In this paper, the machine vision integration method is first investigated to design the coal mine working environment monitoring system for electromechanical equipment. Then, the PID control algorithm is used to remote control various electromechanical equipment in coal mines. Finally, we analyze the effect of the intelligent remote control function combined with machine vision on the coal mine electromechanical monitoring and environmental monitoring, analyze the effect of machine vision device with deviation detection and coal pile detection effect, as well as compare the optimization results and fault separation correct rate of different machine vision fault diagnosis methods for electric power equipment. According to the results, the fault diagnosis adaptability of the optimization results of power equipment fault diagnosis using this paper’s method is approximately 95, and the fault separation accuracy rate is between 92% and 96%.