基于HK-RVM与WCO的电梯门锁故障预测

Failure prediction of elevator door lock based on HK-RVM and WCO

电梯门锁触点故障是电梯发生故障的主要原因之一,为了实现及时监测和有效预警,文中提出了基于世界杯优化算法(WCO)的混合核相关向量机(HK-RVM)预测模型。HK-RVM是一种具有良好概率预测能力的区间预测方法,引入WCO优化参数以提高预测模型的泛化能力和准确性。首先利用门锁触点模拟组块和单片机系统模块,采集可以反映电梯门锁故障的电阻时序信号;其次通过WCO-HK-RVM建立实时时序预测模型,其在保证模型的预测精度和泛化能力的同时,实现了门锁触点的故障预测;最后为了验证了该方法的优越性和适用性,通过实验研究与传统预测算法LSTM和SVM相比较,结果表明WCO-HK-RVM预测方法显著提升了门锁触点故障的预测精度。

Contact failure of elevator door lock is one of the main causes of elevator failure.To monitor and warn effectively in time,a hybrid kernel relevance vector machine(HK-RVM)prediction model based on World Cup optimization algorithm(WCO)is proposed.HK-RVM is an interval prediction method with good probability prediction ability.It introduces WCO optimization parameters to improve the generalization ability and accuracy of the prediction model.Firstly,using the lock contact simulation block and the single chip microcomputer system module,the resistance time sequence signal which can reflect the elevator lock failure was collected;secondly,a real-time time sequence prediction model was established by WCO-HK-RVM,which ensures that the prediction accuracy and generalization ability of the model can be guaranteed,and the door lock contact failure can be predicted;finally,to verify the superiority and applicability of this method,this method was compared with traditional prediction algorithms LSTM and SVM through experimental studies.The results show that WCO-HK-RVM can significantly improve the prediction accuracy of door lock contact failures.