Due to the energy crisis and environmental pollution, electric vehicles have become more and more popular. Compared to traditional fuel vehicles, the electric vehicles are integrated with more high-voltage components, which have potential security risks of insulation. The insulation resistance between the chassis and the direct current bus of the battery pack is easily affected by factors such as temperature, humidity and vibration. In order to ensure the safe and reliable operation of the electric vehicles, it is necessary to detect the insulation resistance of the battery pack. This paper proposes an insulation detection scheme based on low-frequency signal injection method. Considering the insulation detector which can be easily affected by noises, the algorithm based on Kalman filter is proposed. Moreover, the battery pack is always in the states of charging and discharging during driving, which will lead to frequent changes in the voltage of the battery pack and affect the estimation accuracy of insulation detector. Therefore the recursive least squares algorithm is adopted to solve the problem that the detection results of insulation detector mutate with the voltage of the battery pack. The performance of the proposed method is verified by dynamic and static experiments.

由于能源危机和环境污染，电动汽车越来越受欢迎。与传统的燃料汽车相比，电动汽车集成了更多的高压组件，因此存在潜在的绝缘安全隐患。电池组的底盘与直流母线之间的绝缘电阻很容易受到温度，湿度和振动等因素的影响。为了确保电动车辆的安全可靠运行，有必要检测电池组的绝缘电阻。提出了一种基于低频信号注入方法的绝缘检测方案。考虑到绝缘检测器易受噪声影响，提出了一种基于卡尔曼滤波的算法。而且，电池组在行驶过程中始终处于充电和放电状态，这将导致电池组电压的频繁变化并影响绝缘检测器的估算精度。因此，采用递推最小二乘算法来解决绝缘检测器的检测结果随电池组电压的变化而变化的问题。通过动态和静态实验验证了该方法的性能。