Blockages are a major issue for wastewater utilities around the world, causing loss of service, environmental pollution, and significant cleanup costs. Increasing telemetry in combined sewer overflows (CSOs) provides the opportunity for near real-time data-driven modelling of wastewater networks. This paper presents a novel methodology, designed to detect blockages and other unusual events in the proximity of CSO chambers in near real-time. The methodology utilises an evolutionary artificial neural network (EANN) model for short-term CSO level predictions and statistical process control (SPC) techniques to analyse unusual level behaviour. The methodology was evaluated on historic blockage events from several CSOs in the UK and was demonstrated to detect blockage events quickly and reliably, with a low number of false alarms.

This paper proposes a new method to identify the near-optimal response to failures in water distribution networks in near real time. The response method is formulated as a two-objective optimization problem with objectives being the minimization of failure impacts and related operational costs. The new heuristics-based method is developed and used to solve this optimization problem. The method comprises three steps. In the first step, the initial list of available interventions is identified offline. In the second step (online), the narrowed-down list of interventions considered in the optimization is identified. Finally, in the last step (online), a novel heuristic algorithm is applied to identify near-optimal solutions in near real time. The new optimization method was validated and demonstrated in two case studies, a semireal case study based on a C-Town network and an assumed failure event (pipe burst), and a real UK case study involving a complex real pipe network and event caused by shutting down the Water Treatment Works. The Pareto front of response interventions identified by the new heuristics method approximates well the non-dominated sorting genetic algorithm II Pareto front in both cases with the largest differences measured in terms of end-impacts (between relevant solutions for the same cost) being 4% and 9%, respectively. In addition, the new heuristics method is able to identify near-optimal response solutions in a computationally fast manner (15 min and 1 h for the two cases). Therefore, the heuristics method can be used in near real time in real-life situations.

Combined Sewer Overflows (CSOs) are a major source of pollution and urban flooding, spilling untreated wastewater directly into water bodies and the surrounding environment. If overflows can be predicted sufficiently in advance, then techniques are available for mitigation. This paper presents a novel bi-model committee evolutionary artificial neural network (CEANN) designed to forecast water level in a CSO chamber from 15 min to 6 h ahead using inputs of past/current CSO level data, radar rainfall data and forecast forecasted rainfall data. The model is composed of two evolutionary artificial neural network (EANN) models. The two models are trained and optimised for wet and dry weather conditions respectively and their results combined into a single response using a non-linear weighted averaging approach. An evolutionary strategy algorithm is employed to automatically select the optimal artificial neural network (ANN) structure and parameter set, allowing the network to be tailored specifically for different CSO locations and forecast horizons without significant human input. The CEANN model was tested and evaluated on real level data from 4 CSOs located in Northern England and the results compared to three other ANN models. The results demonstrate that the CEANN model is superior in terms of accuracy for almost all forecast horizons considered. It is able to accurately forecast the dry weather and wet weather level, predicting the timing and magnitude of upcoming spill events, thus providing information that is of clear use to a wastewater utility.

The present study proposes a new interactive methodology and an interactive tool for the response to water network failure events facilitating near real-time decision-making. The proposed methodology considers: (i) a structured yet flexible approach supporting and guiding the operator throughout the entire response process to water network failure events, while allowing the operator to have a final say; (ii) a novel interaction with the operator in near real time via the proposed tool (e.g. allowing operators to propose different 'what-if' scenarios without being hydraulic experts); (iii) the provision of automatically generated advice (e.g. optimal response solutions and assessed end-impacts) - although optimal response solutions not identified in near real time yet; and (iv) improved impact assessment using realistic impact indicators that cover different aspects of the event - which are consistently calculated for every proposed response solution (to facilitate easy comparison between different response solutions). The new methodology was applied on a semi-real case study. The results obtained demonstrated the potential of the new response methodology and its application through the interactive tool to improve water utilities' current practice. This was accomplished through supporting/guiding operators in the identification of effective response solutions with low end-impact on the consumers and low cost for the utility.

Near-real-time event detection is crucial for water utilities to be able to detect failure events in their water treatment works (WTW) quickly and efficiently. This paper presents a new method for an automated, near-real-time recognition of failure events at WTWs by the application of combined statistical process control and machine-learning techniques. The resulting novel hybrid CUSUM event recognition system (HC-ERS) uses two distinct detection methodologies: one for fault detection at the level of individual water quality signals and the second for the recognition of faulty processes at the WTW level. HC-ERS was tested and validated on historical failure events at a real-life UK WTW. The new methodology proved to be effective in the detection of failure events, achieving a high true-detection rate of 82% combined with a low false-alarm rate (average 0.3 false alarms per week), reaching a peak F₁ score of 84% as a measure of accuracy. The new method also demonstrated higher accuracy compared with the CANARY detection methodology. When applied to real-world data, the HC-ERS method showed the capability to detect faulty processes at WTW automatically and reliably, and hence potential for practical application in the water industry.