Cloud-edge collaborative lithium battery energy storage power station safety management and control system and method

Abstract

The application discloses a kind of cloud edge coordination's lithium battery energy storage power station safety management and control system and method, wherein system includes battery monitoring analysis subsystem, local computing decision subsystem, cloud computing storage subsystem and electric-thermal-safety coordination execution subsystem.Between which battery monitoring analysis subsystem is used to monitor the electric, heat, gas, smoke and mechanical parameter of energy storage power station in real time;Local computing decision subsystem is used to simulate and obtain the bottom parameter of single battery level according to lithium battery mechanism model, identify instantaneous safety failure and gradual failure from mechanism, issue control instruction;Cloud computing storage subsystem calculates the health state of single battery level according to data-driven model, identifies the gradual failure of battery from data level;Electric-thermal-safety coordination execution subsystem is used to manage and regulate whole station or local battery pack.The above-mentioned system identifies gradual failure through mechanism model and data-driven model, improves the reliability and safety of identification.

Description

Technical Field

[0001] This invention belongs to the field of electrochemical energy storage power station management and control technology, and specifically relates to a safety management and control system and method for lithium battery energy storage power stations based on cloud-edge collaboration. Background Technology

[0002] With the release of the "Guiding Opinions on Accelerating the Development of New Energy Storage" by the State Council, the energy storage industry has begun to move from the early stages of commercialization to large-scale development. Compared with traditional physical energy storage technologies (compressed air, pumped hydro storage, flywheel), electrochemical energy storage systems based on lithium-ion batteries have developed rapidly due to their advantages such as high energy efficiency and fast response speed. They have become an effective solution for energy strategies such as smart energy network construction, electrification of end-use energy, and large-scale access to renewable energy. However, frequent energy storage accidents have made the safety of energy storage systems a key factor restricting their large-scale application.

[0003] Existing lithium-ion battery energy storage systems that rely on local monitoring parameters for safety management cannot identify evolutionary battery faults, which can easily escalate into severe internal short circuits over time, leading to unexpected safety incidents. Using cloud servers to store local monitoring parameters for fault diagnosis is inefficient due to limitations in data volume and frequency, and analysis is often delayed, relying solely on data calculations. Furthermore, existing energy storage systems suffer from a limited number of temperature monitoring points, unavoidable parallel battery connections, and static control parameters, making it difficult to identify temperature extremes within the system, battery aging conditions, and the impact of control parameters. Therefore, accuracy and reliability need improvement. Summary of the Invention

[0004] Purpose of the invention: The purpose of this invention is to propose a cloud-edge collaborative safety management and control system for lithium battery energy storage power stations. Through cloud-edge integration, the local computing decision subsystem obtains the underlying parameters of individual batteries based on real-time parameters and simulation of lithium battery mechanism models. It then judges transient safety faults and some gradual faults based on real-time and simulation parameters. At the same time, the cloud computing storage subsystem uses a data-driven model to identify other gradual faults. The cloud-edge integration improves the accuracy and reliability of safety management and control.

[0005] Another objective of this invention is to propose a control method for the aforementioned cloud-edge collaborative lithium battery energy storage power station safety management and control system.

[0006] The cloud-edge collaborative lithium battery energy storage power station safety management and control system of the present invention includes: a battery monitoring and analysis subsystem for real-time monitoring of the electrical, thermal, gas, smoke, and mechanical parameters of the energy storage power station; a local computing and decision-making subsystem equipped with a whole-cell-scale lithium battery mechanism model for simulating the real-time parameters of the energy storage power station monitored by the battery monitoring and analysis subsystem, obtaining low-level parameters at the individual battery level, identifying transient safety faults and gradual faults from a mechanistic perspective, issuing control commands, and providing early warnings; a cloud computing and storage subsystem equipped with a data-driven model and training algorithm, receiving the real-time monitoring parameters of the energy storage power station and the low-level parameters at the individual battery level obtained from the simulation uploaded by the local computing and decision-making subsystem, calculating the health status of individual batteries based on the data-driven model, identifying gradual faults of the battery from a data perspective, and feeding back the health status and gradual faults at the individual battery level to the local computing and decision-making subsystem; and an electrical-thermal-safety collaborative execution subsystem for receiving control commands issued by the local computing and decision-making subsystem and managing and adjusting the entire station or a portion of the battery pack.

