Lithium-ion Battery Into The Design And Realization Of The Detection System

As the Li-ion battery has many advantages, such as high working-voltage, large -capacity and long circle-life, the demand of Li-ion electric vehicle battery have greatly increased at the speed of 30% each year. At the same time, industrial capacity and high quality battery testing system are demanded by the rapid development of Li-ion electric vehicle battery Based on comprehensive review application actuality of battery formation and testing equipments, a comprehensive system which concluding formation and testing functions was designed and realized based on distributed management. The main research content and results in the paper are as follows:(1) According to the needs of application, an overall planning on the system was carried out, the structures of 3—layer distributed system was designed .The main devices of the system were chosen carefully on the consideration of detection speed, accuracy and high anti-interference performance.(2) The designing and debugging on hardware circuit of formation station and testing station were accomplished. The formation station designed in this paper has 80 cell positions, each of which can be operated separately and in which case the time consumption of data dictation circle is 1.5s with the solution of 12 bits. The testing station with 8 main controllers was designed in a distributed structure, which has wider current adjustment range and faster detection speed. And the data gathered by them is sent to the upper computer by CAN bus through the CAN interface.(3) The development and debugging on the application software for the system was accomplished. The codes for power system, data acquisition and filtering and the driving procedures for CAN interface were designed. The codes for the whole logical of working stations were also accomplished in the paper.(4) Aiming at reducing the impact of the data transmission distortion, a 3—layer BP ANN was designed according to the characteristic of this system. The samples of current and voltage regarded as the network training data were filtered and then a forcast is given to them by the trained ANN on the different occasions.All the results show that the network is easy to realize, the forecast data of which is high in accuracy, and thus is better in reducing the distortion of the sampling data.(5) A test on the system was carried out and the results manifested that the system designed in this paper have many merits such as a high dictation accuracy, reliable and steady data transmission, high resist interference performance, intelligential and lower cost. The system can serve well in massive battery production areas.