介绍了一种基于由电压时间曲线特征点(时间、电压)构成的向量空间电池筛选方法,提出了基于曲率的电压时间曲线特征提取方法,并依据特征点分段线性拟合出标准电池电压时间曲线方程,通过计算非标准电池特征点到标准电池电压时间曲线的距离,从而确定单体电池间的相似度。该方法通过提取电压时间曲线上的时间、电压为特征,建立相似度函数并基于一定的阈值准则,完成电压时间曲线的筛选,并对分选结果的一致性进行了分析,为电池分选提供了一种快速、实用的解决方案。
An efficient and practical cell sieving method on the basis of feature points on the voltage-time curve was discussed.According to curvature,a feature extraction method was proposed to analyze the curve.Then the standard cell voltage-time curve equation was achieved by piecewise linear fitting.By calculating the distance between the non-standard cell feature points and the standard cell voltage-time curve,the similarity between cells was determined.To be exactly,similarity function was established,with certain threshold criteria applied as well,to sieving the voltage-time curves and the consistency of the results were analyzed.