Model-based optimization and scenario planning for a large-scale water resource recovery facility

Several configurations of biological nutrient-removal(BNR) systems have been developed to improve nutrient removal and meet recently established stringent effluent permits for wastewater treatment plants (WWTPs). Since the majority of BNR configurations are modified versions of the conventional activated sludge process (ASP), they are simple to be implemented and cost-effective. Thus, BNR methods have gained great popularity. However, complex, nonlinear and dynamic nature of biological and biochemical processes which takes place in these systems, makes controlling their performance a challenging and not straightforward task. Mathematical models provide a valuable evaluationdecision makingtools for scientists and wastewater engineers to move forward to the optimization of various wastewater treatment processes including BNR. Two to complicated nature of themodeledprocesses, application of WWTP. Since the full data collection required to reducecomplexityWWTPmodeling projects is an expensive and time-consuming task, a data scarcity is the prevailingproblem. Additionally, the path of the WWTP due to its operational and design conditions, the pathway through which it ismodeled is also unique, challenging, and worth investigating. This thesis presents a stepwise approach for the model-based optimization and practical scenario planning of the BNR activated sludge system in Castiglione Torinese WRRF in Italy. Results of the plant from January 2009 to December 2016, in addition to information from the limited number of measurements and sampling campaigns conducted during this study. Knowledge obtained was further integrated with the implementation of WWTP model. Model for active reactors, 1-D Takács model for secondary clarifiers, and aeration, pumping and mixing energy consumption models to be dynamically simulated its performances. The uncertainty of the calibrated models was investigated by the Monte Carlo Analysis.BasedThe RWS. For further investigation of the plant performance, a performance assessment criterion (PAC) consisting of two main types of parameters concerning the effluent quality and energy consumption and production. SRT values ​​(10, 15, 20, 25, 30, 35 and as a part of the optimization study, the impact of the solid retention time (SRT) 40). For scenario planning, results obtained from SRT scenario analysis were implemented. In general, two types of scenarios were considered for various operational modes. First, scenarios proposed to rectify the problem of change from the limits of the operational parameters. Second, practical scenarios as concurrent strategies implemented by plant operators. finally, these practical and problem-solving scenarios were studied by means of the parameters involved in the PAC. These promising model-based optimization solutions are proposed to plant managers for full-scale testing and implementation. This study highlights the role of data collection process, designed for model development and calibration purposes, in performance investigation and resolution of treatment units, which can be achieved by routinely conducted data collections and sampling campaigns in conventional WWTPs. The results obtained in this thesis, evidence the viability of the model-based optimization and problem-solving scenario planning even with the presence of data sacristy and data quality problems. The results confirm the possibility of increasing the plant in the energy savings and simultaneous improvement in pollutant removal, for Castiglione Torinese WRRF. In various steps of modeling and scenario planning, the importance of continuous interaction with the different stakeholders is highlighted. It is proved that this interaction can improve the acceptance of the proposed scenarios.