A distributed energy storage system charging and discharging intelligent control method and cloud platform

Abstract

The present invention relates to the field of power control technology and specifically discloses a method and cloud platform for intelligent charge and discharge control of a distributed energy storage system. The method includes: real-time monitoring of the distributed energy storage system's energy storage status data, operating status data, and power data; obtaining historical abnormal control data of the distributed energy storage system, performing abnormality detection on the distributed energy storage system, determining real-time abnormal data, real-time normal data, and abnormal control instructions; determining energy storage constraints, coordination constraints, and an objective function for maximizing arbitrage returns, and generating optimal charge and discharge control instructions; executing the abnormal control instructions and the optimal charge and discharge control instructions to achieve intelligent charge and discharge control of the distributed energy storage system. This can improve the instruction matching rate and the efficiency of abnormality detection and processing, reduce abnormality response time, optimize charge and discharge strategies, reduce electricity costs, balance economy and safety, and improve system operation efficiency and reliability.

Description

Technical Field

[0001] The present invention relates to the field of power control technology, and in particular to a charging and discharging intelligent control method and a cloud platform for a distributed energy storage system. Background Art

[0002] Early distributed energy storage control relied on single threshold judgments. For example, overcharge protection only set a voltage upper limit, lacking historical data support and resulting in a high rate of misjudgment. After 2010, with the application of lithium iron phosphate batteries in distributed scenarios (such as streetlights and tunnels), the scale of energy storage systems expanded, and traditional control methods struggled to cope with the needs of multi-storage unit coordination and grid interaction. Since 2015, intelligent optimization algorithms (such as genetic algorithms) have begun to be applied to charge and discharge strategy optimization, but they do not integrate equipment life and historical abnormality data. After 2020, with the development of edge computing and 5G communications, distributed energy storage has gradually achieved "cloud-edge-end" collaborative control, but anomaly detection still relies on real-time data and has not formed precise control driven by historical case libraries.

[0003] Therefore, the present invention proposes a charging and discharging intelligent control method and cloud platform for a distributed energy storage system. Summary of the Invention

[0004] The present invention provides a method and cloud platform for intelligent charge and discharge control of a distributed energy storage system. By analyzing detected energy storage status data, operating status data, and acquired historical abnormal control data, the method determines real-time abnormal and normal data, determines and executes abnormal control instructions, analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and an objective function for maximizing arbitrage returns, generates and executes optimal charge and discharge control instructions, and implements intelligent charge and discharge control of the distributed energy storage system. This method improves instruction matching rates and abnormality detection and processing efficiency, reduces abnormality response time, optimizes charge and discharge strategies, reduces electricity costs, balances economy and safety, and improves system operational efficiency and reliability.

[0005] The present invention provides a method for intelligently controlling charging and discharging of a distributed energy storage system, comprising:

[0006] S1: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, and real-time monitoring of the power data in the target area of ​​the distributed storage system;

[0007] S2: Obtain historical abnormal control data of the distributed energy storage system, perform abnormality detection on the distributed energy storage system based on the energy storage status data, operating status data, and historical abnormal control data, determine real-time abnormal data and real-time normal data, and determine abnormal control instructions based on the real-time abnormal data and historical abnormal control data;

[0008] S3: Analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, and generates optimal charge and discharge control instructions;

[0009] S4: Execute abnormal control instructions and optimal charge and discharge control instructions to realize intelligent charge and discharge control of distributed energy storage systems.

[0010] Preferably, a method for intelligent control of charging and discharging of a distributed energy storage system, wherein the energy storage cabinet includes at least a battery unit, an intelligent air conditioner, a fire extinguishing device, and a first monitoring group, wherein the first monitoring group includes at least a voltage sensor, a current sensor, a temperature sensor, a smoke alarm, and a water intrusion alarm;

[0011] The power data at least includes power demand data of a target area of ​​the distributed storage system and a time-of-use electricity price policy.

[0012] Preferably, a method for intelligent control of charging and discharging of a distributed energy storage system monitors the energy storage status data of all energy storage cabinets in a target area of ​​the distributed energy storage system in real time, and monitors the operating status data of the distributed energy storage system in real time, including:

[0013] Real-time monitoring of the energy storage status sub-data of each energy storage cabinet in the target area of ​​the distributed energy storage system, and determining the energy storage status data of the distributed energy storage system based on the energy storage status sub-data of all energy storage cabinets, wherein the energy storage status sub-data includes at least the current remaining power, full charge capacity, health status, charging efficiency, discharging efficiency, maximum charging power, maximum discharging power, minimum state of charge, maximum state of charge, maximum number of charge and discharge switching times, and unit cycle cost;

[0014] The operating status sub-data of each component of the distributed energy storage system is monitored in real time, and the operating status data of the distributed energy storage system is determined based on the operating status sub-data of all components.

[0015] Preferably, a method for intelligent control of charging and discharging of a distributed energy storage system obtains historical abnormal control data of the distributed energy storage system, performs abnormality detection on the distributed energy storage system based on energy storage status data, operating status data, and historical abnormal control data, and determines real-time abnormal data and real-time normal data, including:

[0016] Obtaining historical abnormality control sub-data of the distributed energy storage system within multiple specified operating cycles, and determining historical abnormality control data based on the historical abnormality control sub-data within all specified operating cycles, wherein the historical abnormality control sub-data includes multiple component abnormalities, multiple energy storage cabinet abnormalities, a component abnormality vector for each component abnormality, a component abnormality control instruction, and an energy storage cabinet abnormality vector for each energy storage cabinet abnormality, and an energy storage cabinet abnormality control instruction;

[0017] performing a first cluster analysis on component anomaly vectors of all component anomalies of the same component in all historical anomaly control sub-data in the historical anomaly control data, and determining a plurality of component anomaly categories for each component of the distributed energy storage system based on the first cluster analysis result, wherein each component anomaly category includes a plurality of component anomaly vectors;

[0018] determining a component category vector for each component anomaly category of each component based on all component anomaly vectors for each component anomaly category of each component of the distributed energy storage system;

[0019] Extracting features from the running status sub-data of each component in the running status data to determine a component feature vector of each component;

[0020] performing a second cluster analysis on energy storage cabinet abnormality vectors of all energy storage cabinet abnormalities in all historical abnormality control sub-data in the historical abnormality control data, and determining a plurality of energy storage cabinet abnormality categories of the energy storage cabinets of the distributed energy storage system based on the second cluster analysis results, wherein each energy storage cabinet abnormality category includes a plurality of energy storage cabinet abnormality vectors;

[0021] determining an energy storage cabinet category vector of each energy storage cabinet abnormality category based on all energy storage cabinet abnormality vectors of each energy storage cabinet abnormality category of the energy storage cabinets of the distributed energy storage system;

[0022] Performing feature extraction on the energy storage status sub-data of each energy storage cabinet in the energy storage status data to determine an energy storage cabinet feature vector for each energy storage cabinet;

[0023] Based on the preset component feature matrix and component feature vector of each component, identify whether each component has an abnormality; based on the preset energy storage cabinet feature matrix and energy storage cabinet feature vector of each energy storage cabinet, identify whether each energy storage cabinet has an abnormality;

[0024] Real-time abnormal data is determined based on the operating status sub-data and component feature vectors of all components with abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets with abnormalities. Real-time normal data is determined based on the operating status sub-data and component feature vectors of all components without abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets without abnormalities.

