Cooperative management method and device for high computing power terminal of energy storage power station

Abstract

The invention provides a collaborative management method for a high-computing-power terminal of an energy storage power station, and the method comprises the steps: carrying out the physical integration of a processor calculation unit and a storage unit, building a high-computing-power architecture, and building the high-computing-power terminal of the energy storage power station; an algorithm model is built on the high-computing-power architecture, and cooperation with the high-computing-power architecture is achieved by optimizing the algorithm model; a terminal intelligent management system is constructed based on the high-computing-power terminals, centralized management is carried out on the distributed high-computing-power terminals, algorithm models and data, and end-side cloud collaboration is achieved; the storage and calculation integrated high-computing-power architecture effectively breaks through a memory wall and a power consumption wall, reduces data transmission delay and energy consumption, and remarkably improves the calculation efficiency; through collaborative optimization of an energy storage algorithm and a high computing power architecture, and through an adaptive algorithm model and a data transmission design, the calculation acceleration performance is improved, and meanwhile, the data precision is ensured; the intelligent management system of the energy storage power station terminal improves the intelligent management level of the energy storage power station, and ensures the safety and stability of the operation of the power station.

Description

Technical Field

[0001] This application belongs to the field of energy storage power station technology, and specifically relates to a collaborative management method and device for high computing power terminals of energy storage power stations. Background Technology

[0002] During the operation of energy storage power stations, there are problems such as slow speed, high computing power requirements, high resource consumption, and slow transmission speed in the complex analysis of big data from millions of monitoring points. In traditional data processing methods for energy storage power stations, the storage unit and the computing unit are separated. This leads to high power consumption and slow speed when massive amounts of data are read and written frequently between the CPU processor and the storage particles, forming memory walls and power walls, which seriously affect computing efficiency.

[0003] Meanwhile, existing algorithms have poor compatibility with hardware architectures, failing to fully utilize the computing power advantages of the hardware, resulting in poor overall processing performance. Furthermore, the dispersed high-computing-power terminals within energy storage power stations lack an efficient centralized management system, making it difficult to achieve end-edge-cloud collaboration. This hinders the efficient integration and management of terminals, algorithm models, and data, thus restricting the intelligent operation and safety early warning capabilities of energy storage power stations. Summary of the Invention

[0004] To address the aforementioned problems, this application provides a collaborative management method for high-computing-power terminals in energy storage power stations, the improvement of which includes: Physically integrate processor computing units and storage units to establish a high-computing-power architecture and build a high-computing-power terminal for energy storage power stations; Build algorithm models on high-computing-power architectures and achieve synergy between the algorithm models and the high-computing-power architectures by optimizing the algorithm models. A terminal intelligent management system is built based on high-computing-power terminals to centrally manage distributed high-computing-power terminals, algorithm models, and data, thereby achieving end-edge-cloud collaboration.

[0005] Optionally, the physical integration of the processor computing unit and storage unit to establish a high-performance computing architecture includes: By bringing the processor computing unit and storage unit closer together, a highly integrated module is formed, establishing a high-computing-power architecture.

[0006] Optionally, the step of establishing a high-computing-power terminal for an energy storage power station includes: Select a suitable CPU based on the application scenarios and requirements of the high-computing-power terminal at the edge of the energy storage power station, and design the power supply circuit, functional circuit, peripheral circuit, communication interface and communication protocol of the circuit board based on its physical, electromagnetic and electrical characteristics. By integrating the CPU and high-performance computing architecture onto the circuit board and employing a multi-fin fanless heat dissipation structure, a high-performance computing terminal for energy storage power stations can be established.

[0007] Optionally, the step of building an algorithm model on a high-computing-power architecture and achieving synergy between the algorithm model and the high-computing-power architecture through optimization includes: Establish rules for high-computing-power terminals to acquire data from the in-cabin BMS system, clarify the frequency, format and verification method of data acquisition, and design data transmission rules between high-computing-power terminals and terminal intelligent management systems of energy storage power stations; Based on the characteristics of high-computing-power architecture, a general framework is used to build the algorithm model and unify the expression format of multi-framework algorithm models; The algorithm model optimizes the data layout by using data partitioning and storing data in the order of computation. It reduces the model size and computational redundancy by optimizing operators and compressing the model. It also reduces data loss and accuracy loss by using error compensation and other techniques, and achieves collaboration with high-computing architecture. The terminal intelligent management system is built on high-computing-power terminals. The data transmission rules between the high-computing-power terminals and the terminal intelligent management system include the triggering conditions, transmission protocols, and data synchronization mechanisms for data transmission.

