I. Irizar

Model-based optimisation of Wastewater Treatment Plants design

This paper presents the mathematical basis and some illustrative examples of a model-based decision-making method for the automatic calculation of optimum design parameters in modern Wastewater Treatment Plants (WWTP). The starting point of the proposed methodology is the mathematical modelling of the main processes inside a plants units. The procedure for the automatic calculation of the design parameters is then based on expressing the optimum WWTP design problem as a Mathematical Programming (Optimisation) Problem that can be solved using a non-linear optimisation algorithm (GRG2). The paper shows how the proposed methodology is able to achieve optimum WWTP design using either a steady-state or dynamic mathematical model of the plant and a set of constraints associated with the permitted operational ranges and the required water quality in the effluent. As an illustrative example to show the usefulness of the proposed methodology, the optimum design of the Step-Feed process for nitrogen removal (Alpha) has been analysed by considering two different problems: the optimum plant dimensions, estimated at critical temperature for effluent requirements (Problem 1), and the optimum selection of facultative volumes, fractions of the influent flow-rate and the values of oxygen set-points for long-term plant operation (Problem 2). The proposed decision-making method is intended to facilitate the task of the engineers involved in the design of new WWTP, especially when the complexity of the plant requires a systematic procedure for the selection of the main design parameters.

Optimal predictive control of water transport systems: Arrêt-Darré/Arros case study

This paper proposes the use of predictive optimal control as a suitable methodology to manage efficiently transport water networks. The predictive optimal controller is implemented using MPC control techniques. The Arrêt-Darré/Arros dam-river system located in the Southwest region of France is proposed as case study. A high-fidelity dynamic simulator based on the full Saint-Venant equations and able to reproduce this system is developed in MATLAB/SIMULINK to validate the performance of the developed predictive optimal control system. The control objective in the Arrêt-Darré/Arros dam-river system is to guarantee an ecological flow rate at a control point downstream of the Arrêt-Darré dam by controlling the outflow of this dam in spite of the unmeasured disturbances introduced by rainfalls incomings and farmer withdrawals.

Benchmarking of control strategies for ATAD technology: a first approach to the automatic control of sludge treatment systems.

Autothermal Thermophilic Aerobic Digestion (ATAD technology) is a promising alternative to conventional digestion systems. Aeration is a key factor in the performance of these kinds of reactors, in relation to effluent quality and operating costs. At present, the realisation of automatic control in ATADs is in its infancy. Additionally, the lack of robust sensors also makes the control of these processes difficult: only redox potential and temperature sensors are reliable for operation in full-scale plants. Based as it is on the existing simulation protocols for benchmarking of control strategies for wastewater treatment plants (WWTP), this paper presents the definition and implementation of a similar protocol but specifically adapted to the needs of ATAD technology. The implemented simulation protocol has been used to validate two different control strategies for aeration (ST1 and ST2). In comparison to an open-loop operation for the ATAD, simulation results showed that the ST1 strategy was able to save aeration costs of around 2-4%. Unlike ST1, ST2 achieved maximum sludge stabilisation but at the expense of higher aeration costs.

On-line estimation of suspended solids in biological reactors of WWTPs using a Kalman observer

The total amount of solids in Wastewater Treatment Plants (WWTPs) and their distribution among the different elements and lines play a crucial role in the stability, performance and operational costs of the process. However, an accurate prediction of the evolution of solids concentration in the different elements of a WWTP is not a straightforward task. This paper presents the design, development and validation of a generic Kalman observer for the on-line estimation of solids concentration in the tank reactors of WWTPs. The proposed observer is based on the fact that the information about the evolution of the total amount of solids in the plant can be supplied by the available on-line Suspended Solids (SS) analysers, while their distribution can be simultaneously estimated from the hydraulic pattern of the plant. The proposed observer has been applied to the on-line estimation of SS in the reactors of a pilot-scale Membrane Bio-Reactor (MBR). The results obtained have shown that the experimental information supplied by a sole on-line SS analyser located in the first reactor of the pilot plant, in combination with updated information about internal flow rates data, has been able to give a reasonable estimation of the evolution of the SS concentration in all the tanks.

Standard signal processing using enriched sensor information for WWTP monitoring and control.

Important indicators for monitoring and control of wastewater treatment plants (WWTP) often have to be obtained from the processing of on-line signal trajectories. Therefore, the quality of sensor instantaneous measurements can be improved significantly if they are complemented with valuable information about the geometric features of their trajectories. The present paper describes the design and implementation of a Standard Signal Processing Architecture (SSPA) from which enriched sensor information is generated automatically. The SSPA has been made up of three complementary modules: the pre-processing module, the storage module and the post-processing module. Moreover, the SSPA has been parameterised so as to allow its adaptation to the specifications of every signal. By performing basic calculations on pre-processed signal trajectories, the storage module produces enriched vectors which collect information of the first and second time derivatives, average and variance values, peak values, linear regression parameters, curvature, etc. Then, the enriched information vectors can be exploited to implement customised monitoring and control tools. In this respect, the effectiveness of the SSPA has been demonstrated in three different practical cases: (1) OUR and KLa identification algorithms; (2) processing of measurements for real-time controllers; and, (3) detection of bend-points in on-line signals of SBR processes.

