Design of Intelligent Control System for Cantilever Roadheader Based on BP Neural Network Algorithm

Cantilever roadheader has many advantages. When working in roadway, it can continuously excavate, avoid blasting vibration, and reduce overbreak. Due to the influence of dust and water mist in the working face during construction, it is difficult to control and make decisions on roadheader, which is easy to cause over-excavation and under-excavation, and it is difficult to ensure the forming quality of roadway, so it is urgent to improve the intelligent level of tunneling operation. In order to enrich the autonomous rectification control strategy of roadheader and approach the target of industrial application, this paper establishes a tracking control model with the goal of convergence of posture deviation of the fuselage, and proposes a tracking control method of roadheader based on parameter optimization of back propagation neural network (BPNN) for the purpose of improving the convergence efficiency of the model. The simulation results show that the trajectory tracking algorithm based on improved BPNN has small errors and excellent accuracy, which can reach more than 94% after many iterations. The control model based on BPNN is simple and easy to realize, and the posture deviation of the machine body can converge to zero in limited tracking steps and the speed adjustment process is stable, which proves that the trajectory tracking effect under the control of this model is good.