[0008] Optionally, the terminal intelligent management system includes: The terminal management module is used to install and bind the terminal intelligent management module to distributed high-computing-power terminals; The subscription management module is used to support users' custom terminal data subscription needs; The operation log module is used to record the activity history of the energy storage power station and the operation and change reporting information of the high computing power terminal of the energy storage power station; The algorithm model management module is used to manage the activation, termination, iteration, and batch distribution of various algorithm models. The data management module is used to design data management rules and collect and monitor data from energy storage power stations in real time.

[0009] The edge-cloud collaboration module is used to manage the communication services between the BMS system and high-computing-power terminals, realize the interconnection between high-computing-power terminals in each cabin, establish an encrypted interconnection mechanism between distributed high-computing-power terminals and the terminal intelligent management system, and realize edge-edge device monitoring and edge-cloud data collaboration.

[0010] Based on the same inventive concept, this application also provides a collaborative management device for a high-computing-power terminal of an energy storage power station, the improvement of which includes: High-performance computing architecture unit is used to physically integrate processor computing units and storage units to establish a high-performance computing architecture and build a high-performance computing terminal for energy storage power stations. The algorithm collaboration unit is used to build algorithm models on high-computing-power architectures and achieve collaboration between the algorithm models and the high-computing-power architectures by optimizing the algorithm models. The edge-cloud collaboration unit is used to build a terminal intelligent management system based on high-computing-power terminals, and to centrally manage distributed high-computing-power terminals, algorithm models and data to achieve edge-cloud collaboration.

[0011] Optionally, the physical integration of the processor computing unit and storage unit to establish a high-performance computing architecture includes: By bringing the processor computing unit and storage unit closer together, a highly integrated module is formed, establishing a high-computing-power architecture.

[0012] Optionally, the step of establishing a high-computing-power terminal for an energy storage power station includes: Select a suitable CPU based on the application scenarios and requirements of the high-computing-power terminal at the edge of the energy storage power station, and design the power supply circuit, functional circuit, peripheral circuit, communication interface and communication protocol of the circuit board, taking into account physical characteristics, electromagnetic characteristics and electrical characteristics. By integrating the CPU and high-performance computing architecture onto the circuit board and employing a multi-fin fanless heat dissipation structure, a high-performance computing terminal for energy storage power stations can be established.

[0013] Optionally, the step of building an algorithm model on a high-computing-power architecture and achieving synergy between the algorithm model and the high-computing-power architecture through optimization includes: Establish rules for high-computing-power terminals to acquire data from the in-cabin BMS system, clarify the frequency, format and verification method of data acquisition, and design data transmission rules between high-computing-power terminals and terminal intelligent management systems of energy storage power stations; Based on the characteristics of high-computing-power architecture, a general framework is used to build the algorithm model and unify the expression format of multi-framework algorithm models; The algorithm model optimizes the data layout by using data partitioning and storing data in the order of computation. It reduces the model size and computational redundancy by optimizing operators and compressing the model. It also reduces data loss and accuracy loss by using error compensation and other techniques, and achieves collaboration with high-computing architecture. The terminal intelligent management system is built on high-computing-power terminals. The data transmission rules between the high-computing-power terminals and the terminal intelligent management system include the triggering conditions, transmission protocols, and data synchronization mechanisms for data transmission.

[0014] Optionally, the terminal intelligent management system includes: The terminal management module is used to install and bind the terminal intelligent management module to distributed high-computing-power terminals; The subscription management module is used to support users' custom terminal data subscription needs; The operation log module is used to record the activity history of the energy storage power station and the operation and change reporting information of the high computing power terminal of the energy storage power station; The algorithm model management module is used to manage the activation, termination, iteration, and batch distribution of various algorithm models. The data management module is used to design data management rules and collect and monitor data from energy storage power stations in real time.