[0025] Preferably, a method for intelligent control of charging and discharging of a distributed energy storage system determines abnormal control instructions based on real-time abnormal data and historical abnormal control data, including:

[0026] For each abnormal component in the real-time abnormal data, the component abnormality category corresponding to the component category vector with the greatest similarity to the component feature vector of each component is selected as the real-time component abnormality of each component;

[0027] Selecting, from all component exception vectors in the component exception category corresponding to the real-time component exception of each component, a component exception control instruction corresponding to the component exception vector having the greatest similarity to the component feature vector of each component as the first real-time exception control sub-instruction of each component;

[0028] For each energy storage cabinet with an abnormality in the real-time abnormal data, select the energy storage cabinet abnormality category corresponding to the energy storage cabinet category vector with the greatest similarity to the energy storage cabinet feature vector of each energy storage cabinet as the real-time energy storage cabinet abnormality of each energy storage cabinet;

[0029] From all energy storage cabinet abnormality vectors in the energy storage cabinet abnormality category corresponding to the real-time energy storage cabinet abnormality of each energy storage cabinet, select an energy storage cabinet abnormality control instruction corresponding to the energy storage cabinet abnormality vector having the greatest similarity to the energy storage cabinet feature vector of each energy storage cabinet as the second real-time abnormality control sub-instruction of each energy storage cabinet;

[0030] Based on the first real-time abnormality control sub-instructions of all abnormal components in the real-time abnormality data and the second real-time abnormality control sub-instructions of all abnormal energy storage cabinets, an abnormality control instruction of the distributed energy storage system is determined.

[0031] Preferably, a method for intelligent control of charging and discharging of a distributed energy storage system analyzes real-time normal data and power data to determine energy storage constraints, coordination constraints, and an arbitrage profit maximization objective function, including:

[0032] Based on the time-of-use electricity price policy in the power data, the specified operating cycle is divided into multiple time periods;

[0033] Determine the load demand for each period based on the power demand data and all periods after the designated operating cycle is divided;

[0034] Based on the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data and all time periods after the specified operation cycle is divided, determining the energy storage constraint conditions of each energy storage cabinet without abnormalities in the real-time normal data, where the energy storage constraint conditions include charge and discharge power constraints, state of charge constraints, and energy storage cabinet life constraints;

[0035] Determining coordination constraints based on the load demand for each time period and the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data, where the coordination constraints include grid interaction constraints and power balance constraints;

[0036] Based on power demand data, time-of-use electricity price policy, load demand in all time periods, and the energy storage status sub-data of each energy storage cabinet without abnormalities in real-time normal data, the arbitrage profit maximization objective function is determined.

[0037] Preferably, a method for intelligently controlling charging and discharging of a distributed energy storage system generates optimal charging and discharging control instructions, comprising:

[0038] A control model is constructed based on the charging and discharging power constraints, state of charge constraints, and energy storage cabinet life constraints in the energy storage constraints, as well as the grid interaction constraints, power balance constraints, and arbitrage profit maximization objective function in the coordination constraints.

[0039] The real-time normal data and power data are input into the control model, and the control model generates an optimal charge and discharge control instruction, wherein the optimal charge and discharge control instruction includes an optimal charge and discharge control sub-instruction for each energy storage cabinet in the real-time normal data.

[0040] The present invention provides a distributed energy storage system charging and discharging intelligent control cloud platform, which is used to execute the distributed energy storage system charging and discharging intelligent control method described in any one of Embodiments 1 to 7, including:

[0041] Monitoring module: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, and real-time monitoring of the power data of the target area of ​​the distributed storage system;

[0042] Abnormal control module: obtains historical abnormal control data of the distributed energy storage system, performs abnormality detection on the distributed energy storage system based on energy storage status data, operating status data and historical abnormal control data, determines real-time abnormal data and real-time normal data, and determines abnormal control instructions based on the real-time abnormal data and historical abnormal control data;

[0043] Charge and discharge control module: Analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, and generates optimal charge and discharge control instructions;

[0044] Execution module: executes abnormal control instructions and optimal charge and discharge control instructions to realize intelligent charge and discharge control of distributed energy storage systems.

[0045] Compared to existing technologies, the present invention achieves the following beneficial effects: by analyzing detected energy storage status data, operating status data, and acquired historical abnormal control data, determining real-time abnormal data and real-time normal data, determining abnormal control instructions and executing them, analyzing real-time normal data and power data, determining energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, generating and executing optimal charge and discharge control instructions, and thus realizing intelligent charge and discharge control of distributed energy storage systems. This improves the instruction matching rate and the efficiency of abnormality detection and processing, reduces abnormality response time, optimizes charge and discharge strategies, reduces electricity costs, balances economy and safety, and improves system operational efficiency and reliability.

[0046] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0047] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0049] Figure 1 This is a flow chart of a method for intelligent control of charging and discharging of a distributed energy storage system in an embodiment of the present invention.

[0050] Figure 2 Schematic diagram of a charging and discharging intelligent control cloud platform for a distributed energy storage system in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention. Example 1:

[0052] The present invention provides a method for intelligent control of charge and discharge of distributed energy storage system. Figure 1 ,include:

[0053] S1: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, and real-time monitoring of the power data in the target area of ​​the distributed storage system;

[0054] S2: Obtain historical abnormal control data of the distributed energy storage system, perform abnormality detection on the distributed energy storage system based on the energy storage status data, operating status data, and historical abnormal control data, determine real-time abnormal data and real-time normal data, and determine abnormal control instructions based on the real-time abnormal data and historical abnormal control data;

[0055] S3: Analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, and generates optimal charge and discharge control instructions;

[0056] S4: Execute abnormal control instructions and optimal charge and discharge control instructions to realize intelligent charge and discharge control of distributed energy storage systems.

[0057] In this embodiment, historical abnormal control sub-data within multiple specified operating cycles are extracted from the monitoring system and logs of the distributed energy storage system, including component abnormalities, energy storage cabinet abnormalities, abnormal vectors, and abnormal control instructions.

[0058] In this embodiment, real-time normal data and power data are input into the control model, and an optimization algorithm (such as linear programming, dynamic programming, or genetic algorithm) is used to solve the optimal charge and discharge control instructions. The generated optimal charge and discharge control instructions include the optimal charge and discharge control sub-instructions for each energy storage cabinet.

