Prediction model of high-voltage pulse boring rock-breaking process and intelligent identification of model parameters

Abstract

High-voltage electro-pulse boring (EPB) is featured in high drilling efficiency, low energy consumption and low drilling cost. Firstly, based on the energy conservation law, an EPB process prediction model is established, which is composed of EPB discharge rock-breaking circuit sub-model based on distributed capacitance and time-varying resistance, and shock wave sub-model. The time delay characteristics of a gas discharge switch and the distributed capacitance in a discharge circuit are taken into account in the EPB discharge rock-breaking circuit sub-model to make it closer to the actual discharge circuit. Meanwhile, the model can predict the real-time process parameters such as discharge voltage, power, and energy in the EPB process. The shock wave sub-model can predict the plasma channel radius, shock wave and other process parameters. Secondly, an EPB test system is developed to realize EPB in different rock samples and the detection of electric parameters. A singular value decomposition (SVD) filtering algorithm is developed to extract the useful current signals in the environment with strong noise and interference. Thirdly, based on genetic algorithm (GA) and particle swarm optimization (PSO), an improved hybrid intelligent algorithm is proposed by improving the fitness function and inertia weight. According to the electric parameter curves of the EPB in different rock samples, the parameters of EPB discharge rock-breaking circuit sub-model are identified by the improved PSO-GA. After this algorithm is adopted to identify the model parameters, the current curve is fitted. Compared with other identification hybrid algorithms, the average convergence speed of the improved PSO-GA algorithm is increased by 50.1%, and the average percentage of parameter identification error is reduced by 23.9%. Finally, according to the identified model parameters, the real-time parameters of the shock wave are obtained in the process of the high-voltage EPB. The mechanism of high-voltage EPB process is analyzed and summarized. This research is able to predict the process parameters of high-voltage EPB rock-breaking and summarize the law of high-voltage EPB rock-breaking. Therefore, this research is greatly important to guide the selection of EPB process parameters and improve the drilling efficiency. • An EPB process prediction model is established. • The EPB test system is developed, and the electric curves are filtered. • An improved PSO-GA is proposed. • The mechanism of high-voltage EPB process is analyzed. • The correctness of the prediction model and improved PSO-GA are proved.