ABSTRACT
Contents 7.1 Nature of Industrial Process Control. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 7.2 Bio-Inspired Algorithms in Base Control Layer . . . . . . . . . . . . . . . . . . . . . . . . . . 170 7.3 Advanced Control Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
7.3.1 Artificial Neural Networks in MPC Controllers . . . . . . . . . . . . . . . . . . 174 7.3.2 Hybrid Process Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
7.3.2.1 Step One . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 7.3.2.2 Step Two . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 7.3.2.3 Step Three . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 7.3.2.4 Step Four . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.4 SILO-Immune-Inspired Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 7.4.1 Immune Structure of SILO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
7.4.1.1 Pathogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 7.4.1.2 B Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 7.4.1.3 Antibody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
7.4.2 Basic Concept of SILO Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
7.4.3 Optimization Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 7.4.3.1 Mixed Model-Based Optimization Layer . . . . . . . . . . . . . . . 193 7.4.3.2 Global Model-Based Optimization Layer . . . . . . . . . . . . . . . 197 7.4.3.3 Stochastic Optimization Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 7.4.3.4 Layers Switching Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
7.5 Application of Bio-Inspired Methods in Industrial Process Control . . . . . 200 7.5.1 SILO Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 7.5.2 IVY Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 7.5.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
In this chapter, a layered structure of a control system is presented. Bio-inspired methods implemented in each control layer are introduced. The utilization of an evolutionary algorithm for a proportional-integral-derivative (PID) controller tuning task is briefly discussed. Moreover, an immune-inspired single input-single output (SISO) controller is presented. The model predictive control (MPC) algorithm with an artificial neural network (ANN) is an example of a bio-inspiredmethod applied in an advanced control layer. The design of IVY controller, which is a fuzzy MPC controller with an ANNmodel, is discussed in this chapter. The stochastic immune layer optimizer (SILO) system is another bio-inspired solution developed by the authors. This system is inspired by the operation of the immune system. It is used for online industrial process optimization and control. Results of IVY and SILO operation in real power plants are presented in the last section of this chapter. They confirm that bio-inspired solutions can improve control quality in real, large-scale plants.
