Prediction of TBM operation parameters using machine learning models based on BPSO

•An optimization algorithm BPSO is proposed to improve the random searching ability of a single particle in PSO.•We analyze the convergence of our proposed algorithm mathematically and give a reasonable range of parameters.•The search efficiency of hyper parameters for machine learning models is improved.•A hybrid machine learning model is proposed to predict the speed of screw conveyers. The precision and accuracy of controlling parameters are improved.