[0015] The edge-cloud collaboration module is used to manage the communication services between the BMS system and high-computing-power terminals, realize the interconnection between high-computing-power terminals in each cabin, establish an encrypted interconnection mechanism between distributed high-computing-power terminals and the terminal intelligent management system, and realize edge-edge device monitoring and edge-cloud data collaboration.

[0016] On one hand, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described collaborative management method for a high-computing-power terminal of an energy storage power station.

[0017] In another aspect, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned collaborative management method for a high-computing-power terminal of an energy storage power station.

[0018] This application provides a collaborative management method for high-computing power terminals in energy storage power stations, comprising: physically integrating processor computing units and storage units to establish a high-computing architecture and building a high-computing power terminal for the energy storage power station; building an algorithm model on the high-computing architecture and achieving collaboration between the algorithm model and the high-computing architecture through optimization; constructing a terminal intelligent management system based on the high-computing terminal to centrally manage distributed high-computing terminals, algorithm models, and data, achieving end-edge-cloud collaboration; the integrated storage and computing high-computing architecture brings processor computing units and storage units very close together, and combined with optimized CPU scheduling strategies, effectively overcomes memory and power consumption barriers, reducing data transmission latency. The system significantly improves computing efficiency and energy consumption, solving the problems of slow speed and high resource consumption in complex analysis of millions of data points in energy storage power stations. The collaborative optimization of energy storage algorithms and high-computing-power architecture, through adapted algorithm models and data transmission design, enables the algorithm to fully utilize the advantages of the high-computing-power architecture, improving computational acceleration performance while ensuring data accuracy, further enhancing the efficiency and reliability of data processing. The intelligent management system for energy storage power station terminals achieves centralized management of distributed high-computing-power terminals, algorithm models, and data. Through an end-edge-cloud collaborative mechanism, it improves the intelligent management level of energy storage power stations, ensuring the safety and stability of power station operation.

[0019] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 The following is a flowchart illustrating the implementation of a collaborative management method for a high-computing-power terminal in an energy storage power station provided in this application; Figure 2 A structural flowchart of the optimal embodiment of the collaborative management method for a high-computing-power terminal of an energy storage power station provided in this application is shown; Figure 3 This application provides a system configuration diagram of a collaborative management device for a high-computing-power terminal of an energy storage power station. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] Example 1 This application provides a collaborative management method for high-computing-power terminals in energy storage power stations, such as... Figure 1 ,include: Physically integrate processor computing units and storage units to establish a high-computing-power architecture and build a high-computing-power terminal for energy storage power stations; Build algorithm models on high-computing-power architectures and achieve synergy between the algorithm models and the high-computing-power architectures by optimizing the algorithm models. A terminal intelligent management system is built based on high-computing-power terminals to centrally manage distributed high-computing-power terminals, algorithm models, and data, thereby achieving end-edge-cloud collaboration.

[0024] Optionally, the physical integration of the processor computing unit and storage unit to establish a high-performance computing architecture includes: By bringing the processor computing unit and storage unit closer together, a highly integrated module is formed, establishing a high-computing-power architecture.

[0025] This research aims to overcome the memory wall and power wall problems caused by the traditional separation of processor and storage units. It investigates strategies for CPU scheduling of storage units, optimizing scheduling logic to reduce data transmission latency and energy consumption between computing and storage units, thereby improving computational efficiency.

[0026] Optionally, the step of establishing a high-computing-power terminal for an energy storage power station includes: Based on the application scenarios and needs of high-computing-power terminals at the edge of energy storage power stations, select suitable CPUs. Clarify that CPU selection should consider key factors such as performance, clock speed, functionality, technical compatibility, and scalability. Based on these key factors, establish a comprehensive evaluation system, propose a scientific CPU selection method, and select suitable CPU processors. Based on the application scenarios of energy storage power station terminals, and according to physical, electromagnetic and electrical characteristics, design the power supply circuit, functional circuit, peripheral circuit, communication interface and communication protocol of the circuit board. By integrating the CPU and high-performance computing architecture onto the circuit board and employing a multi-fin fanless heat dissipation structure, a high-performance computing terminal for energy storage power stations can be established.

