A novel battery remote management system based on neural network

Abstract

The application relates to the technical field of battery management, in particular to a novel battery remote management system based on a neural network, which comprises an information acquisition module, a battery marking module and an AI data analysis module; the information acquisition module feeds back collected battery data information to a system cloud end; the system cloud end stores and arranges the information; meanwhile, the system cloud end stores the data information of the rechargeable battery into a database; the AI data analysis module processes the data information in the database by using the algorithm of a machine learning neural network; and the battery marking module classifies and marks the batteries to provide a learning benchmark for the machine learning of the AI data analysis module. The application integrates the charging and discharging of the batteries and specific application devices together by using the technology of the Internet of Things, collects and statistically filters the cloud end data of the Internet of Things, analyzes and manages the data information of the rechargeable battery by using the algorithm of the machine learning neural network of AI.

Description

Technical Field

[0001] This invention relates to the field of battery management technology, specifically to a novel remote battery management system based on neural networks. Background Technology

[0002] Currently, there are many types of rechargeable batteries, and their charging, discharging, and remaining charge / discharging status are statistically based on charge integration. The power of each battery is realized in the BMS (battery management system) circuit system. Generally, rechargeable batteries are composed of multiple individual batteries, which are connected in series and parallel to achieve the application scenarios we need. However, the management of rechargeable batteries in the current technology is costly and has poor system stability (BMS is relatively expensive). Sometimes it may be necessary to remove the battery and send it to a specialized institution for testing and repair, which is inconvenient. Summary of the Invention

[0003] The purpose of this invention is to provide a novel remote battery management system based on neural networks, which can remotely collect and manage data information of rechargeable batteries, thereby solving the shortcomings and unmet technical requirements of existing technologies.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a novel remote battery management system based on neural networks, comprising an information acquisition module, a battery marking module, and an AI data analysis module. The information acquisition module feeds back the acquired battery data to the system cloud, where the system cloud organizes and stores the information, and simultaneously stores the rechargeable battery data in a database. The AI ​​data analysis module uses machine learning neural network algorithms to process the data in the database. The battery marking module classifies and marks the batteries to provide a learning benchmark for the machine learning of the AI ​​data analysis module. This application remotely acquires rechargeable battery data through the information acquisition module, classifies and marks the rechargeable batteries through the battery marking module, stores the rechargeable battery data and classification and marking information in the database, and processes the information in the database using machine learning neural network algorithms.

[0005] Preferably, the information acquisition module collects information through Internet of Things (IoT) technology. The information acquisition module includes a remote data acquisition IoT control module, which includes a controller and a charging / discharging device. The information acquisition module mainly collects the voltage, current, and battery information of the charging / discharging device. This application collects the performance data of the discharging device to achieve a comprehensive understanding of the data information of each discharging device.

[0006] Preferably, the battery marking module includes an MCU and a charging and discharging equipment circuit. The battery marking module classifies and marks the batteries according to the type of rechargeable battery and the discharge period of the rechargeable battery. The rechargeable batteries can be classified into ternary lithium batteries, lithium iron phosphate batteries, lead-acid batteries, cobalt acid batteries, gel batteries, aluminum vanadium batteries, nickel-metal hydride batteries, etc.

[0007] The AI ​​data analysis module described in this application uses a machine learning CNN neural network algorithm, or the AI ​​data analysis module uses a machine learning SVM neural network algorithm, to perform calculations and analysis on the rechargeable battery based on the rechargeable battery information in the database.

[0008] Preferably, the controller and the charging / discharging device are connected to the same circuit. The controller is an MCU+FPGA controller, and the charging / discharging device includes a rechargeable battery, a load, and a power generation device.

[0009] Preferably, the MCU is connected to the charging and discharging device circuit, which is connected to a rechargeable battery, a load, and a power source. The MCU controls the power source and the load to mark and store the current, voltage, and power information in the circuit, and then stores the classification and marking information of the rechargeable battery in a database.

