Precision chemical compounds analysis in chemical tank reactors using conductivity analysis and model predictive control

Abstract

One fundamental property of conductivity is its direct proportion to the electrolytic liquid concentration, and this property is a crucial monitoring component in the Continuous Stirring Tank Reactors (CSTR). The CSTR is an essential industrial chemical preparation tank for chemical product manufacturing and process treatment plants. The paper simulates and tests a response of a concentration control scheme with conductivity as a sensing stimulus. Moreover, it integrates a four-electrode conductivity meter based on Arduino and a functional Model Predictive Control (MPC) algorithm. Conductivity gave good predictability of a solution's concentration changes and good precision with less than 6% error accuracy from the setpoint. Although it showed a slight delay in the response, there were fewer overshoots compared to other controller-based concentration monitoring schemes. The designed conductivity meter was suitable for monitoring moderate and high range concentration changes with an acceptable response time.