Control of Four Tank System Using Model Predictive Controller

Abstract

The four tank process is a non-linear process and it is the de-facto standard for the multiple input and multiple output(MIMO) process. It usually consists of four tanks which are interconnected each other with their inputs and output. Usually in the MIMO process the controlling of the output by changing the single input is difficult. So, the four tank process is a non-linear process. It consists of two pumps which acts as input to the quadruple tank process. The Model Predictive Control (MPC) is a suitable technique for predicting the state of output by using the current values of input. The advantage of using the Model Predictive Control in the non-linear quadruple process is by changing the inputs and output simultaneously. It can stabilize the linear process effectively and it can also be applied to the non-linear process to stabilize the output. The result obtained in the Model Predictive Control will be within the limited tolerance value while stabilizing the output of the non-linear tank process. In Model Predictive Control, it consists of Predictive control ‘P’ and Horizon control ‘M’. By changing or tuning the P and M, the optimized result can be obtained. The results of the above non-linear process is simulated using MATLAB and results shows the good tracking performance.