[0010] The AI ​​data analysis module described in this application uses a machine learning CNN neural network algorithm, or the AI ​​data analysis module uses a machine learning SVM neural network algorithm, to analyze the performance, type, and lifespan of rechargeable batteries during automatic use. Specifically, it uses a CNN algorithm to automatically match the types of rechargeable batteries, calculate the battery usage status and the remaining number of charging cycles, and performs statistical analysis on other information.

[0011] Preferably, the controller is equipped with a G communication module to enable remote acquisition of battery information.

[0012] Preferably, in addition to classifying and labeling rechargeable batteries according to their type, the MCU also classifies and labels rechargeable batteries according to their charge status for different charging methods. The MCU classifies and labels rechargeable batteries according to their charging methods as pre-charging, constant current charging, constant voltage charging, and micro-charging. This application matches different charging methods according to the real-time status of the battery.

[0013] Preferably, the information acquisition module can collect data from multiple rechargeable batteries simultaneously to perform regression analysis on the data from the same batch of rechargeable batteries.

[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: This application utilizes Internet of Things (IoT) technology to integrate battery charging and discharging with specific application devices. An information collection module collects data on the voltage / current / battery of charging and discharging devices, such as rechargeable batteries, power supplies, and loads, enabling edge data processing. Leveraging cloud-based data collection and statistical filtering via IoT, and employing AI machine learning neural network algorithms, it automatically matches battery types to rechargeable batteries, calculates battery usage status and remaining charging cycles, and performs regression analysis on data from the same batch of batteries. This application is characterized by low cost and high efficiency, while simultaneously enabling more comprehensive information collection and more detailed data management, thus achieving remote battery management. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the system of the present invention;

[0016] Figure 2 This is a schematic diagram of the Internet of Things control module in this new type of device;

[0017] Figure 3 This is a schematic diagram of the battery marking principle in this invention;

[0018] In the diagram: 1. Information acquisition module; 2. Battery marking module; 3. AI data analysis module; 4. Database; 5. Remote data acquisition IoT control module; 6. Controller; 7. Rechargeable device; 8. MCU; 9. Rechargeable device circuit; 10. MCU+FPGA controller; 11. Rechargeable battery; 12. Load; 13. Power generation equipment; 14. Power supply; 15. 4G communication module. Detailed Implementation

[0019] The following will refer to the appendices in the embodiments of the present invention. Figure 1-3 The technical solutions in the embodiments of the present invention are clearly and completely described herein. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1-3 Embodiments of the present invention

[0021] like Figure 1As shown: A novel remote battery management system based on neural networks includes an information acquisition module 1, a battery marking module 2, and an AI data analysis module 3. The information acquisition module 1 feeds back the collected battery data to the system cloud, where the information is organized and stored. Simultaneously, the system cloud stores the rechargeable battery data 11 in a database 4. The AI ​​data analysis module 3 uses machine learning neural network algorithms to process the data in the database 4. The battery marking module 2 classifies and marks the batteries to provide a learning benchmark for the machine learning of the AI ​​data analysis module 3. This application utilizes the AI ​​data analysis module 3 to learn neural network algorithms for accurate and comprehensive analysis and processing of data.

[0022] like Figure 2 As shown: The information acquisition module 1 collects information through Internet of Things (IoT) technology. The information acquisition module 1 includes a remote data acquisition IoT control module 5, which includes a controller 6 and a charging / discharging device 8. The information acquisition module 1 mainly collects the voltage, current, and battery information of the charging / discharging device 8. This application utilizes the information acquisition module 1 to comprehensively collect information from the charging / discharging device and uses the remote data acquisition IoT control module 5 to remotely transmit the information, providing a data source for subsequent data analysis and processing.