[0027] Optionally, the step of building an algorithm model on a high-computing-power architecture and achieving synergy between the algorithm model and the high-computing-power architecture through optimization includes: Establish rules for high-computing terminals to acquire data from the in-cabin BMS system, clarify the frequency, format and verification method of data acquisition, and design data transmission rules between high-computing terminals and terminal intelligent management systems of energy storage power stations to ensure the timeliness and consistency of data. Based on the characteristics of high-performance computing architecture, while ensuring the integrity and accuracy of the model, a general framework is adopted to build the algorithm model, unify the expression format of multi-framework algorithm models, make the algorithm model compatible and adaptable with high-performance computing architecture, and make full use of the computing advantages of the architecture. The algorithm optimizes the data layout in the algorithm model by using data partitioning and storing data in the order of computation, thereby reducing data access latency and conflicts. By optimizing operators and compressing the model, the model size and computational redundancy are reduced. While improving computational acceleration performance, error compensation and other techniques are used to reduce data loss and accuracy loss, and to achieve synergy with high-computing-power architecture. The terminal intelligent management system is built on high-computing-power terminals. The data transmission rules between the high-computing-power terminals and the terminal intelligent management system include the triggering conditions, transmission protocols, and data synchronization mechanisms for data transmission.

[0028] Optionally, the terminal intelligent management system includes: The terminal management module is used to enable the installation, binding, instance creation, deletion, modification, and querying of distributed high-computing-power terminals through the terminal intelligent management module. The subscription management module is used to support users' custom terminal data subscription needs; The operation log module is used to record the activity history of the energy storage power station and the operation and change reporting information of the high computing power terminal of the energy storage power station, so as to realize efficient management of the high computing power terminal; The algorithm model management module is used to develop algorithm model management functions, and to manage the activation, termination, iteration and batch distribution of various algorithm models to ensure the effective operation and timely update of algorithm models on various high-computing terminals. The data management module is used to design data management rules and develop modules such as power monitoring, SOC monitoring, intelligent early warning, and alarm center. It collects and monitors data such as the operating status, health status, degradation trend, and equipment failure status of each energy storage battery compartment, battery pack, and battery cell in real time, providing data support for the safe operation of the power station.

[0029] The edge-cloud collaboration module manages the communication services between the BMS battery management system and high-computing terminals, ensuring stable data transmission from end devices, enabling interconnection between high-computing terminals in each compartment, supporting data sharing and collaborative computing between terminals, establishing an encrypted interconnection mechanism between distributed high-computing terminals and the terminal intelligent management system, and transmitting analysis results data to the central end after encryption and isolation processing, thereby realizing early warning monitoring of edge devices and collaborative linkage of edge-cloud data.

[0030] Example 2 Based on the same inventive concept, this application also provides a preferred embodiment of a collaborative management method for high-computing-power terminals in energy storage power stations, such as... Figure 2 ,include: 1. Development and implementation of in-memory computing high-performance architecture In high-performance computing architectures, 3D integration technology is used to physically integrate processor computing units and storage units, keeping the distance between them within 10. Within this range. For edge computing scenarios in battery energy storage power stations, analysis of data processing volume and computational task types clarifies that the CPU must support multi-threaded parallel computing, have a clock speed of at least 2.5GHz, and possess good PCIe scalability. Based on this, a specific ARM architecture CPU was selected, whose performance meets the computing requirements, has low power consumption, and is suitable for energy storage scenarios.

[0031] The circuit board design incorporates a wide voltage input design, supporting 12-24V DC input to ensure stable operation in the complex power supply environment of energy storage power stations. Communication interfaces include Ethernet and RS485 to meet the communication requirements of BMS and management systems. During integration, a multi-fin aluminum heat dissipation structure with a 2mm fin spacing is used, achieving heat dissipation through natural convection to ensure the terminal operates within -20°C. Up to 60 It operates normally in the environment.

