Neural Network Predictive Controller For Temperature Process Station

Abstract

The importance of the temperature control system in industrial processes is growing more and more apparent. Recently, a significant amount of research has been undertaken on temperature control systems that use a variety of control algorithms. Because of their powerful self-learning and parameter altering capabilities, neural networks (NN) have been frequently used to solve highly nonlinear control difficulties in industrial processes. The temperature sensor is responsible for measuring the temperature of the process. The mathematical model of the system is identified based on the input and output data from the process itself. After that, the NN predictive controller is constructed to maintain the process temperature at the predetermined temperature. The constructed controller is evaluated using a variety of different inputs. The performance of the controller is evaluated using the metrics such as integral square error, integral absolute error, and integral time absolute error. The response time of the system is taken into consideration in addition to this. The following goals must be met for the project to be a success and put into action. Measure the temperature of the process using RTD, Design an NN predictive controller to maintain the temperature of the process based on the setpoint, and evaluate the controller using ISE, IAE, and ITAE.