[0059] In this embodiment, executing abnormal control instructions may include executing safety protection instructions, such as disconnecting a faulty battery pack, starting a cooling system, and sounding an alarm; executing fault isolation instructions to ensure normal operation of other parts of the system, etc.

[0060] In this embodiment, optimal charge and discharge control instructions are sent to the energy storage system to control the charging and discharging operations of the energy storage cabinet. The energy storage system's operating status is monitored in real time to ensure the effectiveness of the control instructions. Based on the real-time monitoring data, the effectiveness of the control instructions is evaluated. Based on the evaluation results, the control instructions are adjusted to optimize system operation.

[0061] In this embodiment, the arbitrage profit maximization objective function is a mathematical expression used to describe how a distributed energy storage system maximizes arbitrage profits by rationally arbitrage-discharging strategies under certain conditions. It comprehensively considers multiple factors, including the price mechanism of the electricity market, load demand, and the state of the energy storage system itself.

[0062] The beneficial effects of the above technology include: analyzing detected energy storage status data, operating status data, and historical abnormal control data to determine real-time abnormal data and real-time normal data, determining abnormal control instructions and executing them, analyzing real-time normal data and power data to determine energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, generating and executing optimal charge and discharge control instructions, and realizing intelligent charge and discharge control of distributed energy storage systems. This can improve the instruction matching rate and the efficiency of abnormality detection and processing, reduce abnormality response time, optimize charge and discharge strategies, reduce electricity costs, balance economy and safety, and improve system operation efficiency and reliability. Example 2:

[0063] Based on Example 1, a method for intelligent control of charging and discharging of a distributed energy storage system is provided, wherein the energy storage cabinet includes at least a battery unit, an intelligent air conditioner, a fire extinguishing device, and a first monitoring group, wherein the first monitoring group includes at least a voltage sensor, a current sensor, a temperature sensor, a smoke alarm, and a water intrusion alarm;

[0064] The power data at least includes power demand data of a target area of ​​the distributed storage system and a time-of-use electricity price policy.

[0065] In this embodiment, the fire extinguishing device may be an aerosol fire extinguishing system.

[0066] In this embodiment, the intelligent air conditioner uses air cooling technology to adjust the temperature inside the energy storage cabinet.

[0067] In this embodiment, a smoke alarm is provided on the top of the energy storage cabinet. This device is one of the most advanced alarm devices. Once triggered, it will immediately stop all states of the energy storage equipment and notify the on-duty manager via a short message on his mobile phone.

[0068] In this embodiment, a water intrusion alarm is provided at the bottom of the energy storage cabinet. This device is one of the most advanced alarm devices. Once triggered, it will immediately stop all states of the energy storage device and notify the on-duty manager via a short message on his mobile phone.

[0069] In this embodiment, the local power grid's time-of-use electricity price data (e.g., 1.2 yuan / kWh for peak hours (8:00 AM - 10:00 PM) and 0.4 yuan / kWh for off-peak hours (22:00 PM - 8:00 AM)) is read from the meter via the 485 interface and stored in the BAU memory. 485 stands for RS-485, a serial communication interface standard that uses differential signaling to effectively suppress common-mode interference and maintain high data transmission rates (up to 10 Mbps, with longer distances resulting in lower rates) over longer distances (up to 1200 meters). It supports multi-node communication, allowing up to 32 devices to connect on a single bus. Address encoding distinguishes devices, allowing multiple devices to exchange data on the same bus.

[0070] The beneficial effects of the above technology are: determining power data and energy storage cabinets, monitoring energy storage status data, and providing data support for determining constraints and objective functions. Example 3:

[0071] Based on Example 2, a method for intelligently controlling charging and discharging of a distributed energy storage system is provided, which monitors the energy storage status data of all energy storage cabinets in a target area of ​​the distributed energy storage system in real time, and monitors the operating status data of the distributed energy storage system in real time, including:

[0072] Real-time monitoring of the energy storage status sub-data of each energy storage cabinet in the target area of ​​the distributed energy storage system, and determining the energy storage status data of the distributed energy storage system based on the energy storage status sub-data of all energy storage cabinets, wherein the energy storage status sub-data includes at least the current remaining power, full charge capacity, health status, charging efficiency, discharging efficiency, maximum charging power, maximum discharging power, minimum state of charge, maximum state of charge, maximum number of charge and discharge switching times, and unit cycle cost;

[0073] The operating status sub-data of each component of the distributed energy storage system is monitored in real time, and the operating status data of the distributed energy storage system is determined based on the operating status sub-data of all components.

[0074] In this embodiment, the current remaining power is monitored in real time by a fuel meter or coulomb meter in a battery management system (BMS) for the current remaining power of the battery cells in each energy storage cabinet.

[0075] In this embodiment, the full-charge capacity is recorded by the BMS through the nominal capacity of the battery unit, and the estimated value of the full-charge capacity is dynamically adjusted based on actual usage.

[0076] In this embodiment, the health status of the battery is evaluated by monitoring parameters such as the internal resistance, cycle number, and capacity decay of the battery.

[0077] In this embodiment, the charging efficiency is calculated by monitoring the ratio of the input power and the actual stored power during the charging process.

[0078] In this embodiment, the maximum charging power is obtained through communication between the BMS and the power storage converter (PCS) to obtain the maximum charging power limit of the battery unit.

[0079] In this embodiment, the maximum discharge power obtains the maximum discharge power limit of the battery unit through communication between the BMS and the PCS.

[0080] In this embodiment, the minimum state of charge sets the lowest state of charge allowed for the battery cell to avoid over-discharge.

[0081] In this embodiment, the maximum state of charge sets the highest state of charge allowed for the battery cell to avoid overcharging.

[0082] In this embodiment, the maximum charge-discharge switching times records the number of times the charge-discharge state of the battery unit is switched within a certain period of time, and is used to evaluate the usage frequency and life of the battery.

[0083] In this embodiment, the unit cycle cost is evaluated by calculating the cost of each charge and discharge cycle and combining the capacity and service life of the battery.

[0084] In this embodiment, the components of the distributed energy storage system may include an energy storage converter, a battery management unit (BMU), a main control unit (BCU), an overall control unit (BAU), an energy management system (EMS), and switchgear.

[0085] In this embodiment, the energy storage converter is used to realize the conversion between direct current and alternating current, and is an interface device between the energy storage cabinet and the power grid or load.

[0086] In this embodiment, the battery unit in each energy storage cabinet of the branch control unit (BMU) is integrated with one, collects the voltage and temperature of the single battery and uploads them to the BCU, and receives the fan start and stop instructions issued by the BCU.

[0087] In this embodiment, the main control unit (BCU) collects the total voltage on the battery side, the total voltage on the PCS side, the charge and discharge current, estimates the SOC, etc.; controls the closing or opening of the main positive, main negative, and pre-charge contactors; receives battery data from the BMU via CAN; exchanges information with the BAU via CAN; and performs battery protection logic judgment.

