Supercritical Carbon Dioxide Equipment Neural Network Predictive Control Research

The technology of Supercritical Carbon Dioxide dyeing is new technology which uses supercritical carbon dioxide to be the medium of dyeing. It not only shortens the dyeing time, and eliminates the environmental pollution. It belongs to green production process, and it will bring revolution in the textile dyeing process.However, the supercritical technology needs pressure above 30MPa and precise control of temperature, and it increases the complexity of the system. Conventional controller can not meet the requirements of high-precision parameters of dyeing process. It is unable to raise the quality and efficiency of staining. The complexity of the plant displays highly nonlinear, high noise interference, and with its dynamic characteristics of the changes in operating conditions change drastically, etc. It is difficult to describe the complex using the precise mathematical model. The requirement of the control system is much higher and higher, and we need improve the level of Intelligent Control System urgently. Therefore, we introduce a very good method to solve this problem.This thesis is about the temperature plant controller research of the supercritical CO2 dyeing equipment in Dalian Institute of Light Industry textile professionals. In the thesis I access to a number of related documents, and introduce a neural network predictive control algorithm to solve these problems, and then I conduct deep research and discussion. First, this thesis is analyzed supercritical CO2 dyeing process and the characteristics of devices, and I understand the dynamic characteristics of objects, control requirements and every process parameters on the impact of technology. Secondly, in view of the temperature of supercritical CO2 dyeing equipment I build the neural network modeling and have deep research on the neural network predictive control algorithm. Finally, through simulation with MATLAB, I compare the result with conventional PID control algorithm with the neural network predictive control algorithm. I get a fact that control algorithm in the application of the temperature control of supercritical dyeing equipment can achieve the desired results in this thesis.The design of the controller has a good dynamic performance, anti-disturbance, robustness and tracking performance in this thesis. This new type of controller will be in practice once, and the technology of supercritical CO2 dyeing will be used in industry production, and it will bring huge economic benefits society and social benefits.