文献类型：期刊文献

中文题名：基于加权核偏最小二乘的低压断路器热脱扣时间预测方法

英文题名：Prediction method of heat tripping time of low voltage circuit breaker based on weighted kernel partial least squares

作者：操春名[1];张凌波[1];徐震浩[1]

机构：[1]华东理工大学自动化研究所,上海200237

年份：2020

卷号：57

期号：10

起止页码：140

中文期刊名：电测与仪表

外文期刊名：Electrical Measurement & Instrumentation

收录：CSTPCD;;北大核心:【北大核心2017】；

基金：国家自然科学基金资助项目(61573144,61773165,61673175)。

语种：中文

中文关键词：低压断路器;热脱扣时间;统计特性;加权核偏最小二乘;预测模型

外文关键词：low voltage circuit breaker;heat tripping time;statistical characteristics;weighted kernel partial least squares;prediction model

摘要：热脱扣时间是低压断路器产品质量的关键指标。为实现断路器热脱扣时间的实时预测,提出了一种加权核偏最小二乘(WKPLS)预测方法。该方法根据生产样本数据的分布与统计特性确定样本的权值,以解决断路器生产数据分布不平衡问题。以电流、环境温度、双金属片初始位置与焊接位置为输入量,利用WKPLS方法建立低压断路器热脱扣时间模型。利用采集的断路器生产数据进行仿真实验,结果表明基于WKPLS方法建立的断路器热脱扣时间模型预测精度较高,且该模型可为断路器生产过程中的控制参数调整和异常监测提供一定参考。
Heat tripping time is a key specification of product quality of low voltage circuit breaker. In order to realize the real-time prediction of its heat tripping time,this paper puts forward a prediction method of weighted kernel partial squares( WKPLS). Firstly,the weight values of the samples are determined according to the distribution and statistical characteristic of the production sample data to solve the problem of unbalanced distribution of production data of the circuit breaker.Secondly,taking the current,ambient temperature,initial position and welding position of bimetallic sheet as input,the heat tripping time model of the low voltage circuit breaker is established by means of WKPLS method. Finally,simulation experiment is carried out with the production data of the collected circuit breaker,which shows that the heat tripping time model of circuit breaker based on WKPLS method is of high prediction accuracy. The model can provide reference for the adjustments of control parameters and abnormal monitoring in the production process of circuit breaker.