[0088] In this embodiment, the master control unit (BAU) collects the operating position status of each switch and rotary switch; exchanges information with the BCU via CAN; communicates with the PCS via CAN to upload battery information; communicates with the PCS via 485 to control the PCS operating mode and control PCS operation using time policies; communicates with the LCD via 485 to upload data and receive LCD setting information; communicates with the meter via 485 to read meter data; and exchanges data with the data terminal via 232, simultaneously receiving backend timing information. 232, also known as RS-232, is one of the earliest serial communication standards and uses single-ended signaling. Its transmission distance is relatively short, generally within 15 meters, and its data transmission rate can reach up to 20 kbps. RS-232 is typically used for point-to-point communication, meaning one transmitter corresponds to one receiver.

[0089] The beneficial effects of the above technology are: real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, which can realize the refined monitoring and precise control of the entire energy storage system, and improve data support for determining real-time abnormal data and real-time normal data. Example 4:

[0090] Based on Example 3, a method for intelligent charge and discharge control of a distributed energy storage system is provided. The method obtains historical abnormal control data of the distributed energy storage system, performs abnormality detection on the distributed energy storage system based on energy storage status data, operating status data, and historical abnormal control data, and determines real-time abnormal data and real-time normal data. The method includes:

[0091] Obtaining historical abnormality control sub-data of the distributed energy storage system within multiple specified operating cycles, and determining historical abnormality control data based on the historical abnormality control sub-data within all specified operating cycles, wherein the historical abnormality control sub-data includes multiple component abnormalities, multiple energy storage cabinet abnormalities, a component abnormality vector for each component abnormality, a component abnormality control instruction, and an energy storage cabinet abnormality vector for each energy storage cabinet abnormality, and an energy storage cabinet abnormality control instruction;

[0092] performing a first cluster analysis on component anomaly vectors of all component anomalies of the same component in all historical anomaly control sub-data in the historical anomaly control data, and determining a plurality of component anomaly categories for each component of the distributed energy storage system based on the first cluster analysis result, wherein each component anomaly category includes a plurality of component anomaly vectors;

[0093] determining a component category vector for each component anomaly category of each component based on all component anomaly vectors for each component anomaly category of each component of the distributed energy storage system;

[0094] Extracting features from the running status sub-data of each component in the running status data to determine a component feature vector of each component;

[0095] performing a second cluster analysis on energy storage cabinet abnormality vectors of all energy storage cabinet abnormalities in all historical abnormality control sub-data in the historical abnormality control data, and determining a plurality of energy storage cabinet abnormality categories of the energy storage cabinets of the distributed energy storage system based on the second cluster analysis results, wherein each energy storage cabinet abnormality category includes a plurality of energy storage cabinet abnormality vectors;

[0096] determining an energy storage cabinet category vector of each energy storage cabinet abnormality category based on all energy storage cabinet abnormality vectors of each energy storage cabinet abnormality category of the energy storage cabinets of the distributed energy storage system;

[0097] Performing feature extraction on the energy storage status sub-data of each energy storage cabinet in the energy storage status data to determine an energy storage cabinet feature vector for each energy storage cabinet;

[0098] Based on the preset component feature matrix and component feature vector of each component, identify whether each component has an abnormality; based on the preset energy storage cabinet feature matrix and energy storage cabinet feature vector of each energy storage cabinet, identify whether each energy storage cabinet has an abnormality;

[0099] Real-time abnormal data is determined based on the operating status sub-data and component feature vectors of all components with abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets with abnormalities. Real-time normal data is determined based on the operating status sub-data and component feature vectors of all components without abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets without abnormalities.

[0100] In this embodiment, historical abnormality control sub-data within multiple specified operating cycles are extracted from the monitoring system and logs of the distributed energy storage system. These data include: multiple components with abnormalities: recording the abnormal conditions of each component with abnormalities in historical operation; multiple energy storage cabinets with abnormalities: recording the abnormal conditions of each energy storage cabinet with abnormalities in historical operation; component abnormality vectors: quantizing the abnormal characteristics of each component with abnormalities into vector form, including all operating characteristics of the component when the abnormality occurred; component abnormality control instructions: recording the control instructions taken in response to component abnormalities, such as restart and isolation; energy storage cabinet abnormality vectors: quantizing the abnormal characteristics of each energy storage cabinet with abnormalities into vector form, including all state characteristics of the energy storage cabinet when the abnormality occurred; energy storage cabinet abnormality control instructions: recording the control instructions taken in response to energy storage cabinet abnormalities, such as cutting off power and activating fire extinguishing devices.

[0101] In this embodiment, the historical abnormality control sub-data within all specified operating cycles are aggregated to form a complete historical abnormality control data set.

[0102] In this embodiment, a cluster analysis is performed on the component anomaly vectors of all component anomalies of the same component in all historical anomaly control sub-data in the historical anomaly control data, and a clustering algorithm (such as K-means, DBSCAN, etc.) is used to classify similar component anomaly vectors into one category to form multiple component anomaly categories, each of which contains multiple similar component anomaly vectors.

[0103] In this embodiment, statistical analysis is performed on all component anomaly vectors of each component anomaly category of each component, and a central vector of each category is calculated (for example, an average vector is calculated) to form a component category vector, which is used to characterize the characteristics of each component anomaly category.

[0104] In this embodiment, cluster analysis is performed on the energy storage cabinet anomaly vectors of all energy storage cabinet anomalies in all historical anomaly control sub-data in the historical anomaly control data. A clustering algorithm is used to group similar energy storage cabinet anomaly vectors into one category, forming multiple energy storage cabinet anomaly categories, each of which contains multiple similar energy storage cabinet anomaly vectors.

[0105] In this embodiment, statistical analysis is performed on all energy storage cabinet abnormality vectors of each energy storage cabinet abnormality category, and the central vector of each category is calculated to form an energy storage cabinet category vector. The energy storage cabinet category vector is used to characterize the characteristics of each energy storage cabinet abnormality category.

[0106] In this embodiment, feature extraction is performed on the operating status sub-data of each component in the operating status data. The extracted features include temperature, pressure, power conversion efficiency, etc. The extracted features are quantized into vector form to form a component feature vector for each component.

[0107] In this embodiment, feature extraction is performed on the energy storage status sub-data of each energy storage cabinet in the energy storage status data. The extracted features include the current remaining power, full charge capacity, health status, etc. The extracted features are quantized into vector form to form an energy storage cabinet feature vector for each energy storage cabinet.

[0108] In this embodiment, the component preset feature matrix of each component is determined based on a statistical analysis of all component normal vectors of the component under normal operation of the distributed energy storage system. The columns of the component preset feature matrix represent the range of each component feature of the component, the first row of the component preset feature matrix represents the lower limit of the values ​​of all component features of the component, and the second row of the component preset feature matrix represents the upper limit of the values ​​of all component features of the component.

