Investigation of the influence of network-induced time delays on the activated sludge process behavior in the networked wastewater distributed systems

Abstract

This paper examines the effects of networked induced time delays on the dissolved oxygen (DO) concentration in the activated sludge process (ASP) of a networked wastewater treatment plant (WWTP). This is a situation in which the controller and the wastewater plant are separated by wide geographical distance. This is a new type of WWTP control that allows two or more WWTPs to be controlled by a single controller placed in a remote location. The objective is to achieve flexibility of control and to reduce its cost. The communication medium between the controller and the WWTPs introduces communication drawbacks into the control system. The influences of network-induced time delays [controller to actuator delay (τca) and the sensor to controller delay (τsc)] over the behavior of the DO process controlled by both nonlinear linearizing and proportional-integral controllers are investigated for constant and random delays. Investigation of the DO process under random delays was also performed with varying linear controller parameters [proportional gain (Kp ) and integral time (TI )]. Simulation results reveal that large network-induced time delays in the closed-loop DO process leads to depletion of the amount of oxygen available for microorganism metabolism, leading to inefficiency of the ASP. The critical delay during which the DO process becomes unstable due to communication drawbacks was also determined for constant and random delays. These values are found to vary depending on the delay type (constant/random), delay magnitude, and the linear controller parameters Kp and TI . The results of this study would provide useful information for process performance and form the basis for the design of a robust networked control for the DO process capable of mitigating communication drawbacks in a networked wastewater distributed systems