摘要
针对国内游乐设施行业载人设备运行状态监测不够完善、故障诊断多数情况下需要依赖技术人员经验的现状,分析了基于机理模型、基于知识和基于数据驱动的载人设备故障预警方法,研究了载人设备故障预测的多元线性回归算法,并利用矿山车的时间系数、人数系数、温度系数,建立了矿山车载人设备的多元线性回归模型。通过矿山车15个同步点区间用时预测案例,在可视化平台上显示预测和实际运行曲线,实现了对矿山车或过山车类载人设备运行故障的有效预警。
In view of the insufficiency of monitoring the running status of manned equipment and rely on the experience of technicians mostly in domestic rides,the paper analyses the means of manned equipment fault warning,including the mechanism model,knowledge-based and data-driven.By studying the multivariate linear regression algorithm for manned equipment fault prediction,it establishes a multiple linear regression prediction model,which uses the time coefficient,number of people coefficient,temperature coefficient of the mining roller coaster.According to the case of predicting the time interval of 15 synchronization points of mining roller coaster,it displays the prediction and actual running curves on the visualization platform and realizes an effective warning for the running failure.
引文
[1]严波.大型游乐设施安全评价技术研究[J].中国高新技术企业,2016,(29):170-171.
[2]CHEN B,CHEN S.A study on applications of multisensor information fusion in pulsed—GTAW[J],Industrial Robot,2010,37(2):168-176.
[3]宋欣,王翠荣.基于线性回归的无线传感器网络分布式数据采集优化策略[J].计算机学报,2012,35(03):568-580.
[4]田垅,刘宗田.最小二乘法分段直线拟合[J].计算机科学,2012,39(S1):482-484.