Study on Optimization Method for CNC Machining Plastic-Shaped Appliances Based on ICOA Algorithm

The issue of deformation during the secondary processing of plastic plate is becoming increasingly concerning with the expanding use of plastic-shaped appliances. This is especially evident due to work-hardening and thermal softening, leading to problems such as hot-melt adhesion and distortion. To tackle this challenge, a study employed the Theory of Inventive Problem Solving (TRIZ) to conduct a systematic analysis of the problem. It identified contradictions and conflicts therein and proposed an innovative optimization strategy based on the Improved Coati Optimization Algorithm (ICOA). Initially, five crucial processing parameters—spindle speed (A_S), cutting tool material (A_CM), feed rate (A_vf), plastic plate material (A_PM), and cutting depth (A_ap)—are determined through grey relational analysis and utilized as input for the prediction model. Subsequently, a processing state prediction model is constructed on the MATLAB software platform, and the ICOA algorithm is used to optimize the model parameters of the convolutional gated recurrent unit network (CNN-GRU), thereby enhancing its robustness, generalization ability, and prediction accuracy. Experimental validation on the PT-1325 CNC machine tool demonstrated that the CNN-GRU model optimized by ICOA attained a prediction accuracy of 96.67% and effectively enhanced machining efficiency through optimizing the machining parameters.