[0109] In this embodiment, it is identified whether each component has an abnormality, that is, it is determined whether each component feature in the component feature vector of each component is within the range of the component feature of the component. If all are within the range, it is determined that the component has no abnormality; otherwise, it is determined that the component has an abnormality.

[0110] In this embodiment, the energy storage cabinet preset feature matrix is ​​determined by statistically analyzing the normal vectors of all energy storage cabinets in the normally operating distributed energy storage system. The columns of the energy storage cabinet preset feature matrix represent the range of each energy storage cabinet feature of the energy storage cabinet. The first row of the energy storage cabinet preset feature matrix represents the lower limit of the values ​​of all energy storage cabinet features of the energy storage cabinet, and the second row of the energy storage cabinet preset feature matrix represents the upper limit of the values ​​of all energy storage cabinet features of the energy storage cabinet.

[0111] In this embodiment, identifying whether an energy storage cabinet has an abnormality involves determining whether each energy storage cabinet feature in the energy storage cabinet feature vector of each energy storage cabinet is within the range of the energy storage cabinet features of the energy storage cabinet. If all of the features are within the range, it is determined that the energy storage cabinet does not have an abnormality; otherwise, it is determined that the energy storage cabinet has an abnormality.

[0112] The beneficial effects of the above technology are: obtaining historical abnormal control data of the distributed energy storage system, performing anomaly detection on the distributed energy storage system based on energy storage status data, operating status data and historical abnormal control data, determining real-time abnormal data and real-time normal data, and being able to quickly locate and handle anomalies, improve the accuracy of anomaly detection, reduce abnormal response time, improve system reliability and operating efficiency, and reduce operation and maintenance costs. Example 5:

[0113] Based on Example 4, a method for intelligent charge and discharge control of a distributed energy storage system is provided, which determines an abnormal control instruction based on real-time abnormal data and historical abnormal control data, including:

[0114] For each abnormal component in the real-time abnormal data, the component abnormality category corresponding to the component category vector with the greatest similarity to the component feature vector of each component is selected as the real-time component abnormality of each component;

[0115] Selecting, from all component exception vectors in the component exception category corresponding to the real-time component exception of each component, a component exception control instruction corresponding to the component exception vector having the greatest similarity to the component feature vector of each component as the first real-time exception control sub-instruction of each component;

[0116] For each energy storage cabinet with an abnormality in the real-time abnormal data, select the energy storage cabinet abnormality category corresponding to the energy storage cabinet category vector with the greatest similarity to the energy storage cabinet feature vector of each energy storage cabinet as the real-time energy storage cabinet abnormality of each energy storage cabinet;

[0117] From all energy storage cabinet abnormality vectors in the energy storage cabinet abnormality category corresponding to the real-time energy storage cabinet abnormality of each energy storage cabinet, select an energy storage cabinet abnormality control instruction corresponding to the energy storage cabinet abnormality vector having the greatest similarity to the energy storage cabinet feature vector of each energy storage cabinet as the second real-time abnormality control sub-instruction of each energy storage cabinet;

[0118] Based on the first real-time abnormality control sub-instructions of all abnormal components in the real-time abnormality data and the second real-time abnormality control sub-instructions of all abnormal energy storage cabinets, an abnormality control instruction of the distributed energy storage system is determined.

[0119] In this embodiment, for each component with an abnormality, its component feature vector is extracted from the real-time abnormality data. These feature vectors contain information about the current operating status of the component, such as temperature, power conversion efficiency, fault code, etc. A similarity calculation is performed between the component feature vector of each component and the component category vector determined in the historical abnormality control data. The similarity between vectors can be measured using methods such as Euclidean distance and cosine similarity. The component abnormality category corresponding to the component category vector with the greatest similarity to the component feature vector is selected as the real-time component abnormality of the component. This means that the abnormal characteristics of the current component match the most similar abnormality category in the historical data.

[0120] In this embodiment, for each component's real-time component anomaly category, all component anomaly vectors within that category are extracted. These vectors contain specific characteristics of historical anomalies. A similarity calculation is performed between each component's component feature vector and each component anomaly vector within that category, and the component anomaly vector with the greatest similarity to the component feature vector is found. The component anomaly control instruction corresponding to the component anomaly vector with the greatest similarity to the component feature vector is selected as the first real-time anomaly control sub-instruction for that component. These instructions are generated based on successful experience handling similar anomalies in historical data.

[0121] In this embodiment, for each energy storage cabinet with an abnormality, its energy storage cabinet feature vector is extracted from the real-time abnormality data. These feature vectors contain information about the current operating status of the energy storage cabinet, such as the current remaining power, health status, temperature, etc. In this embodiment, a similarity calculation is performed between the energy storage cabinet feature vector of each energy storage cabinet and the energy storage cabinet category vector determined in the historical abnormality control data. Methods such as Euclidean distance and cosine similarity can be used to measure the similarity between vectors. The energy storage cabinet abnormality category corresponding to the energy storage cabinet category vector with the greatest similarity to the energy storage cabinet feature vector is selected as the real-time energy storage cabinet abnormality of the energy storage cabinet. This means that the abnormal characteristics of the current energy storage cabinet match the most similar abnormal category in the historical data.

[0122] In this embodiment, for each energy storage cabinet's real-time energy storage cabinet anomaly category, all energy storage cabinet anomaly vectors within that category are extracted. These vectors contain the specific characteristics of historical anomalies. A similarity calculation is performed between the energy storage cabinet feature vector of each energy storage cabinet and each energy storage cabinet anomaly vector within that category, and the energy storage cabinet anomaly vector with the greatest similarity to the energy storage cabinet feature vector is found. The energy storage cabinet anomaly control instruction corresponding to the energy storage cabinet anomaly vector with the greatest similarity to the energy storage cabinet feature vector is selected as the second real-time anomaly control sub-instruction for that energy storage cabinet. These instructions are generated based on successful experience handling similar anomalies in historical data.

[0123] In this embodiment, the first real-time abnormality control sub-instructions for all abnormal components and the second real-time abnormality control sub-instructions for all abnormal energy storage cabinets are summarized. Based on the summarized abnormality control sub-instructions, abnormality control instructions for the distributed energy storage system are generated. These instructions guide the system to take appropriate measures to handle the current abnormal situation, such as isolating the faulty component, activating the cooling system, and cutting off the power supply.

[0124] The beneficial effects of the above technology are: determining abnormal control instructions based on real-time abnormal data and historical abnormal control data can quickly respond to system failures, reduce the impact of abnormalities on system operation, realize accurate calling of abnormal control instructions, improve instruction matching rate and abnormality handling efficiency, and enhance system stability and reliability. Example 6:

[0125] Based on Example 4, a method for intelligently controlling charging and discharging of a distributed energy storage system analyzes real-time normal data and power data to determine energy storage constraints, coordination constraints, and an arbitrage profit maximization objective function, including:

[0126] Based on the time-of-use electricity price policy in the power data, the specified operating cycle is divided into multiple time periods;

[0127] Determine the load demand for each period based on the power demand data and all periods after the designated operating cycle is divided;

[0128] Based on the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data and all time periods after the specified operation cycle is divided, determining the energy storage constraint conditions of each energy storage cabinet without abnormalities in the real-time normal data, where the energy storage constraint conditions include charge and discharge power constraints, state of charge constraints, and energy storage cabinet life constraints;

[0129] Determining coordination constraints based on the load demand for each time period and the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data, where the coordination constraints include grid interaction constraints and power balance constraints;

[0130] Based on power demand data, time-of-use electricity price policy, load demand in all time periods, and the energy storage status sub-data of each energy storage cabinet without abnormalities in real-time normal data, the arbitrage profit maximization objective function is determined.