[0032] 2. Collaborative Optimization Implementation of Energy Storage Algorithms and High-Computing-Power Architecture For data transmission, the high-performance computing terminal is set to acquire data from the BMS system at a frequency of 100ms / time. The data format is JSON, containing key parameters such as battery voltage, current, and temperature, and CRC checksums are used to ensure data integrity. The high-performance computing terminal and the terminal intelligent management system use the MQTT protocol for data transmission. When the terminal detects abnormal data, transmission is triggered immediately. Under normal circumstances, data is transmitted every 5 seconds to summarize the data.

[0033] In algorithm model development, a safety early warning algorithm model was built based on the TensorFlow Lite framework. The original battery fault diagnosis algorithm models under the Caffe and PyTorch frameworks were converted to the unified TensorFlow Lite format to achieve compatibility with high-performance computing architectures. During model optimization, the data was processed in blocks, with the size of each block matching the CPU cache capacity. Model quantization technology was used to convert 32-bit floating-point weights to 16-bit, reducing the model size. Simultaneously, precision loss was compensated for by fine-tuning model parameters, maintaining model accuracy above 95%.

[0034] 3. Implementation of intelligent management system for energy storage power station terminals The terminal management module adopts a B / S architecture, allowing administrators to install and bind terminals via a web interface. During binding, the module automatically retrieves the terminal's hardware information and network address. The operation log module summarizes and stores terminal operation data hourly for at least one year. The algorithm model management module supports uploading new algorithm models via an API interface and can distribute them in batches to specified terminals. Models are validated during distribution to ensure their integrity.

[0035] The data management module collects data from the battery compartment in real time, calculates the battery charging and discharging power through the power monitoring module, and estimates the battery's state of charge using a Kalman filter algorithm through the SOC monitoring module. The intelligent early warning module analyzes the results based on the algorithm model and, when an anomaly is detected, alerts management personnel via sound, light, and SMS through the alarm center. Regarding end-edge-cloud collaboration, the BMS system communicates with terminals using the Modbus protocol, terminals are interconnected via a local area network, and terminals communicate with the management system via encrypted Ethernet. Data transmission uses the AES256 encryption algorithm to ensure data security.

[0036] Example 3 Based on the same inventive concept, this application also provides a collaborative management device for a high-computing-power terminal of an energy storage power station, such as... Figure 3 ,include: High-performance computing architecture unit is used to physically integrate processor computing units and storage units to establish a high-performance computing architecture and build a high-performance computing terminal for energy storage power stations. The algorithm collaboration unit is used to build algorithm models on high-computing-power architectures and achieve collaboration between the algorithm models and the high-computing-power architectures by optimizing the algorithm models. The edge-cloud collaboration unit is used to build a terminal intelligent management system based on high-computing-power terminals, and to centrally manage distributed high-computing-power terminals, algorithm models and data to achieve edge-cloud collaboration.

[0037] Optionally, the physical integration of the processor computing unit and storage unit to establish a high-performance computing architecture includes: By bringing the processor computing unit and storage unit closer together, a highly integrated module is formed, establishing a high-computing-power architecture.

[0038] Optionally, the step of establishing a high-computing-power terminal for an energy storage power station includes: Select a suitable CPU based on the application scenarios and requirements of the high-computing-power terminal at the edge of the energy storage power station, and design the power supply circuit, functional circuit, peripheral circuit, communication interface and communication protocol of the circuit board, taking into account physical characteristics, electromagnetic characteristics and electrical characteristics. By integrating the CPU and high-performance computing architecture onto the circuit board and employing a multi-fin fanless heat dissipation structure, a high-performance computing terminal for energy storage power stations can be established.

[0039] Optionally, the step of building an algorithm model on a high-computing-power architecture and achieving synergy between the algorithm model and the high-computing-power architecture through optimization includes: Establish rules for high-computing-power terminals to acquire data from the in-cabin BMS system, clarify the frequency, format and verification method of data acquisition, and design data transmission rules between high-computing-power terminals and terminal intelligent management systems of energy storage power stations; Based on the characteristics of high-computing-power architecture, a general framework is used to build the algorithm model and unify the expression format of multi-framework algorithm models; The algorithm model optimizes the data layout by using data partitioning and storing data in the order of computation. It reduces the model size and computational redundancy by optimizing operators and compressing the model. It also reduces data loss and accuracy loss by using error compensation and other techniques, and achieves collaboration with high-computing architecture. The terminal intelligent management system is built on high-computing-power terminals. The data transmission rules between the high-computing-power terminals and the terminal intelligent management system include the triggering conditions, transmission protocols, and data synchronization mechanisms for data transmission.

