Abstract: This paper develops a nickel-metal hydride battery management system for hybrid electric vehicles. The system adopts a distributed communication structure based on a CAN network with multiple nodes. Each node module performs functions such as parameter acquisition, analysis and processing, and monitoring and diagnosis, making the management system more advanced, more accurate, and improving its reliability and scalability. Keywords: CAN bus, battery management system, residual energy 1 Introduction Accurate measurement of the remaining capacity of a battery has always been a critical issue in the development of electric vehicles. An effective battery management system is beneficial to improving battery life. Therefore, accurate estimation of the battery's SOC has become the central issue of electric vehicle battery energy management systems. If the battery's SOC can be correctly estimated, the electrical energy provided by the battery can be rationally utilized, extending the battery pack's lifespan. The scheme adopts a bus-based networking method, using a fieldbus to complete data exchange between nodes. In the distributed solution, the multi-energy controller is the master ECU, which communicates with multiple lower-level ECUs via a fieldbus. During operation, the communication submodule of each controller runs in the background using timers or interrupts to complete data transmission and reception, saving resources in the main process. As shown in Figure 1, the battery's SOC value is sent to the multi-energy controller by the battery controller via the CAN bus, while the vehicle's operating mode is determined by the multi-energy controller through a specific logical algorithm based on information collected from each ECU. Once these parameters are determined, we can decide whether to start or stop the engine, and also determine the motor's operating state. For example, when the battery's SOC value is between 50% and 70%, the multi-energy controller calculates the vehicle's operating mode as start-up mode, indicating sufficient electrical energy in the system, no need to start the engine, and the motor can operate in drive mode. 2. System Hardware Composition As shown in Figure 2, the battery controller can communicate with other control systems in the external vehicle via the CAN bus network. It consists of one battery management ECU (electronic control unit) and four battery pack information detection ECUs; the individual batteries used are combined into 24 battery packs. We configure one measurement unit for every six battery packs, i.e., battery pack ECU1 to ECU4. Four battery pack ECUs and the battery stack ECU form a CAN bus network. One CAN controller and the battery pack ECUs form the internal CAN network of the battery management system. Another CAN controller and other control systems in the vehicle form a whole-vehicle fiber optic CAN bus network. Figure 2 shows the structural block diagram of the battery management ECU, and Figure 3 shows the structure. The embedded microcontroller used in the battery pack ECU is the P87C591 microcontroller, which integrates a CAN controller and an A/D converter module. Each battery pack ECU manages six battery packs, measuring the voltage and temperature information of the six battery packs and sending the collected information to the battery management ECU via the CAN bus. The voltage of each of the six battery packs is connected to the six A/D input ports of the P87C591 after passing through a voltage conditioning circuit. The signal lines of the six temperature sensors are connected to the same I/O port of the P87C591. Figure 3 shows the circuit structure of the battery pack ECU. Figure 3 shows the circuit design of the CAN interface. In this design, the P87C591 is used as the microcontroller. The interface circuit design between the P87C591 and the CAN driver chip is shown in Figure 4. It mainly consists of three parts: the P87C591, an opto-isolation circuit, and the CAN driver. Opto-isolation circuit: To further suppress interference, an opto-isolation circuit is often used in the CAN bus interface. The opto-isolation circuit is generally located between the CAN controller and the transceiver. Figure 4 shows the hardware design circuit diagram of the CAN communication module. The overall system program includes an initialization program and a main loop program, and its flowchart is shown in Figure 5: The system first powers on, then initializes the CAN and timer. The system waits for an interrupt. If there is an interrupt, the interrupt type is determined. If it is an interrupt from the SJA1000 controller, the data from the SJA1000 controller is read, and the buffer is released. After the interrupt is completed, the system returns. If it is a 50ms timer period interrupt, the voltage and current data are converted to analog (AD) values, the SOC value is calculated, and the relevant data is sent by the CAN bus. After the interrupt is completed, the system returns to the main function `main()`. Figure 5 shows the main program. Figure 4 concludes: CAN bus-based data communication technology has high reliability, real-time performance, and flexibility. CAN bus has broad application prospects and development space in the application of nickel-metal hydride battery management systems for hybrid electric vehicles. References [1] Zou Kuanming. CAN bus principle and application system design. Beijing University of Aeronautics and Astronautics Press, 1997 [2] Hu Minghui, Qin Datong, Shu Hong. Evaluation of battery management system SOC for hybrid electric vehicles. Journal of Chongqing University: April 2003, Vol. 26, No. 4. [3] Wang Yifeng, Li Lingqi. Step-by-step data acquisition and control system based on CAN bus. Industrial Control Computer: 2000.5 Liu Qian, female, (1978.6.30-) Han nationality, native of Wuhan, Huazhong Agricultural University, College of Engineering and Technology, 430070, Lecturer, Master, research direction: communication control, information system Xiong Lirong, female Han nationality, native of Wuhan, Huazhong Agricultural University, College of Engineering and Technology, 430070, Lecturer, Master