AqquaScan: design and implementation of an Internet-based service for the remote monitoring and management of decentralised WWTPs

This paper describes the design and implementation of AqquaScan, an Internet-based service for remote monitoring and integrated management of decentralised WWTPs. AqquaScan is a multi-user and multi-WWTP service. It has been built according to criteria such as flexibility, scalability and interoperability with the idea of providing an open environment suited to quickly accommodate future scenarios (e.g. incorporation of new plants or upgrading of existing installations). Both, the management of plant information and users interfaces have been implemented in distributed software components that communicate with one another via web services. The implemented web services can be exploited to develop customised user interfaces for visualising the monitored data. By default, a customised web-based client module has been programmed in order for users to be able to exploit the facilities offered within AqquaScan: (1) real-time monitoring of on-line signals; (2) visualisation of historical data; (3) changing operational parameters; (4) notification of time-event information; and (5) storage of measurements from laboratory analysis. At present, AqquaScan is fully operative and is offering supervision services to eleven industrial WWTPs distributed around Northern Spain.

From sewage sludge and agri-food waste to VFA: Individual acid production potential and up-scaling

Volatile fatty acid (VFA) production through anaerobic fermentation may constitute an innovative solution for organic waste management within the context of circular economy. In the present study, the evolution of individual VFA during laboratory-scale fermentation of sewage sludge (SS), winery wastewater (Www) and meat and bone meal (MBM) was assessed, focusing on the effect of pH (5.5 and 10) and temperature (35 and 55 °C). Up-scaling of the fermentation process was evaluated in batch operation. The latter showed that specific VFA could be produced, giving similar individual evolution to lab-scale testing. To be precise, acetic acid percentage ranged within 30-65% and increased up to 5900 mg O2 L-1 during SS fermentation at 55 °C and pH 9. In addition, 60% butyric acid was reached during Www acid fermentation at 55 °C, which corresponded to 6670 mg O2 L-1 concentration in the fermentation broth. Regarding valeric acid, over 20% proportion and 2700 mg O2 L-1 were reached in MBM acid fermentation at 35 °C. Finally, iso-valeric maximum level ranged within 15-17% in SS alkaline fermentation at 55 °C, which represented a concentration close to 2000 mg O2 L-1. Interestingly, co-fermentation of agri-food waste and SS at thermophilic temperature and alkaline pH, boosted the VFA concentration 1.7-2 fold, which suggests that anaerobic co-fermentation of substrates from different nature could give promising outcomes in full-scale operation.

Robust tuning of bending-points detection algorithms in-batch-operated processes

The literature shows a diversity of real-time algorithms for automatic detection of bending-points in batch-operated waste treatment systems. In this study a new methodology is proposed for tuning the parameters of these algorithms when uncertainty specifications are considered at the outset. In this method the effects of slow and fast dynamic responses on the shape of signal trajectories were treated separately in order to cover via simulation all possible operating scenarios for a real situation. Long-term uncertainty and steady-state simulations were combined to derive probability distributions for biomasses. These probability distributions were then merged with short-term uncertainty to run one-cycle random simulations with which to reproduce the entire diversity of signal trajectories. Finally, an optimisation problem was formulated in terms of the algorithm parameters. The methodology was satisfactorily applied to tune an algorithm for detection of bending-points in an Autothermal Thermophilic Aerobic Digestion (ATAD) process. Methodology for tuning of bending-point detection algorithms in batch-operated systems.Application of the method to a specific bending-point detection algorithm for ATAD technology.Short and long-term uncertainty to describe the population of both slow and fast response states.Steady-state and one-cycle simulations to generate realistic signal trajectories.

Online monitoring of OUR, KLa and OTE indicators: practical implementation in full-scale industrial WWTPs

Based on on/off aeration strategies, this paper describes all the steps involved in the development and implementation of three identification algorithms aimed at monitoring the oxygen uptake rate (OUR), the oxygen mass-transfer coefficient (K(L)a), and oxygen transfer efficiency (OTE) in aerated biological reactors. Firstly, a detailed explanation of the theoretical background behind every algorithm is given. In addition, practical issues have also been taken into account in order to guarantee the quality of estimations. Finally, the three algorithms have been implemented and validated in a full-scale industrial wastewater treatment plant with satisfactory results. Although short-term noise has been observed in the estimated data (especially at high OURs), the medium and long-term data trajectories have been correctly reproduced.