[0040] Optionally, the terminal intelligent management system includes: The terminal management module is used to install and bind the terminal intelligent management module to distributed high-computing-power terminals; The subscription management module is used to support users' custom terminal data subscription needs; The operation log module is used to record the activity history of the energy storage power station and the operation and change reporting information of the high computing power terminal of the energy storage power station; The algorithm model management module is used to manage the activation, termination, iteration, and batch distribution of various algorithm models. The data management module is used to design data management rules and collect and monitor data from energy storage power stations in real time.

[0041] The edge-cloud collaboration module is used to manage the communication services between the BMS system and high-computing-power terminals, realize the interconnection between high-computing-power terminals in each cabin, establish an encrypted interconnection mechanism between distributed high-computing-power terminals and the terminal intelligent management system, and realize edge-edge device monitoring and edge-cloud data collaboration.

[0042] Based on the above disclosure, the present invention also provides an electronic device. The electronic device of this embodiment includes at least one processor and at least one storage medium electrically connected to each other. The storage medium is electrically connected to the processor, wherein the storage medium stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described above.

[0043] Based on the same inventive concept, the present invention also provides a storage medium storing instructions executable by at least one processor, the instructions being executed by at least one processor to enable at least one processor to perform the method described above.

[0044] Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A collaborative management method for high-computing-power terminals in an energy storage power station, characterized in that, include: Physically integrate processor computing units and storage units to establish a high-computing-power architecture and build a high-computing-power terminal for energy storage power stations; Build algorithm models on high-computing-power architectures and achieve synergy between the algorithm models and the high-computing-power architectures by optimizing the algorithm models. A terminal intelligent management system is built based on high-computing-power terminals to centrally manage distributed high-computing-power terminals, algorithm models, and data, thereby achieving end-edge-cloud collaboration.

2. The method as described in claim 1, characterized in that, The physical integration of processor computing units and storage units to establish a high-performance computing architecture includes: By bringing the processor computing unit and storage unit closer together, a highly integrated module is formed, establishing a high-computing-power architecture.

3. The method as described in claim 2, characterized in that, The aforementioned includes establishing a high-computing-power terminal for energy storage power stations, including: Select a suitable CPU based on the application scenarios and requirements of the high-computing-power terminal at the edge of the energy storage power station, and design the power supply circuit, functional circuit, peripheral circuit, communication interface and communication protocol of the circuit board based on its physical, electromagnetic and electrical characteristics. By integrating the CPU and high-performance computing architecture onto the circuit board and employing a multi-fin fanless heat dissipation structure, a high-performance computing terminal for energy storage power stations can be established.

4. The method as described in claim 1, characterized in that, The process of building an algorithm model on a high-computing-power architecture and achieving synergy between the algorithm model and the high-computing-power architecture through optimization includes: Establish rules for high-computing-power terminals to acquire data from the in-cabin BMS system, clarify the frequency, format and verification method of data acquisition, and design data transmission rules between high-computing-power terminals and terminal intelligent management systems of energy storage power stations; Based on the characteristics of high-computing-power architecture, a general framework is used to build the algorithm model and unify the expression format of multi-framework algorithm models; The algorithm model optimizes the data layout by using data partitioning and storing data in the order of computation. It reduces the model size and computational redundancy by optimizing operators and compressing the model. It also reduces data loss and accuracy loss by using error compensation and other techniques, and achieves collaboration with high-computing architecture. The terminal intelligent management system is built on high-computing-power terminals. The data transmission rules between the high-computing-power terminals and the terminal intelligent management system include the triggering conditions, transmission protocols, and data synchronization mechanisms for data transmission.

