A real-time optimal control method of deep neural network for injection molding machine

Abstract

The present invention relates to the technical field of injection molding control, and more specifically, relates to a real-time optimal control method of a deep neural network for an injection molding machine, including: S10. Establishing a dynamic mathematical model of the injection filling process of the injection molding machine, and solving the flow rate control problem of the injection molding machine Convert to solving the optimal control problem with constraints; S20. Perform iterative offline optimization on the dynamic mathematical model to generate optimal state-control data sets based on different initial state starting points; S30. Use the optimal state-control data set for training Deep neural network, the deep neural network learns the mathematical relationship of the nonlinear mapping between the input state and the output optimal action; S40. Collect the current state data of the injection molding machine, input the trained deep neural network, and output the control signal of the injection molding machine. The invention combines the optimal control with the deep neural network, so that the current system state of the injection molding machine can quickly respond to the current optimal input control signal of the servo valve motor of the injection molding machine in the next step.

Description

technical field [0001] The invention relates to the technical field of injection molding control, and more specifically, to a real-time optimal control method of a deep neural network for an injection molding machine. Background technique [0002] Injection molding technology is a processing technology that transforms thermoplastic and thermosetting materials into plastic products. Injection molding machines are used as professional working machines for processing plastic parts and other plastic industries. 70% of plastic parts are produced by it. Important technical equipment in high-tech fields such as electrical and optoelectronic communications provide important equipment support for high-end manufacturing industries such as new energy, new materials, energy conservation and environmental protection, and biomedicine. The injection flow rate of the molten polymer inside the injection molding machine is one of the key control process parameters in the injection molding pro...

AI Technical Summary

Problems solved by technology

Although the above scheme can convert the injection molding process parameters into control parameters, and the control parameters are directly used in the injection molding process, however, the above scheme adjusts the internal parameters of the injection molding machine offline for different conditions and working conditions, and the optimal input parameters are online. This process is time-consuming and labor-intensive, and the robustness is poor; once the initial conditions of the system change or the working conditions change temporarily, it is necessary to re-control the internal control of the system according to the new situation Correction and optimization of parameters, there are problems such as poor stability, long time, high cost, etc., and the online real-time feedback optimal control of the injection molding machine cannot be achieved

Images