| The operating temperature of the battery has a major effect on property and safety of EV.The working environment of EV are extremely intricacy,leading to complex thermal characteristics of battery.Therefore,in the development of electric vehicles,the development of thermal management system for power batteries is extremely important.The main work of this paper is to design a new energy vehicle battery liquid cooling test bench,develop thermal management test and analysis software based on LabVIEW,and a controller based on STM32 according to the needs of the bench,and rely on KULI builds a control strategy based on intelligent algorithms,laying the foundation for subsequent experimental research on the thermal management system.First,combined with the new energy vehicles on the market,the typical structure of the indirect liquid cooling system was analyzed,and the experimental bench structure design of the battery thermal management system was completed based on this,and it was decided to develop thermal management test and analysis software and thermal management controller for new energy vehicles.Based on this,the selection of main components and test instruments was completed,and the communication protocol of the system was selected.After the completion of battery liquid-cooling testbed design,the functions of the EV thermal management test and analysis software are sorted out,and the overall structure of the software is designed,and the software is divided into five major modules.Based on the Modbus communication protocol,a control communication module for communicating with the controller was developed,and a data acquisition module for acquiring experimental data was developed based on NI-DAQ.After analyzing the test data characteristics of the thermal management system,the median and mean filtering algorithms were selected to develop the data processing module,and the support vector regression machine(SVR)and the BP neural network intelligent algorithm were selected to develop the data analysis module.According to the needs of each module to read and write data,a data management module was developed.Finally,adhering to the artistic and easy-to-use design philosophy,the software’s UI was designed.In order to meet the control needs of the thermal management system,STM32F103VET6 was chosen as the MCU of the liquid-cooling controller,andμC/OS-Ⅱ was used as the OS of the liquid-cooling controller.After sorting out the software and hardware functions that the controller needs to implement,the software and hardware structure of liquid-cooling controller was designed,and the circuit design of optocoupler,drive,temperature measurement,power,communication and other modules are finished.The software is segmented into three modules including temperature calculation,control and communication.A test was carried out with the test and analysis software.The test result proves that the controller can adjust the output signal based on the temperature signal,the communication of liquid-cooling system is normal,and the control commands and collected data is realizedFinally,based on the designed liquid-cooling testbed,a thermal management model was established through KULI and the model verification was completed through the test data of the bench.Combined with the thermal management model,the cooling performance of the thermal management system under different fan speeds and compressor speeds are analyzed.On this basis,the input model of various working conditions is used for simulation calculation,and the speed that can meet the heat dissipation requirements is obtained.Using this data as the sample space,the fan speed and compressor speed prediction models were trained using the SVR intelligent algorithm,and then the prediction performance of the model was verified through the test set.Finally,simplify the speed prediction model,calculate the input-output relationship table,draw the control map based on the table,and write it into the control module of the controller. |