5. A method as described in claim 4, characterized in that, The terminal intelligent management system includes: The terminal management module is used to install and bind the terminal intelligent management module to distributed high-computing-power terminals; The subscription management module is used to support users' custom terminal data subscription needs; The operation log module is used to record the activity history of the energy storage power station and the operation and change reporting information of the high computing power terminal of the energy storage power station; The algorithm model management module is used to manage the activation, termination, iteration, and batch distribution of various algorithm models. The data management module is used to design data management rules and collect and monitor data from energy storage power stations in real time. The edge-cloud collaboration module is used to manage the communication services between the BMS system and high-computing-power terminals, realize the interconnection between high-computing-power terminals in each cabin, establish an encrypted interconnection mechanism between distributed high-computing-power terminals and the terminal intelligent management system, and realize edge-edge device monitoring and edge-cloud data collaboration.

6. A collaborative management device for a high-computing-power terminal of an energy storage power station, characterized in that, include: High-performance computing architecture unit is used to physically integrate processor computing units and storage units to establish a high-performance computing architecture and build a high-performance computing terminal for energy storage power stations. The algorithm collaboration unit is used to build algorithm models on high-computing-power architectures and achieve collaboration between the algorithm models and the high-computing-power architectures by optimizing the algorithm models. The edge-cloud collaboration unit is used to build a terminal intelligent management system based on high-computing-power terminals, and to centrally manage distributed high-computing-power terminals, algorithm models and data to achieve edge-cloud collaboration.

7. The apparatus as described in claim 6, characterized in that, The physical integration of processor computing units and storage units to establish a high-performance computing architecture includes: By bringing the processor computing unit and storage unit closer together, a highly integrated module is formed, establishing a high-computing-power architecture.

8. The apparatus as described in claim 7, characterized in that, The aforementioned includes establishing a high-computing-power terminal for energy storage power stations, including: Select a suitable CPU based on the application scenarios and requirements of the high-computing-power terminal at the edge of the energy storage power station, and design the power supply circuit, functional circuit, peripheral circuit, communication interface and communication protocol of the circuit board, taking into account physical characteristics, electromagnetic characteristics and electrical characteristics. By integrating the CPU and high-performance computing architecture onto the circuit board and employing a multi-fin fanless heat dissipation structure, a high-performance computing terminal for energy storage power stations can be established.

9. The apparatus as described in claim 6, characterized in that, The process of building an algorithm model on a high-computing-power architecture and achieving synergy between the algorithm model and the high-computing-power architecture through optimization includes: Establish rules for high-computing-power terminals to acquire data from the in-cabin BMS system, clarify the frequency, format and verification method of data acquisition, and design data transmission rules between high-computing-power terminals and terminal intelligent management systems of energy storage power stations; Based on the characteristics of high-computing-power architecture, a general framework is used to build the algorithm model and unify the expression format of multi-framework algorithm models; The algorithm model optimizes the data layout by using data partitioning and storing data in the order of computation. It reduces the model size and computational redundancy by optimizing operators and compressing the model. It also reduces data loss and accuracy loss by using error compensation and other techniques, and achieves collaboration with high-computing architecture. The terminal intelligent management system is built on high-computing-power terminals. The data transmission rules between the high-computing-power terminals and the terminal intelligent management system include the triggering conditions, transmission protocols, and data synchronization mechanisms for data transmission.

10. The apparatus as described in claim 9, characterized in that, The terminal intelligent management system includes: The terminal management module is used to install and bind the terminal intelligent management module to distributed high-computing-power terminals; The subscription management module is used to support users' custom terminal data subscription needs; The operation log module is used to record the activity history of the energy storage power station and the operation and change reporting information of the high computing power terminal of the energy storage power station; The algorithm model management module is used to manage the activation, termination, iteration, and batch distribution of various algorithm models. The data management module is used to design data management rules and collect and monitor data from energy storage power stations in real time. The edge-cloud collaboration module is used to manage the communication services between the BMS system and high-computing-power terminals, realize the interconnection between high-computing-power terminals in each cabin, establish an encrypted interconnection mechanism between distributed high-computing-power terminals and the terminal intelligent management system, and realize edge-edge device monitoring and edge-cloud data collaboration.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the collaborative management method for a high-computing-power terminal of an energy storage power station as described in any one of claims 1-5.

12. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in a memory, it implements the steps of the collaborative management method for a high-computing-power terminal of an energy storage power station as described in any one of claims 1-5.

Images