[0007] Furthermore, the battery monitoring and analysis subsystem includes a battery system general management system and an energy management system for monitoring the electrical and thermal parameters of the energy storage power station, as well as gas sensors, smoke sensors, and torque sensors for monitoring the gas, smoke, and mechanical parameters of the energy storage power station.

[0008] Furthermore, the battery system master management system has multiple battery pack management units connected in parallel, and each battery pack management unit has multiple individual battery cell management units connected in parallel.

[0009] Furthermore, the electric-thermal-safety coordinated execution subsystem includes an energy storage inverter system, an air conditioning and refrigeration system, a liquid cooling circulation system, and a fire extinguishing system.

[0010] Furthermore, the air conditioning refrigeration system, liquid cooling circulation system, and fire extinguishing system are integrated into a single design. The cold source of the liquid cooling circulation system is provided by heat exchange before refrigerant evaporation in the air conditioning refrigeration system, and the heat source of the liquid cooling circulation system is provided by heat exchange before refrigerant condensation in the air conditioning refrigeration system. The fire sprinklers of the fire extinguishing system are located at the ends of the liquid cooling pipes extending from the liquid cooling circulation system to each battery rack or battery cabinet.

[0011] Furthermore, the lithium battery mechanism models set in the local computing decision subsystem include a battery thermal model, a battery electro-thermal model, a battery electrochemical-thermal model, a battery electro-thermal-fluid model, and / or an open-circuit potential-state-of-charge curve.

[0012] Furthermore, the local computing decision-making subsystem includes an edge computing terminal, a simulation computer, and a comprehensive management system. The edge computing terminal is used to receive real-time data monitored by the battery monitoring and analysis subsystem and low-level data obtained by simulation by the simulation computer, and uploads the received data to the cloud computing storage subsystem after preprocessing. The simulation computer is used to obtain the low-level parameters of the battery based on the real-time data and the mechanism model of the lithium battery. The comprehensive management system is used to determine transient and gradual faults based on the real-time data and low-level data, and controls the electric-thermal-safety coordinated execution subsystem to adjust the energy storage power station in conjunction with the battery health status and gradual faults evaluated by the cloud computing storage subsystem.

[0013] The cloud-edge collaborative lithium battery energy storage power station safety management method of the present invention includes the following steps:

[0014] S1: The battery monitoring and analysis subsystem monitors the electrical, thermal, gas, smoke, and mechanical parameters of the energy storage power station and transmits them to the local computing and decision-making subsystem;

[0015] S2: The local computing decision subsystem obtains the low-level parameters of a single cell through the mechanism model simulation of lithium battery, and uploads the received real-time parameters of the energy storage power station and the low-level parameters obtained by simulation to the cloud computing storage subsystem after preprocessing.

[0016] S3: The cloud computing storage subsystem uses a data-driven model to evaluate the health status of each individual battery based on real-time parameters and underlying parameters uploaded by the local computing decision subsystem, and identifies gradual failures of the battery, and sends the health status and gradual failures to the local computing decision subsystem.

[0017] S4: The local computing decision subsystem compares real-time parameters and underlying parameters with the corresponding set thresholds to determine the instantaneous faults of the battery, and determines the gradual faults of the battery based on the changing trends of real-time parameters and underlying parameters.

[0018] S5: The local computing decision subsystem combines its own fault assessment with the health status and gradual fault identification by the cloud computing storage subsystem to instruct the electrical-thermal-safety coordinated execution subsystem to adjust the energy storage power station.

[0019] Furthermore, in step S3, the cloud computing storage subsystem identifies gradual battery failures using an outlier detection algorithm.

[0020] Furthermore, the threshold values ​​corresponding to each parameter in step S4 are dynamic threshold values, set according to the health status of the battery.

