改进随机森林法在断路器机械故障诊断中的应用

Application of Improved Random Forest Method in Diagnosis of Mechanical Faults of Circuit Breakers

高压断路器机械故障时有发生,严重影响设备及电力系统的安全稳定运行。针对目前高压断路器机械故障诊断准确率不理想的现状,将改进后的随机森林方法应用于高压断路器机械故障诊断。结合高压断路器分合闸线圈电流信号的原理分析,提取出可有效反映断路器机械故障的相应特征量,建立高压断路器机械故障诊断模型,通过不同噪声干扰强度下高压断路器机械故障诊断实例的对比分析。结果表明,随机森林方法具有更加优良的准确性和抗干扰能力。同时验证了基于改进随机森林方法的高压断路器机械故障诊断模型的有效性和优越性。

Mechanical failures of high-voltage circuit breakers occur from time to time,thus seriously affecting safe and stable operation of equipment and power system.Under consideration of non-ideal accuracy of current diagnosis of mechanical faults of high-voltage circuit breakers,the improved random forest method was applied to the diagnosis.With the help of an analysis of the principle of the current signal of the breaking-closing coil of the circuit breaker,corresponding characteristic quantities capable of effective reflection of mechanical faults of circuit breakers were extracted,and a diagnosis model was established.Results of comparative analysis of fault diagnosis examples under different noise interference intensities indicated that the proposed approach had better accuracy and anti-interference ability,and the effectiveness and superiority of the diagnosis model for mechanical faults of high-voltage circuit breakers based on the improved random forest method was verified.