[0131] In this embodiment, the time-of-use electricity price policy is extracted from the power data, including the specific time ranges and corresponding electricity prices of the peak period, the flat period and the off-peak period.

[0132] In this embodiment, a designated operating cycle (e.g., a 24-hour day) is divided into multiple time periods based on a time-of-use electricity price policy. For example, a day is divided into 0:00-8:00 (off-peak period), 8:00-14:00 (normal period), 14:00-22:00 (peak period), and 22:00-24:00 (normal period).

[0133] In this embodiment, real-time power demand data for the target area is extracted from the power data, including total power demand and power change rate. For example, the total power demand for the target area during peak hours is 1000kW, during flat hours is 800kW, and during off-peak hours is 500kW. Based on the power demand data and time period divisions, the average load demand for each time period is calculated. For example, the average load demand during peak hours is 1000kW, during flat hours is 800kW, and during off-peak hours is 500kW.

[0134] In this embodiment, based on the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data and all time periods after the specified operation cycle is divided, the energy storage constraint conditions of each energy storage cabinet without abnormalities in the real-time normal data are determined. The calculation formulas for the charge and discharge power constraint, the state of charge constraint, and the energy storage cabinet life constraint can be expressed as follows:

[0135] Charge and discharge power constraints:

[0136] ;

[0137] State of charge constraints:

[0138] ;

[0139] Energy storage cabinet life constraints:

[0140] ;

[0141] in, represents the charging power of the i-th energy storage cabinet in the real-time normal data of the t-th period, represents the discharge power of the i-th energy storage cabinet in the real-time normal data of the t-th period, Indicates the charge state of the i-th energy storage cabinet in the real-time normal data of the t-th period, represents the full charge capacity of the i-th energy storage cabinet in real-time normal data, Indicates the state of charge of the i-th energy storage cabinet in the real-time normal data of the t-1th period, 、 Respectively represent the charging efficiency and discharging efficiency of the i-th energy storage cabinet in real-time normal data, 、 They represent the maximum charging power and maximum discharging power of the i-th energy storage cabinet in real-time normal data, 、 They represent the minimum state of charge and maximum state of charge of the i-th energy storage cabinet in the real-time normal data, Indicates the health status of the i-th energy storage cabinet in real-time normal data, represents the safety margin of the energy storage cabinet, N1 represents the number of time periods in a specified operating cycle, β represents the energy storage cabinet aging acceleration factor, Indicates the maximum number of charge and discharge switching times of the i-th energy storage cabinet in real-time normal data, Indicates the length of the t-th period.

[0142] In this embodiment, the energy storage cabinet aging acceleration factor β may have a value range of 1.2-1.8.

[0143] In this embodiment, .

[0144] In this embodiment, the safety margin of the energy storage cabinet The value is 0.01 to 0.1, which can prevent the battery from being overcharged or deeply discharged, and reduce the risk of electrochemical aging or instability.

[0145] In this embodiment, based on the load demand of each time period and the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data, the coordination constraint conditions are determined. The calculation formulas for the grid interaction constraint and the power balance constraint can be expressed as follows:

[0146] Grid interaction constraints:

[0147] ;

[0148] Power balance constraints:

[0149] ;

[0150] in, Indicates the maximum allowable power of the grid interface, represents the power regulation coefficient of the power grid, represents the electricity price in the tth period, TC represents the electricity price threshold during the peak and valley periods, represents the power purchased and sold by the power grid in the tth period, represents the load demand in the tth period, represents the power loss of the distributed energy storage system, and N2 represents the number of energy storage cabinets.

[0151] In this embodiment, the power grid power regulation coefficient Indicates the power exchange regulation index between the distributed energy storage system and the power grid.

[0152] In this embodiment, the peak-valley electricity price threshold is used to distinguish the critical value of the peak-valley period, for example, TC=0.6 yuan / kWh.

[0153] In this embodiment, the power grid purchases and sells power A positive value indicates purchasing electricity from the grid. A negative value indicates that electricity is sold from the grid.

[0154] In this embodiment, the arbitrage profit maximization objective function is determined based on the power demand data, the time-of-use electricity price policy, the load demand of all time periods, and the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data. The arbitrage profit maximization objective function can be expressed as:

[0155] ;

[0156] ;

[0157] in, represents the electricity purchase price in the tth period, represents the unit cycle cost of the i-th energy storage cabinet in real-time normal data, Indicates the equivalent life loss of the i-th energy storage cabinet in the specified operating cycle in the real-time normal data, represents the life loss sensitivity coefficient of the i-th energy storage cabinet in real-time normal data, represents the aging index of the battery unit of the i-th energy storage cabinet in the real-time normal data, represents the electricity price in the tth period, It represents the average load demand of all periods in the specified operating cycle. represents the standard deviation of load demand in all periods within a specified operating cycle, θ represents the attenuation impact factor, α represents the price adjustment coefficient under critical load conditions, and CBa represents the benchmark electricity price in the time-of-use electricity price policy. represents the indicator function at the tth period, Indicates the safety margin of power capacity.

[0158] In this embodiment, the life loss sensitivity coefficient of the energy storage cabinet is The value range is arrive , reflecting the aging rate or loss degree of the battery unit of the energy storage cabinet.

[0159] In this embodiment, the aging index of the battery unit of the energy storage cabinet is The value range is 0.5-2, which is used to describe the nonlinear effect of the change in charge state on the life loss of the battery cell during the cyclic charge and discharge process. The specific value needs to be determined by the aging test data of the battery cell or the cycle life curve fitting provided by the manufacturer.

[0160] In this embodiment, the value range of the electricity price adjustment coefficient α under critical load conditions is 0.2-1, so as to increase the electricity sales revenue when the load is tight and encourage the energy storage system to discharge more.

[0161] In this embodiment, the safety margin of power capacity The value range is 0.05 to 0.2 , which represents the maximum allowable power of the grid interface A reserve margin is set aside to ensure safe operation when the load is close to the maximum capacity to prevent overload or instability.

[0162] In this embodiment, the unit cycle cost Indicates the cycle cost per unit capacity of the energy storage cabinet.

[0163] In this embodiment, the attenuation influence factor θ ranges from 0.5 to 2, which controls the nonlinear effect of the health status of the energy storage cabinet on the electricity price.