[0021] Beneficial Effects: Compared with existing technologies, this invention has the following advantages: 1. By identifying gradual faults through both locally deployed mechanistic models and cloud-deployed data-driven models, the reliability and safety of identification are improved, overcoming the shortcomings of traditional energy storage power stations that only manage transient faults. 2. The combination of simulation data obtained from the locally deployed mechanistic model and real-time data collection solves the problem of limited monitoring in traditional energy storage power stations, accelerates fault identification and diagnosis, and improves the safety and operation and maintenance capabilities of the power station. 3. The integrated design of the air conditioning, liquid cooling, and fire protection system facilitates rapid cooling after battery overheating and the synergistic effect of liquid cooling and fire protection, achieving effective control of power station safety in the early stages of thermal runaway. 4. The mechanistic model can also be used to simulate the state of aging batteries, obtaining the threshold boundaries of key battery parameters under aging conditions, ensuring the accuracy and reliability of transient fault judgment under battery property aging changes. Attached Figure Description

[0022] Figure 1 This is a system block diagram of a cloud-edge collaborative lithium battery energy storage station safety management and control system according to an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the early warning and control process according to an embodiment of the present invention. Detailed Implementation

[0024] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0025] Reference Figure 1 According to an embodiment of the present invention, a cloud-edge collaborative lithium battery energy storage station safety management and control system includes a battery monitoring and analysis subsystem, a local monitoring and analysis subsystem, a cloud computing storage subsystem, and an electricity-heat-safety collaborative execution subsystem. The battery monitoring and analysis subsystem is used to monitor the electrical, thermal, gas, smoke, and mechanical parameters of the energy storage station in real time. The local computing and decision-making subsystem is equipped with a whole-cell-scale lithium battery mechanism model, used to simulate the real-time parameters of the energy storage station monitored by the battery monitoring and analysis subsystem, obtain low-level parameters at the individual battery level, identify transient safety faults and gradual faults from a mechanistic perspective, issue control commands, and provide early warnings. The cloud computing storage subsystem is equipped with a data-driven model and training algorithm, receives the real-time monitoring parameters of the energy storage station uploaded by the local computing and decision-making subsystem and the low-level parameters at the individual battery level obtained from the simulation, uses the data-driven model to calculate the health status of individual batteries, identifies gradual faults of the battery at the data level, and feeds back the health status and gradual faults at the individual battery level to the local computing and decision-making subsystem. The electric-thermal-safety collaborative execution subsystem receives control commands from the local computing and decision-making subsystem to manage and adjust the entire station or a portion of the battery pack. The battery monitoring and analysis subsystem, the local monitoring and analysis subsystem, and the electric-thermal-safety collaborative execution subsystem are deployed locally, while the cloud computing storage subsystem is deployed in the cloud.

[0026] The aforementioned safety management system uses a lithium battery mechanism model to simulate and obtain underlying parameters that are difficult to monitor directly based on real-time collected parameters. This data is then combined with the real-time parameters to identify transient and gradual battery faults, improving the accuracy of fault identification. Simultaneously, the cloud computing storage subsystem employs a data-driven model, incorporating historical data to identify gradual faults that are difficult to identify through the parameters themselves, further enhancing the accuracy and reliability of safety management. Furthermore, cloud-edge collaboration—with local identification of transient faults and cloud-edge collaborative identification of gradual faults—ensures rapid identification of transient faults while also providing early warnings and pre-adjustments for gradual faults, preventing the accumulation of gradual faults over time from leading to transient faults.

[0027] Reference Figure 1 In this embodiment, the battery monitoring and analysis subsystem includes a Battery System Management System (BMS) and an Energy Management System (EMS) for monitoring the electrical and thermal parameters of the energy storage power station, as well as gas sensors, smoke sensors, and torque sensors for monitoring the gas, smoke, and force parameters of the energy storage power station. The EMS is responsible for analyzing the power station's energy status and monitoring historical curve changes in data. It has the function of sharing fault and anomaly information with the local computing and decision-making subsystem, and sending instructions to the power storage inverter system (PCS) in the electric-thermal-safety collaborative subsystem. The torque sensor monitors the expansion modulus of each battery, providing identification parameters for gas expansion faults; the smoke sensor monitors the smoke concentration after a battery fire, providing identification parameters for fire control; the gas sensor monitors the concentration of volatile organic compounds (VOCs) and gases (carbon monoxide, hydrogen, and ethylene, etc.) generated by overheating side reactions during battery leakage, providing parameters for identifying gradual leakage faults and instantaneous overheating faults. A Battery Management System (BMS) can monitor and acquire electrical and thermal parameters at the individual battery cell level. Specifically, a Battery System Management Unit (BAMS) is set up in parallel under the BMS, multiple Battery Pack Management Units (BCMUs) are set up in parallel under the BAMS, and multiple Individual Battery Cell Management Units (BMUs) are set up in parallel under the BCMUs, forming a monitoring and management system for each level of the lithium battery.

