and the curve evaluation value is 1, 2 or 0; Analyze the non-conformity function through the curve evaluation model to obtain the curve evaluation value of the non-conformity function in the corresponding unit period; Mark the curve segments whose curve evaluation values are not equal to 0 as optimized segments; Optimize materials based on the optimized segment collection strategy.

7. The artificial intelligence-based energy storage management system according to claim 1, characterized in that: It also includes an abnormality monitoring module, which is used to perform abnormality monitoring.

8. The artificial intelligence-based energy storage management system according to claim 7, characterized in that: Methods for anomaly monitoring include: Obtain the calibration record sheet, calculate the calibration failure rate of the corresponding time in real time based on the calibration record sheet, and generate a calibration analysis curve based on the obtained calibration failure rate. The horizontal axis of the calibration analysis curve is time and the vertical axis is the calibration failure rate; issue an abnormal warning based on the calibration analysis curve.

Images