Fuzzy Model-Based Predictive Control applied to multivariable level control of multi tank system

Abstract

In this study issues related to applicability of ModelBased Predictive Control (MBPC) to nonlinear and complex processes are addressed. A tank system is taken as an exemplary process, and its prediction model is used for control purposes. Obtained results are applied for level control of a tank process. A Takagi-Sugeno type fuzzy neural network is used to model the nonlinear system. The obtained model is represented in statespace implementation. It is embedded into a model predictive control scheme and ensures the optimization procedure of MPC. Furthermore, thus formulated MPC strategy can be treated as a quadratic programming (QP) problem. It ensures ability to handle physical constraints of the system. Optimization objectives in MPC include minimization of the difference between the predicted and desired response trajectories, and the control effort subjected to prescribed constraints. The case study is implemented in MATLAB&Simulink environment.

Data-based Tuning of PI Controllers for Vertical Three-Tank Systems

“…Vertical three-tank systems are nonlinear Multi Input-Multi Output (MIMO) benchmarks which convincingly illustrate control design, fault detection and diagnosis problems. Some current approaches to the level control of vertical three-tank systems include fault diagnosis using sliding mode observers (Orani et al, 2010), fuzzy model-based predictive control (Ahmed et al, 2010), discrete-time model identification (Nikolić et al, 2010), sensitivity analysis of process models (Antić et al, 2011) and optimal tuning of PID controllers by nature-inspired algorithms (Kumar and Dhiman, 2011).…”