掘进机姿态调整模型辨识方法与精准控制

Method of Model Identification and Precise Control for Tunnel Boring Machine Body Posture Adjustment

针对因掘进机姿态控制精度较差导致巷道掘进轨迹偏离进而影响煤矿生产安全与效率的问题,提出了一种基于粒子群算法的掘进机姿态调整控制模型辨识方法及模糊神经网络PID姿态控制方法。首先分析了掘进机姿态角误差与巷道断面之间关系,基于姿态运动学与液压模型确定了姿态控制系统传递函数结构,通过粒子群算法的寻优特性与输入输出信号的响应对传递函数参数进行拟合进而实现辨识。仿真结果表明,与采用最小二乘和遗传算法的辨识方法相比,所提采用粒子群算法的模型辨识方法的辨识精度分别提高了96.75%和95.15%。此外,利用模糊控制的强非线性适应性与神经网络的自学习能力设计了姿态调整的PID控制系统。搭建了EBZ-55型掘进机俯仰角控制试验平台,并在不同的试验工况下进行了多组俯仰角控制试验。试验结果表明,基于粒子群模型辨识方法的模糊神经网络PID控制方法比模糊PID控制方法最多可降低57.1%的超调误差以及53.5%的响应时间,并能稳定实现1°以内的姿态角控制。该方法可为掘进机及大型液压设备的姿态控制提供有效的技术支撑与参考。

Aiming at the derivation of the tunnel excavation track and the coal production inefficiency and safety problem caused by poor posture control accuracy of tunnel boring machine(TBM),a model identification method based on the particle swarm optimization(PSO)and a fuzzy neural network control method of attitude control are proposed.Firstly,the relation between the posture error and the tunnel section is analyzed,and the structure of the control system transfer function is determined according to the kinematic and hydraulic system model.Through the PSO searching character and the response of the input and output signals,the transfer function parameters could be fitted.Simulation results show that the identification accuracy based on PSO is 96.75%and 95.15%higher,respectively,than that based on least square or genetic algorithm identification method.Making use of the fuzzy logic control’strong nonlinearity adaptability and the neural network’self-learning ability,a posture PID control algorithm is proposed,and a EBZ-55 tunnel boring machine attitude control experiment system is designed and established.Experiment shows that compared with the fuzzy PID,the proposed fuzzy neural network PID control system could mostly reduce 57.1%overshoot error and 53.5%response time in different conditions,and the posture of the tunneling boring machine could be controlled in 1°stably.