[0023] like Figure 3 As shown: The battery marking module 2 includes an MCU8 and a charging and discharging device circuit 9. The battery marking module 2 classifies and marks the batteries according to the type of rechargeable battery 11 and the discharge period of the rechargeable battery 11. In this application, the rechargeable battery 11 can be divided into ternary lithium batteries, lithium iron phosphate batteries, lead-acid batteries, cobalt acid batteries, gel batteries, aluminum vanadium batteries, nickel-metal hydride batteries, etc. The battery marking module 2 is used to further classify and mark the rechargeable battery 11, providing a benchmark for subsequent data information processing.

[0024] like Figure 1 As shown: The AI ​​data analysis module 3 uses the machine learning CNN neural network algorithm to perform calculations and analysis on the rechargeable battery 11 based on the information of the rechargeable battery 11 in the database 4. This application uses the AI ​​data analysis module 3 to learn the CNN neural algorithm to accurately analyze and process the data stored in the database 4.

[0025] like Figure 2 As shown: The controller 6 and the charging and discharging device 8 are connected to the same circuit. The controller 6 adopts an MCU+FPGA controller 10. The charging and discharging device 8 includes a rechargeable battery 11, a load 12 and a power generation device 13. The scope of information collected in this application includes all rechargeable and discharging devices 7 in the circuit that can affect the rechargeable battery 11, and comprehensive data collection is carried out.

[0026] like Figure 2 As shown: The MCU8 is connected to the charging and discharging device circuit 9, which is connected to a rechargeable battery 11, a load 12, and a power supply 14. The MCU8 controls the power supply 14 and the load 12 to mark and store the current, voltage, and power information in the circuit. Then, the classification and marking information of the rechargeable battery 11 is stored in the database 4. This application stores various information of the various rechargeable and discharging devices 7 collected in the cloud database 4. The database 4 performs statistics and filtering on the input data information.

[0027] In this application, the AI ​​data analysis module 3 uses a machine learning CNN neural network algorithm to analyze the performance, type, and lifespan of the rechargeable battery 11 during automatic use. Specifically, this application uses the CNN algorithm to automatically match the battery types of the rechargeable battery 11, calculate the battery usage status and the remaining number of charging cycles of the rechargeable battery 11, and at the same time perform statistical analysis on other information, such as battery voltage, battery current, battery charging power, load power, and ambient temperature.

[0028] like Figure 2 As shown: The controller 6 is equipped with a 4G communication module 15 to realize remote collection of battery information.

[0029] like Figure 3 As shown: In addition to classifying and labeling the rechargeable battery 11 according to its type, the MCU8 also classifies and labels the rechargeable battery 11 according to its power status for different charging methods. The MCU8 classifies and labels the rechargeable battery 11 for charging methods as pre-charging, constant current charging, constant voltage charging, and micro-charging. When the battery voltage is relatively low, it corresponds to the pre-charging charging method. When fast charging is required, constant current charging is used. When normal charging is required, constant voltage charging is used. Finally, micro-charging is used for charging.

[0030] In this application, the information acquisition module 1 can simultaneously collect data information from multiple rechargeable batteries 11 to achieve regression analysis of data information from the same batch of rechargeable batteries 11. This application can also collect information from the circuits containing multiple rechargeable batteries 11.

[0031] Working principle: In this application, the information acquisition module 1 comprehensively collects various data information of the rechargeable and discharging device 7 in the circuit where the rechargeable battery 11 is located, and inputs the collected data information into the cloud database 4 through the Internet of Things. At the same time, the battery marking module 2 accurately and meticulously classifies the rechargeable battery 11, and inputs the marked battery information into the cloud database 4 through the Internet of Things. The database 4 filters and statistically analyzes the input data information. Then, the AI ​​data analysis module 3 in this application learns the neural network algorithm to process and analyze the data in the database 4, and generates an accurate and comprehensive data report to achieve remote management of the rechargeable battery 11.