[0028] The electro-thermal-safety coordinated execution subsystem, in addition to the PCS used for regulating electrical parameters of the energy storage power station, also includes an air conditioning and refrigeration system, a liquid cooling circulation system, and a fire suppression system for thermal management of the energy storage power station. In this embodiment, the air conditioning and refrigeration system, liquid cooling circulation system, and fire suppression system are integrated into a single design. The cold source for the liquid cooling circulation system is provided by heat exchange before refrigerant evaporation in the air conditioning and refrigeration system, and the heat source is provided by heat exchange before refrigerant condensation in the air conditioning and refrigeration system. The fire sprinklers of the fire suppression system are located at the ends of the liquid cooling pipes extending from the liquid cooling circulation system to each battery rack or battery cabinet. This integrated design of the air conditioning, liquid cooling, and fire suppression systems can solve the problem of rapid cooling of overheated batteries through the synergistic effect of liquid cooling and fire suppression, achieving effective control of the power station's safety in the early stages of thermal runaway.

[0029] Local decision-making involves edge computing terminals, simulation computers, and a comprehensive management system. The edge computing terminals primarily handle data preprocessing, performing operations such as differentiation, summation, difference, quotient calculation, extreme value extraction, and sorting on the acquired real-time data (i.e., the underlying parameters obtained from the simulation) to facilitate use by the simulation computer, the comprehensive management system, and the cloud storage subsystem. The simulation computer uses stored lithium battery mechanism models, combined with collected real-time parameters, to simulate and obtain underlying parameters that are difficult to monitor. Mechanism models include physical models, three-dimensional thermal models, electro-thermal models, electrochemical-thermal models, and electro-thermal-fluid models of the energy storage power station's battery compartment. These models are used to calculate the current magnitude and temperature distribution of individual cells based on dynamically changing real-time monitoring parameters. The open-circuit potential-state-of-charge (OCV-SOC) curve is also used to calculate the battery's state of charge (SOH). The comprehensive management system is used locally to determine transient and gradual faults based on real-time and underlying parameters. Transient faults are primarily identified by comparing set thresholds for each parameter with real-time parameters, combined with fault tree analysis, expert systems, and fuzzy logic. Gradual faults are mainly identified based on the changing trends of parameters. The integrated management system operates independently from the BMS and EMS. It combines locally identified faults with the health status and gradual fault assessments from the cloud storage subsystem, issuing adjustment commands to the electrical-thermal-safety collaborative execution subsystem. This system works in conjunction with the BMS and EMS to regulate the energy storage power station and maintain its safety and stability.

[0030] The cloud computing storage subsystem includes a cloud server and remote access terminals. The cloud server assesses battery health and identifies gradual failures based on parameters uploaded by the local computing decision-making subsystem. Users can access data from the cloud server via remote access terminals to monitor the operational status of the energy storage power station. The cloud computing storage subsystem needs to maintain a monitoring list and scheduling priorities for edge nodes to ensure business continuity when offline; it also provides data migration, backup, and recovery capabilities. The cloud server employs methods including, but not limited to, mirror caching, edge mirror site acceleration, and P2P distribution to rapidly process massive amounts of operational data.

