Piecewise Linear Neural Network for Process Control in Industrial Environment

The assumption, that once a nonlinear problem is modeled by a piecewise-linear model, it can be advantageously solved by many efficient techniques, was proposed and proved many times. Beside, authors of this contribution have recently published a new smart linearization technique based on neural networks. This technique provides a set of linear equations and each of these equations is valid in some region of state space of considered system. Together, they approximate the whole nonlinear process. The technique has been comprehensively described and its advantages have been demonstrated in previous authors' publications. The mentioned technique is encouraged to be used especially in process control for controllers tuning. The issue is, that hardware used in industrial applications is reliable and robust, but rather simple and with low performance. Thus, the aim of this paper is to show, whether the proposed technique is capable to be applied in industrial environment. To be more specific, one particular programmable logic controller is used to control a real nonlinear continuous system and for the control law evaluation, mentioned technique is used. The process is then reviewed from the perspective of computational complexity and functional use.