Multi-Objective Optimization of the Process Parameters of a Grinding Robot Using LSTM-MLP-NSGAII

Grinding robots are widely used in the automotive, mechanical processing, aerospace industries, among others, due to their strong adaptability, high safety and intelligence. The grinding process parameters main factors that affect quality efficiency of robots. However, it is difficult obtain optimal combination only by manual experience. This study proposes an artificial intelligence-based method for optimizing a robot using neural networks genetic algorithm, with aim reduce workpiece surface roughness shorten time. Specifically, this first utilizing multi-objective optimization approach optimize robot. Based on experimental data ROKAE XB7, long short-term memory (LSTM) multilayer perceptron (MLP) were trained fit quantitative relationships between robot, such as feed rate, spindle pressure pneumatic motor pressure, result After that, non-dominated sorting algorithm II (NSGA-II) was calculate Pareto parameter combinations LSTM MPL model objective function. Compared based experience, optimized achieved reduction at least 13.62% whole time 28%. excellent results obtained validated feasibility proposed robots’ practical manufacturing applications.