[0031] Reference Figure 2 According to the safety management system of the present invention, the safety management of the energy storage power station is achieved through the following method:

[0032] S1: The battery monitoring and analysis subsystem monitors the electrical, thermal, gas, smoke, and mechanical parameters of the energy storage power station and transmits them to the local computing and decision-making subsystem;

[0033] S2: The local computing decision subsystem obtains the low-level parameters of a single cell through the mechanism model simulation of lithium battery, and uploads the received real-time parameters of the energy storage power station and the low-level parameters obtained by simulation to the cloud computing storage subsystem after preprocessing.

[0034] S3: The cloud computing storage subsystem uses a data-driven model to evaluate the health status of each individual battery based on real-time parameters and underlying parameters uploaded by the local computing decision subsystem, and identifies gradual failures of the battery, and sends the health status and gradual failures to the local computing decision subsystem.

[0035] S4: The local computing decision subsystem compares real-time parameters and underlying parameters with the corresponding set thresholds to determine the instantaneous faults of the battery, and determines the gradual faults of the battery based on the changing trends of real-time parameters and underlying parameters.

[0036] S5: The local computing decision subsystem combines its own fault assessment with the health status and identified gradual faults evaluated by the cloud computing storage subsystem, and instructs the electrical-thermal-safety coordinated execution subsystem to adjust the energy storage power station.

[0037] The parameters include voltage (U), current (I), battery pack internal temperature (T), state of charge (SOC), insulation resistance (R), gas concentration (C), force parameters (F), and equalization current (I_equalization), enabling multi-dimensional fault monitoring. The local computing and decision-making subsystem uses an algorithm to calculate the self-discharge point capacity based on the equalization current, identifying gradual faults in micro-short circuits from a mechanistic perspective, issuing warnings, indicating and isolating the fault location, and guiding maintenance and inspection. The cloud computing storage subsystem communicates with the local computing and decision-making subsystem to track and analyze the target batteries corresponding to abnormal data. Based on historical charge-discharge curves and the self-discharge rate of the target batteries, it distinguishes whether the abnormal data cells are short-term cells (parallel cells with excessively rapid capacity degradation) or gradual faults. The cloud computing storage subsystem identifies gradual faults in micro-short circuits through outlier detection algorithms, including but not limited to LOF, IFOres, DBSCAN clustering, and RCF. The training of the data-driven model in the cloud computing storage subsystem can use simulation data of artificially induced faults from the mechanistic model in the local computing decision-making subsystem as samples to improve the fault identification capability and accuracy. In this embodiment, the threshold for key parameters used to determine transient faults is a dynamic threshold. This threshold is set based on the parameters simulated under different health conditions by the mechanistic model, and is further refined according to the real-time dynamic changes in the battery's health status as assessed by the cloud computing storage subsystem. This ensures the accuracy and reliability of transient fault judgment under battery aging conditions.

Claims

1. A cloud-edge collaborative safety management system for lithium battery energy storage power stations, characterized in that, include: The battery monitoring and analysis subsystem is used to monitor the electrical, thermal, gas, smoke, and mechanical parameters of the energy storage power station in real time. The local computing decision-making subsystem is equipped with a whole-cell-scale lithium battery mechanism model, which is used to simulate the parameters of the real-time energy storage power station monitored by the battery monitoring and analysis subsystem, obtain the underlying parameters at the individual battery level, identify instantaneous safety faults and gradual faults from the mechanism, issue control commands and provide early warnings. The cloud computing storage subsystem is equipped with a data-driven model and training algorithm. It receives real-time monitoring parameters of the energy storage power station and low-level parameters at the individual battery level obtained from simulation from the local computing decision subsystem. It is used to calculate the health status of individual batteries based on the data-driven model, identify gradual failures of batteries from the data level, and feed back the health status and gradual failures of individual batteries to the local computing decision subsystem. The electric-thermal-safety collaborative execution subsystem is used to receive control commands issued by the local computing and decision-making subsystem and to manage and adjust the entire station or a portion of the battery pack. The local computing decision subsystem also includes an integrated management system, which is used to determine transient and gradual faults based on real-time and underlying data, and to control the electric-thermal-safety coordinated execution subsystem to adjust the energy storage power station in conjunction with the battery health status and gradual faults assessed by the cloud computing storage subsystem.

