当前位置: X-MOL 学术Trans. Inst. Meas. Control › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Coordinated optimization control of shield machine based on dynamic fuzzy neural network direct inverse control
Transactions of the Institute of Measurement and Control ( IF 1.8 ) Pub Date : 2020-12-24 , DOI: 10.1177/0142331220980274
Xuanyu Liu 1 , Wentao Wang 1 , Yudong Wang 2 , Cheng Shao 3 , Qiumei Cong 1
Affiliation  

During shield machine tunneling, the earth pressure in the sealed cabin must be kept balanced to ensure construction safety. As there is a strong nonlinear coupling relationship among the tunneling parameters, it is difficult to control the balance between the amount of soil entered and the amount discharged in the sealed cabin. So, the control effect of excavation face stability is poor. For this purpose, a coordinated optimization control method of shield machine based on dynamic fuzzy neural network (D-FNN) direct inverse control is proposed. The cutter head torque, advance speed, thrust, screw conveyor speed and earth pressure difference in the sealed cabin are selected as inputs, and the D-FNN control model of the control parameters is established, whose output are screw conveyor speed and advance speed at the next moment. The error reduction rate method is introduced to trim and identify the network structure to optimize the control model. On this basis, an optimal control system for earth pressure balance (EPB) of shield machine is established based on the direct inverse control method. The simulation results show that the method can optimize the control parameters coordinately according to the changes of the construction environment, effectively reduce the earth pressure fluctuations during shield tunneling, and can better control the stability of the excavation surface.



中文翻译:

基于动态模糊神经网络直接逆控制的盾构协调优化控制

在盾构机隧道掘进过程中,必须保持密封室内的土压力平衡,以确保施工安全。由于掘进参数之间存在很强的非线性耦合关系,因此很难控制进入的土壤量与密闭车厢中排放量之间的平衡。因此,开挖面稳定性的控制效果差。为此,提出了一种基于动态模糊神经网络(D-FNN)直接逆控制的盾构协调优化控制方法。选择密封室中的刀头扭矩,前进速度,推力,螺旋输送机速度和土压差作为输入,建立控制参数的D-FNN控制模型,其输出为螺旋输送机速度和前进速度。下一刻。引入误差减少率方法来修整和识别网络结构,以优化控制模型。在此基础上,建立了基于直接逆控制方法的盾构机土压力平衡最优控制系统。仿真结果表明,该方法可以根据施工环境的变化而协调地优化控制参数,有效减少盾构掘进过程中的土压力波动,可以更好地控制基坑开挖面的稳定性。

更新日期:2020-12-24
down
wechat
bug