矿井无人驾驶单轨吊安全性能关键参数识别

Safety performance of unmanned monorail cranes in mines key parameters identification research

单轨吊在复杂深部矿井环境的辅助运输系统中具有不可替代的作用,由于目前无法有效精确识别单轨吊的荷载质量和轨道坡度,直接影响了运输安全性能及能量高效利用。因此,笔者提出了矿井无人驾驶单轨吊安全性能关键参数识别方法。针对单轨吊结构特性和轨道运输特点,对具有强耦合关系的荷载质量和轨道坡度建立了纵向动力学模型;基于运行数据和带有动态遗忘因子的递推最小二乘算法(DFFRLS)对纵向动力学模型参数进行实时在线识别,实现荷载质量和轨道坡度的精准解耦;并基于解耦的纵向动力学模型和识别的模型参数,动态修正当前的荷载质量识别值,以消除误差,完成荷载质量的高精度识别;由识别的纵向动力学模型参数、运行数据,应用Sage-Husa自适应扩展卡尔曼滤波算法(AEKF)对系统噪声协方差和误差协方差进行动态更新,滤除环境噪声干扰,实时调节和修正当前轨道坡度值,保证轨道坡度识别的精准性。在多种工况下,仿真与实际应用表明,基于DFFRLS-AEKF方法的荷载质量识别值与实际值的误差在3.2%以内,运行轨道坡度识别值与实际值的误差在5.3%以内。该方法可实现无人驾驶单轨吊安全性能关键参数的实时精准获取,有效减少无人驾驶单轨吊安全事故的发生,显著提升无人驾驶单轨吊的能量高效利用。

Monorail cranes have an irreplaceable role in the auxiliary transport system in complex deep mine environment,which directly affects the transport safety performance and efficient energy utilization due to the current inability to effectively and accurately identify the load quality and track slope of monorail cranes.For this reason,this paper proposes a method for identifying key parameters of safety performance of unmanned monorail cranes in mines based on DFFRLS-AEKF.Based on the structural characteristics of the monorail crane and the characteristics of track transportation,a longitudinal dynamics model is established for the load mass and track slope with strong coupling relationship;the parameters of the longitudinal dynamics model are identified online in real time based on the operational data and the recursive least squares algorithm with dynamic forgetting factor to achieve the accurate decoupling of the load mass and track slope;and based on the decoupled longitudinal dynamics model and the identified model parameters,the current load mass identification is dynamically modified.Based on the decoupled longitudinal dynamics model and the identified model parameters,the current load quality recognition value is dynamically corrected to eliminate the error and complete the high-precision recognition of load quality;from the identified longitudinal dynamics model parameters and operation data,the Sage-Husa adaptive extended Kalman filter algorithm is applied to dynamically update the system noise covariance and error covariance,filter out the environmental noise interference,adjust and correct the current track slope value in real time,and ensure the accuracy of track slope recognition.Accuracy.The simulation and practical application show that the error between the load mass recognition value and the actual value is within 3.2%and the error between the running track slope recognition value and the actual value is within 5.3%under various working conditions.The method achieves real-time and accurate acquisition of key parameters of the safety performance of the unmanned monorail crane,effectively reduces the occurrence of safety accidents of the unmanned monorail crane,and significantly improves the efficient utilization of energy of the unmanned monorail crane.