[0032] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0033] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A novel battery remote management system based on neural networks, characterized in that, It includes an information acquisition module (1), a battery marking module (2) and an AI data analysis module (3). The information acquisition module (1) feeds back the collected battery data information to the system cloud. The system cloud organizes and stores the information. At the same time, the system cloud stores the rechargeable battery data information into the database (4). The AI ​​data analysis module (3) uses the algorithm of machine learning neural network to process the data information in the database (4). The battery marking module (2) classifies and marks the batteries to provide a learning benchmark for the machine learning of the AI ​​data analysis module (3). The battery marking module (2) includes an MCU (8) and a charging and discharging equipment circuit (9). The battery marking module (2) classifies and marks batteries according to the type of rechargeable battery and the discharge period of the rechargeable battery. Rechargeable batteries can be classified into ternary lithium batteries, lithium iron phosphate batteries, lead-acid batteries, cobalt acid batteries, gel batteries, aluminum vanadium batteries, and nickel-metal hydride batteries. The AI ​​data analysis module (3) uses the machine learning CNN neural network algorithm to calculate and analyze the rechargeable battery based on the rechargeable battery information in the database (4). Specifically, it uses the CNN algorithm to automatically match the battery types, calculate the battery usage status and the remaining number of recharges of the rechargeable battery, and at the same time performs statistical analysis on other information such as battery voltage, battery current, battery charging power, load power, and ambient temperature. Alternatively, the AI ​​data analysis module (3) uses the machine learning SVM neural network algorithm to calculate and analyze the rechargeable battery based on the rechargeable battery information in the database. Specifically, it uses the CNN algorithm to automatically match the battery types of the rechargeable battery, calculate the battery usage status and the remaining number of charging times of the rechargeable battery, and at the same time perform statistical analysis on other information, such as battery voltage, battery current, battery charging power, load power, and ambient temperature. In addition to classifying and labeling rechargeable batteries according to their type, the MCU (8) also classifies and labels rechargeable batteries according to their charge status for different charging methods. The MCU (8) classifies and labels rechargeable batteries according to their charging methods as pre-charging, constant current charging, constant voltage charging, and micro-charging.

2. The novel battery remote management system based on neural networks according to claim 1, characterized in that, The information acquisition module (1) collects information through Internet of Things (IoT) technology. The information acquisition module (1) includes a remote data acquisition IoT control module (5). The remote data acquisition IoT control module (5) includes a controller (6) and a charging and discharging device (7). The information acquisition module (1) mainly collects the voltage, current and battery information of the charging and discharging device (7).

3. The novel battery remote management system based on neural networks according to claim 2, characterized in that, The controller (6) is connected to the charging and discharging device (7) in the same circuit, and the controller (6) adopts an MCU+FPGA controller (10); The charging and discharging device (7) includes a rechargeable battery (11), a load (12), and a power generation device (13).

4. The novel battery remote management system based on neural networks according to claim 1, characterized in that, The MCU (8) is connected to the charging and discharging equipment circuit (9), which is connected to a rechargeable battery (11), a load (12) and a power supply (14). The MCU (8) marks and stores the current, voltage and power information in the circuit by controlling the power supply (14) and the load (12), and then stores the classification and marking information of the rechargeable battery in the database.

5. The novel battery remote management system based on neural networks according to claim 1, characterized in that, The AI ​​data analysis module (3) uses the machine learning CNN neural network algorithm to analyze the performance, type and lifespan of the rechargeable battery in automatic use; or the AI ​​data analysis module (3) uses the machine learning SVM neural network algorithm to analyze the performance, type and lifespan of the rechargeable battery in automatic use.

6. The novel battery remote management system based on neural networks according to claim 3, characterized in that, The controller (6) is equipped with a 4G communication module (15) to enable remote collection of battery information.

7. The novel battery remote management system based on neural networks according to claim 1, characterized in that, The information acquisition module (1) can collect data information from multiple rechargeable batteries at the same time, so as to realize regression analysis of data information of the same batch of rechargeable batteries.

Images