[0164] The beneficial effects of the above technologies are: analyzing real-time normal data and power data, determining energy storage constraints, coordination constraints, and the objective function for maximizing arbitrage returns, which can improve peak-valley arbitrage returns, extend the cycle life of energy storage cabinets, improve the adaptability of control strategies, optimize the charging and discharging strategies of distributed energy storage systems, improve system operating efficiency, reduce electricity costs, and ensure system safety and reliability. Example 7:

[0165] Based on Example 6, a method for intelligently controlling charging and discharging of a distributed energy storage system generates optimal charging and discharging control instructions, including:

[0166] A control model is constructed based on the charging and discharging power constraints, state of charge constraints, and energy storage cabinet life constraints in the energy storage constraints, as well as the grid interaction constraints, power balance constraints, and arbitrage profit maximization objective function in the coordination constraints.

[0167] The real-time normal data and power data are input into the control model, and the control model generates an optimal charge and discharge control instruction, wherein the optimal charge and discharge control instruction includes an optimal charge and discharge control sub-instruction for each energy storage cabinet in the real-time normal data.

[0168] In this embodiment, real-time normal data and power data are input into the control model. Based on this input data, the control model combines energy storage constraints, coordination constraints, and the objective function of maximizing arbitrage returns to solve the optimal charge and discharge control instructions using an optimization algorithm (such as linear programming, dynamic programming, or genetic algorithm).

[0169] In this embodiment, the optimal charge and discharge control instruction generated by the control model includes an optimal charge and discharge control sub-instruction for each energy storage cabinet.

[0170] The beneficial effects of the above technologies are: generating optimal charge and discharge control instructions, balancing economy and safety, ensuring that the distributed energy storage system maximizes economic benefits while meeting safety and reliability, and improving the system's intelligent operation level. Example 8:

[0171] The present invention provides a distributed energy storage system charging and discharging intelligent control cloud platform for executing the distributed energy storage system charging and discharging intelligent control method described in any one of embodiments 1 to 7, with reference to Figure 2 ,include:

[0172] Monitoring module: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, and real-time monitoring of the power data of the target area of ​​the distributed storage system;

[0173] Abnormal control module: obtains historical abnormal control data of the distributed energy storage system, performs abnormality detection on the distributed energy storage system based on energy storage status data, operating status data and historical abnormal control data, determines real-time abnormal data and real-time normal data, and determines abnormal control instructions based on the real-time abnormal data and historical abnormal control data;

[0174] Charge and discharge control module: Analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, and generates optimal charge and discharge control instructions;

[0175] Execution module: executes abnormal control instructions and optimal charge and discharge control instructions to realize intelligent charge and discharge control of distributed energy storage systems.

[0176] The beneficial effects of the above technology include: analyzing detected energy storage status data, operating status data, and historical abnormal control data to determine real-time abnormal data and real-time normal data, determining abnormal control instructions and executing them, analyzing real-time normal data and power data to determine energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, generating and executing optimal charge and discharge control instructions, and realizing intelligent charge and discharge control of distributed energy storage systems. This can improve the instruction matching rate and the efficiency of abnormality detection and processing, reduce abnormality response time, optimize charge and discharge strategies, reduce electricity costs, balance economy and safety, and improve system operation efficiency and reliability.

[0177] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for intelligent control of charge and discharge of a distributed energy storage system, characterized in that: include: S1: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, and real-time monitoring of the power data in the target area of ​​the distributed storage system; S2: Obtain historical abnormal control data of the distributed energy storage system, perform abnormality detection on the distributed energy storage system based on the energy storage status data, operating status data, and historical abnormal control data, determine real-time abnormal data and real-time normal data, and determine abnormal control instructions based on the real-time abnormal data and historical abnormal control data; S3: Analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, and generates optimal charge and discharge control instructions; S4: Execute abnormal control instructions and optimal charge and discharge control instructions to achieve intelligent charge and discharge control of the distributed energy storage system; The process involves obtaining historical abnormal control data of the distributed energy storage system, performing abnormality detection on the distributed energy storage system based on the energy storage status data, operating status data, and historical abnormal control data, and determining real-time abnormal data and real-time normal data, including: Acquire historical abnormality control sub-data of the distributed energy storage system within a plurality of specified operating cycles, and determine historical abnormality control data based on the historical abnormality control sub-data within all specified operating cycles; performing a first cluster analysis on component anomaly vectors of all component anomalies of the same component in all historical anomaly control sub-data in the historical anomaly control data, and determining multiple component anomaly categories of each component of the distributed energy storage system based on the first cluster analysis result; determining a component category vector for each component anomaly category of each component based on all component anomaly vectors for each component anomaly category of each component of the distributed energy storage system; Extracting features from the running status sub-data of each component in the running status data to determine a component feature vector of each component; performing a second cluster analysis on energy storage cabinet abnormality vectors of all energy storage cabinet abnormalities in all historical abnormality control sub-data in the historical abnormality control data, and determining a plurality of energy storage cabinet abnormality categories of the energy storage cabinets of the distributed energy storage system based on the second cluster analysis results; determining an energy storage cabinet category vector of each energy storage cabinet abnormality category based on all energy storage cabinet abnormality vectors of each energy storage cabinet abnormality category of the energy storage cabinets of the distributed energy storage system; Performing feature extraction on the energy storage status sub-data of each energy storage cabinet in the energy storage status data to determine an energy storage cabinet feature vector for each energy storage cabinet; Based on the preset component feature matrix and component feature vector of each component, identify whether each component has an abnormality; based on the preset energy storage cabinet feature matrix and energy storage cabinet feature vector of each energy storage cabinet, identify whether each energy storage cabinet has an abnormality; Real-time abnormal data is determined based on the operating status sub-data and component feature vectors of all components with abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets with abnormalities. Real-time normal data is determined based on the operating status sub-data and component feature vectors of all components without abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets without abnormalities.

2. The intelligent charge and discharge control method for a distributed energy storage system according to claim 1, characterized in that: The energy storage cabinet includes at least a battery unit, an intelligent air conditioner, a fire extinguishing device, and a first monitoring group, wherein the first monitoring group includes at least a voltage sensor, a current sensor, a temperature sensor, a smoke alarm, and a water intrusion alarm; The power data at least includes power demand data of a target area of ​​the distributed storage system and a time-of-use electricity price policy.

3. The intelligent charge and discharge control method for a distributed energy storage system according to claim 2, characterized in that: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, and real-time monitoring of the operating status data of the distributed energy storage system, including: Real-time monitoring of the energy storage status sub-data of each energy storage cabinet in the target area of ​​the distributed energy storage system, and determining the energy storage status data of the distributed energy storage system based on the energy storage status sub-data of all energy storage cabinets, wherein the energy storage status sub-data includes at least the current remaining power, full charge capacity, health status, charging efficiency, discharging efficiency, maximum charging power, maximum discharging power, minimum state of charge, maximum state of charge, maximum number of charge and discharge switching times, and unit cycle cost; The operating status sub-data of each component of the distributed energy storage system is monitored in real time, and the operating status data of the distributed energy storage system is determined based on the operating status sub-data of all components.

