Level Control of Coupled Tank System Based on Neural Network Techniques

Abstract

The control design of coupled tanks is not an easy task due to the nonlinear characteristic of the valves, and the interactions between the controlled variables. Those features pose a challenge in the automatic control, so that linear controllers, such as conventional PID, might not work properly for regulating this MIMO system. Some advanced control techniques (e. g. control based on neural networks) can be used since neural networks are universal approximators which can deal with nonlinearities and interactions between process variables. In the present work, an experimental investigation was performed presenting a comparison between two neural network-based techniques and testing the feasibility of these techniques in the coupled tanks system. First principles simulations helped to find suitable parameters for the controllers. The results showed that the model predictive control based on artificial neural networks presented the best performance for supervisory tests. On the other hand, the inverse neural network needed a very accurate model and small plant-model mismatches led to undesirable offsets.

Design of 3×3 Multi Input Multi Output (MIMO) Decoupling on Coupled Tank System with PID Controller

“…In the process sector, there are various types of tanks, including coupled tanks and solitary tanks, both of which are separate from other tanks and do not have a connection (Sadli, 2014). Coupled tanks are categorized as MIMO (multiple input, multiple output) systems, where MIMO system settings are more challenging than SISO (single input, single output) systems (Ogata & Brewer, 2010;Sousa et al, 2020). A nonlinear coupled tank is one that has two tanks connected by a pipe, such as the coupled tank system (Pangestu et al, 2022;Saputra et al, 2018;Ulum, 2017).…”

Desarrollo de un sistema de predicción y control de solventes durante la fermentación de cerveza con S. Cerevisiae mediante redes neuronales y lógica difusa

“…Realizando una revisión bibliográfica, encontramos otros trabajos en los cuales esta técnica ha sido aplicada con éxito; por ejemplo, en el control de nivel de tanques, prediciendo los parámetros requeridos por el controlador (Sousa, Silva, & Fileti, 2019); control de la temperatura de un biorreactor para producir etanol (Imtiaz, Assadzadeh, Jamuar, & Sahu, 2013) y un modelamiento para el proceso de fermentación de la penicilina (Wang, B., Ji, & Zhuang, 2015), entre otras, pero ninguna centrada en los compuestos solventes en la fermentación alcohólica.…”