Model predictive control and fuzzy control in a hierarchical structure for wastewater treatment plants

Abstract

This paper presents a hierarchical two-level control structure for a biological wastewater treatment plant (WWTP), with the goal of simultaneously improving the effluent quality and reducing operational costs. The Benchmark Simulation Model No.1 is used as working scenario.The idea is to adjust the dissolved oxygen in the fifth tank (DO5) according with the present working conditions, instead of keeping it in a fixed value. Thus is concreted by a hierarchical structure. MPC with inlet flow rate feedforward control (MPC + FF) is proposed for the lower level to control nitrate nitrogen concentration of the second tank (NO2) and DO5. The high level is the responsible for adjusting the DO setpoint in the fifth tank (DO5 setpoint) of the controller of the low level, according with the ammonium and ammonia nitrogen concentration in the fifth tank (NH5). The controller for the high level is based on a Fuzzy control. A tuning region has been determined for the high level controller, in which the results show a simultaneously improvement of the effluent quality and operational cost.

Hierarchical predictive control of Wastewater Treatment Plants

“…In [18] a modified version of GPC algorithm based on a positional form for both the process model and controller cost function is presented. In this paper only the cost function (13) is considered in the positional form and the CARIMA process model is considered. Thus, the minimization of the cost function (13) with the output of the controller replaced by u(t)=u(t-1)+Δu(t) where u(t-1) is known (u (t-1)=u') leads to:…”

“…Thus, optimizing the aeration process actually means determining the optimal value of the dissolved oxygen setpoint. Several strategies have been proposed and Model Predictive Control (MPC) proved to be a suitable one [12,13,14]. Due to specific features of WWTP a two-layer hierarchical structure is developed: local process control loops as a lower layer and a higher layer of process optimization.…”

Fuzzy Control and Model Predictive Control Configurations for Effluent Violations Removal in Wastewater Treatment Plants

“…Due to the presence of strong disturbances in WWTPs, MPC has difficulties in keeping the controlled variables at their reference level. To compensate the disturbances, a feedforward control is added, as in refs − and . MPC provides options for the feedforward compensation of the measured disturbances, in the same way as for the reference signals.…”

“…On the contrary, when NH decreases, less DO is required, producing less NO. NH5 is not forced to keep a fixed value since it has been observed that it is not possible by manipulating the DO set point due to the large disturbance − , .…”

“…Other studies deal with the basic control strategy (DO of the aerated tanks and NO of the last anoxic tank), but testing with different controllers such Model Predictive Controller (MPC) and fuzzy controller. − These methods provide an acceptable balance between quality and costs. Finally, other investigations propose a hierarchical control that regulates the DO set points, depending on some states of the plant, usually NH and NO concentration values in any tank or in the influent − or DO in other tanks.…”