Deep Learning Based System Identification and Nonlinear Model Predictive Control of pH Neutralization Process

An essential step in the progression of nonlinear systemidentificationis the inception of recurrent and convolution-type deep learning methodsin industrial units. Many chemical/pharmaceutical/wastewater processunits employ pH neutralization schemes to check the acidity and alkalinityof the product before bringing them for industrial production. Theinherent nonlinear dynamics, especially during the neutralizationof strong acid by a strong base, pose rigorous difficulties to modeluncertainties, making it highly challenging to implement automaticpH control. This research endeavor focuses on building a deep TemporalConvolution Network (TCN) with a larger receptive field to learn thedynamics of the pH neutralization process. This method uses a 1-Dcausal convolution strategy with a residual learning framework toperform dilated causal convolutions. The simulation studies of theproposed TCN-based identification-scheme are adopted and executedin a Python environment for two important case studies, namely, (a)single tank pH process and (b) ETP (effluent treatment plant)-pH process.The proposed TCN framework manifested supremacy over predicted modelresponses of LSTM (long short-term memory) and MLP (multilayer-perceptron)architectures in terms of accuracy while requiring less training time.Furthermore, in this research, a state-of-art TCN-based NMPC (nonlinearmodel predictive control) and LSTM-based NMPC schemes are designed,implemented, and investigated where the control performance revealedthe precision of the former.