P. HIRSCH, R. PIOTROWSKI, K. DUZINKIEWICZ, M. GROCHOWSKI
Gdansk University of Technology, Faculty of Electrical and Control Engineering,
Narutowicza 11/12, 80-233 Gdansk, Poland
piotr.hirsch@pg.edu.pl, robert.piotrowski@pg.edu.pl,
kazimierz.duzinkiewicz@pg.edu.pl, michal.grochowski@pg.edu.pl

Abstract: The paper presents the design of the integrated control system applied to Sequencing Batch Reactor (SBR) in a biological Wastewater Treatment Plant (WWTP) in Swarzewo, which operates under activated sludge technology. Based on the real data records, ASM2d biological processes model and aeration system model, hierarchical control system for dissolved oxygen tracking and cycle management is designed. Internal Model Controller (IMC) was applied to control the air flow at the lower control level. Higher level dissolved oxygen controller is based on Direct Model Reference Adaptive Control (DMRAC) method. The supervisory system performs management of reactor work cycle, determines the phase length, controls sludge age, calculates setpoint of dissolved oxygen and adapts parameters of the lower control layer. Proposed control system allowed to: increase the efficiency, improve the quality of outflow and reduce the cost of aeration and chemical treatment plant, in relation to existing solutions in case study plant.

Keywords: activated sludge, aeration, control, wastewater treatment, dynamic simulation, sequencing batch reactor.

>>Full text
CITE THIS PAPER AS:
P. HIRSCH, R. PIOTROWSKI, K. DUZINKIEWICZ, M. GROCHOWSKI, Supervisory Control System for Adaptive Phase and Work Cycle Management of Sequencing Wastewater Treatment Plant, Studies in Informatics and Control, ISSN 1220-1766, vol. 25(2), pp. 153-162, 2016. https://doi.org/10.24846/v25i2y201602

1. Introduction

1.1 General information

In industrial practice, two different types of activated sludge Wastewater Treatment Plant (WWTP) are used: Sequencing Batch Reactor (SBR) and WWTP with a continuous flow throughout the plant. In SBR, all biochemical processes occurs in one tank, in the predefined sequence. In the second type of plant, several tanks are connected by recirculation flows (internal and external). In this paper the SBR type of the WWTP is considered.

The SBR technology is widely used under small wastewater inflows conditions and may be designed using a single or multiple tanks in parallel. A typical work cycle involves five operational phases: filling, biochemical reactions (aerobic, anoxic, anaerobic), sedimentation, decantation and idle state.

A variety of different control strategies may be set up by selecting of number and sequence of SBR operational phases, in order to adjust SBR to specific working conditions of each WWTP. Because of this SBR gains operational flexibility, that allows this kind of WWTP to benefit from control system modernization. It has been shown that SBR pollutant removal efficiency can be greatly improved by application of step-feed control strategy [16]. A variant of this strategy is implemented in Swarzewo WWTP, Northern Poland, which is the case study considered in this paper. During the biochemical reaction phases, pollutions in the form of organic matter and nutrients (mainly phosphorus and nitrogen compounds) are degraded during the mineralization, nitrification, denitrification and enhanced biological phosphorus removal. Efficiency of those processes relay strongly on oxygen concentration, therefore high quality of DO control is crucial for fulfilment of discharge limits. The DO control is also important for WWTP energy efficiency, as the aeration system is responsible for 50-75% of plant power usage [1].

The supervisory control in continuous flow and SBR WWTPs is of a different nature. The main objective of the continuous flow WWTP supervisory controller is to determine the oxygen setpoint and recirculation flows, that will be sufficient to efficiently treat the wastewater. In the case of the SBR, objective of the supervisory controller is to manage the work cycle, namely the sequence and length of operational phases. Therefore SBR supervisory controller may be called work cycle controller.

1.2 Survey of related works

Broad range of control technologies have been researched and applied to DO control, e.g. adaptive controller [4,9,13], predictive controller [5,7,11,12,18], multivariable PID controller [19], fuzzy controller [3,14], fuzzy predictive controller [15,20]. Extensive review of DO control can be found in [1].

In [4] Direct Model Reference Adaptive Control (DMRAC) technology for DO tracking in Swarzewo case study was presented. Furthermore, analysis of the parameters of the adaptive controller for control quality was examined [13]. The authors used calibrated SBR Activated Sludge Model No. 2d (ASM2d) [8] coupled with validated aeration system model [14]. Simulations proved that the proposed solution provides satisfactory control quality in the case of most basic work cycles based on single filling pattern [4]. However, application of more complex cycle control strategies showed that discussed control system lacks adaptability. Classic DMRAC is ineffective when applied to SBR operating with step-feed strategy in occurrence of influent fluctuations. Hence, additional stepwise adaptation of DMRAC parameters (learning rates, anti-windup filter gains and starting points) was applied and examined in [9].

The most basic applications of SBR work cycle controllers are based on a fixed duration and composition of the work cycle and subsequent phases. Numerous studies were conducted on the optimal selection of these scenarios. In [2] the effect of total cycle time, phases duration and phase sequence on nutrient removal performance were examined. Systematic approach to examine and to determine the optimal operation strategy for nutrient removal was developed [16]. On-line method of phase duration optimization was presented in [10]. Discussed algorithm was based on measurements of oxidation reduction potential, dissolved oxygen and pH.