Study on Online Battery Monitoring Method Based on Multilayer Support Vector Machine
The online monitoring nodes for batteries are conventionally deployed in independent or regional forms which have limited monitoring ranges,resulting in a decrease in the obtained average online monitoring frequency.Therefore a design and validation study of a multi-layer support vector machine based online battery monitoring method is proposed.Based on the current measurement,first the values of battery electromotive force and circuit voltage are collected,and by adopting an adaptive approach,the adaptive online monitoring nodes are deployed.An online monitoring model for batter-ies using multi-layer support vector machines is constructed,with strengthened online monitoring through continuous tracking and early warning processing.The test results show that for the selected four batteries,three types of resistance wires with diameters of 0.8 mm,1.2 mm,and 2.1 mm are implanted in order to form different resistivities.After meas-urement and calculation,the average online monitoring frequency obtained can all reach 150 Hz or above.This indicates that the designed online monitoring method for batteries is more stable and safe,with stronger overall adaptability in dif-ferent environments and significantly improved monitoring efficiency.
multi-layer support vector machinebatteryonline monitoringdirectional identificationremote control
NETL
NSTL
万方数据