4. The method for intelligent charge and discharge control of a distributed energy storage system according to claim 3, characterized in that: The historical abnormality control sub-data includes multiple component abnormalities, multiple energy storage cabinet abnormalities, component abnormality vectors for each component abnormality, component abnormality control instructions, and energy storage cabinet abnormality vectors for each energy storage cabinet abnormality, and energy storage cabinet abnormality control instructions; Each component anomaly category includes multiple component anomaly vectors; Each energy storage cabinet abnormality category includes multiple energy storage cabinet abnormality vectors.

5. The intelligent charge and discharge control method for a distributed energy storage system according to claim 4, characterized in that: Determine abnormal control instructions based on real-time abnormal data and historical abnormal control data, including: For each abnormal component in the real-time abnormal data, the component abnormality category corresponding to the component category vector with the greatest similarity to the component feature vector of each component is selected as the real-time component abnormality of each component; Selecting, from all component exception vectors in the component exception category corresponding to the real-time component exception of each component, a component exception control instruction corresponding to the component exception vector having the greatest similarity to the component feature vector of each component as the first real-time exception control sub-instruction of each component; For each energy storage cabinet with an abnormality in the real-time abnormal data, select the energy storage cabinet abnormality category corresponding to the energy storage cabinet category vector with the greatest similarity to the energy storage cabinet feature vector of each energy storage cabinet as the real-time energy storage cabinet abnormality of each energy storage cabinet; From all energy storage cabinet abnormality vectors in the energy storage cabinet abnormality category corresponding to the real-time energy storage cabinet abnormality of each energy storage cabinet, select an energy storage cabinet abnormality control instruction corresponding to the energy storage cabinet abnormality vector having the greatest similarity to the energy storage cabinet feature vector of each energy storage cabinet as the second real-time abnormality control sub-instruction of each energy storage cabinet; Based on the first real-time abnormality control sub-instructions of all abnormal components in the real-time abnormality data and the second real-time abnormality control sub-instructions of all abnormal energy storage cabinets, an abnormality control instruction of the distributed energy storage system is determined.

6. The intelligent charge and discharge control method for a distributed energy storage system according to claim 4, characterized in that: Analyze real-time normal data and power data to determine energy storage constraints, coordination constraints, and the objective function for maximizing arbitrage returns, including: Based on the time-of-use electricity price policy in the power data, the specified operating cycle is divided into multiple time periods; Determine the load demand for each period based on the power demand data and all periods after the designated operating cycle is divided; Based on the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data and all time periods after the specified operation cycle is divided, determining the energy storage constraint conditions of each energy storage cabinet without abnormalities in the real-time normal data, where the energy storage constraint conditions include charge and discharge power constraints, state of charge constraints, and energy storage cabinet life constraints; Determining coordination constraints based on the load demand for each time period and the energy storage status sub-data of each energy storage cabinet without abnormalities in the real-time normal data, where the coordination constraints include grid interaction constraints and power balance constraints; Based on power demand data, time-of-use electricity price policy, load demand in all time periods, and the energy storage status sub-data of each energy storage cabinet without abnormalities in real-time normal data, the arbitrage profit maximization objective function is determined.

7. The intelligent charge and discharge control method for a distributed energy storage system according to claim 6, characterized in that: Generate optimal charge and discharge control instructions, including: A control model is constructed based on the charging and discharging power constraints, state of charge constraints, and energy storage cabinet life constraints in the energy storage constraints, as well as the grid interaction constraints, power balance constraints, and arbitrage profit maximization objective function in the coordination constraints. The real-time normal data and power data are input into the control model, and the control model generates an optimal charge and discharge control instruction, wherein the optimal charge and discharge control instruction includes an optimal charge and discharge control sub-instruction for each energy storage cabinet in the real-time normal data.

8. A distributed energy storage system charging and discharging intelligent control cloud platform, characterized in that: The method for intelligently controlling charging and discharging of a distributed energy storage system according to any one of claims 1 to 7 comprises: Monitoring module: Real-time monitoring of the energy storage status data of all energy storage cabinets in the target area of ​​the distributed energy storage system, real-time monitoring of the operating status data of the distributed energy storage system, and real-time monitoring of the power data of the target area of ​​the distributed storage system; Abnormal control module: obtains historical abnormal control data of the distributed energy storage system, performs abnormality detection on the distributed energy storage system based on energy storage status data, operating status data and historical abnormal control data, determines real-time abnormal data and real-time normal data, and determines abnormal control instructions based on the real-time abnormal data and historical abnormal control data; Charge and discharge control module: Analyzes real-time normal data and power data, determines energy storage constraints, coordination constraints, and the arbitrage profit maximization objective function, and generates optimal charge and discharge control instructions; Execution module: executes abnormal control instructions and optimal charge and discharge control instructions to realize intelligent charge and discharge control of distributed energy storage systems; The abnormality control module includes: Acquire historical abnormality control sub-data of the distributed energy storage system within a plurality of specified operating cycles, and determine historical abnormality control data based on the historical abnormality control sub-data within all specified operating cycles; performing a first cluster analysis on component anomaly vectors of all component anomalies of the same component in all historical anomaly control sub-data in the historical anomaly control data, and determining multiple component anomaly categories of each component of the distributed energy storage system based on the first cluster analysis result; determining a component category vector for each component anomaly category of each component based on all component anomaly vectors for each component anomaly category of each component of the distributed energy storage system; Extracting features from the running status sub-data of each component in the running status data to determine a component feature vector of each component; performing a second cluster analysis on energy storage cabinet abnormality vectors of all energy storage cabinet abnormalities in all historical abnormality control sub-data in the historical abnormality control data, and determining a plurality of energy storage cabinet abnormality categories of the energy storage cabinets of the distributed energy storage system based on the second cluster analysis results; determining an energy storage cabinet category vector of each energy storage cabinet abnormality category based on all energy storage cabinet abnormality vectors of each energy storage cabinet abnormality category of the energy storage cabinets of the distributed energy storage system; Performing feature extraction on the energy storage status sub-data of each energy storage cabinet in the energy storage status data to determine an energy storage cabinet feature vector for each energy storage cabinet; Based on the preset component feature matrix and component feature vector of each component, identify whether each component has an abnormality; based on the preset energy storage cabinet feature matrix and energy storage cabinet feature vector of each energy storage cabinet, identify whether each energy storage cabinet has an abnormality; Real-time abnormal data is determined based on the operating status sub-data and component feature vectors of all components with abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets with abnormalities. Real-time normal data is determined based on the operating status sub-data and component feature vectors of all components without abnormalities and the energy storage status sub-data and energy storage cabinet feature vectors of all energy storage cabinets without abnormalities.

Images