2. The cloud-edge collaborative lithium battery energy storage power station safety management and control system according to claim 1, characterized in that, The battery monitoring and analysis subsystem includes a battery system general management system and an energy management system for monitoring the electrical and thermal parameters of the energy storage power station, as well as gas sensors, smoke sensors, and torque sensors for monitoring the gas, smoke, and mechanical parameters of the energy storage power station.

3. The cloud-edge collaborative lithium battery energy storage power station safety management and control system according to claim 2, characterized in that, The battery system's overall management system has multiple battery pack management units connected in parallel, and each battery pack management unit has multiple individual battery cell management units connected in parallel.

4. The cloud-edge collaborative lithium battery energy storage power station safety management and control system according to claim 1, characterized in that, The electro-thermal-safety coordinated execution subsystem includes an energy storage inverter system, an air conditioning and refrigeration system, a liquid cooling circulation system, and a fire extinguishing system.

5. The cloud-edge collaborative lithium battery energy storage power station safety management and control system according to claim 4, characterized in that, The air conditioning refrigeration system, liquid cooling circulation system, and fire extinguishing system are integrated into one design. The cold source of the liquid cooling circulation system is provided by the heat exchange before the refrigerant evaporates in the air conditioning refrigeration system, and the heat source of the liquid cooling circulation system is provided by the heat exchange before the refrigerant condenses in the air conditioning refrigeration system. The fire sprinklers of the fire extinguishing system are located at the end of the liquid cooling pipes that extend from the liquid cooling circulation system to each battery rack or battery cabinet.

6. The cloud-edge collaborative lithium battery energy storage power station safety management and control system according to claim 1, characterized in that, The lithium battery mechanism models set in the local computing decision subsystem include a battery thermal model, a battery electro-thermal model, a battery electrochemical-thermal model, a battery electro-thermal-fluid model, and / or an open-circuit potential-charge state curve.

7. The cloud-edge collaborative lithium battery energy storage power station safety management and control system according to claim 1, characterized in that, The local computing decision subsystem also includes an edge computing terminal and a simulation computer. The edge computing terminal is used to receive real-time data monitored by the battery monitoring and analysis subsystem and low-level data obtained by simulation by the simulation computer, and uploads the received data to the cloud computing storage subsystem after preprocessing. The simulation computer is used to obtain the low-level parameters of the battery based on the real-time data and the mechanism model of the lithium battery.

8. A cloud-edge collaborative method for the safety management and control of lithium battery energy storage power stations, characterized in that, Includes the following steps: S1: The battery monitoring and analysis subsystem monitors the electrical, thermal, gas, smoke, and mechanical parameters of the energy storage power station and transmits them to the local computing and decision-making subsystem; S2: The local computing decision subsystem obtains the low-level parameters of a single cell through the mechanism model simulation of lithium battery, and uploads the received real-time parameters of the energy storage power station and the low-level parameters obtained by simulation to the cloud computing storage subsystem after preprocessing. S3: The cloud computing storage subsystem uses a data-driven model to evaluate the health status of each individual battery based on real-time parameters and underlying parameters uploaded by the local computing decision subsystem, and identifies gradual failures of the battery, and sends the health status and gradual failures to the local computing decision subsystem. S4: The local computing decision subsystem compares real-time parameters and underlying parameters with the corresponding set thresholds to determine the instantaneous faults of the battery, and determines the gradual faults of the battery based on the changing trends of real-time parameters and underlying parameters. S5: The local computing decision subsystem combines its own fault assessment with the health status and gradual fault identification by the cloud computing storage subsystem to instruct the electrical-thermal-safety coordinated execution subsystem to adjust the energy storage power station.

9. The cloud-edge collaborative lithium battery energy storage power station safety management method according to claim 8, characterized in that, In step S3, the cloud computing storage subsystem identifies gradual battery failures using an outlier detection algorithm.

10. The cloud-edge collaborative lithium battery energy storage power station safety management method according to claim 8, characterized in that, The threshold values ​​corresponding to each parameter in step S4 are dynamic threshold values, which are set according to the threshold boundaries obtained by mechanistic model simulation of the battery under the health state obtained in the evaluation.

Images