Calibration is a process of comparing model results with field data and making the appropriate adjustments so that both results agree. Calibration methods can involve formal optimization methods or manual methods in which the modeler informally examines alternative model parameters. The development of a calibration framework typically involves the following: (1) definition of the model variables, coefficients, and equations; (2) selection of an objective function to measure the quality of the calibration; (3) selection of the set of data to be used for the calibration process; and (4) selection of an optimization/manual scheme for altering the coefficient values in the direction of reducing the objective function. Hydraulic calibration usually involves the modification of system demands, fine-tuning the roughness values of pipes, altering pump operation characteristics, and adjusting other model attributes that affect simulation results, in particular those that have significant uncertainty assoc...

The paper describes a procedure for the robust design of water distribution networks which incorporates the uncertainty of nodal water demands and pipe roughness in a multiobjective optimization scheme aimed at minimizing costs and maximizing hydraulic reliability. The methodology begins with a deterministic system design in order to generate a set of optimal networks that serves as the initial population for subsequent multiobjective stochastic design. This approach does not depend on the choice of multiobjective optimizer (for example, a multiobjective genetic algorithm is used here) and can drastically reduce the number of “stochastic” runs needed for searching robust solutions. A collection of probability density functions based on the β function is introduced and applied to modeling variable uncertainty according to different physical requirements. The approach is tested in a case study involving a real network, illustrating its computational advantages.

Artificial neural networks (ANNs) are general-purpose techniques that can be used for nonlinear data-driven rainfall–runoff modelling. The key issue to construct a good model by means of ANNs is to understand their structural features and the problems related to their construction. Indeed, the quantity and quality of data, the type of noise and the mathematical properties of the algorithm for estimating the usual large number of parameters (weights) are crucial for the generalization performances of ANNs. However, it is well known that ANNs may suffer from poor generalization properties due to the high number of parameters and non-Gaussian data noise. Therefore, in the first part of this paper, the features and problems of ANNs are discussed. Eight Avoiding Overfitting Techniques are then presented, considering that these are methods for improving the generalization of ANNs. For this reason, they have been tested on two case studies—rainfall–runoff data from two drainage basins in the south of It...

The division of water distribution networks (WDNs) into district metering areas (DMAs) is a challenging issue and can be effective for analysis, planning, and management purposes. This contribution proposes a novel two-step strategy for DMA planning. The first step is optimal segmentation design by maximizing the WDN-oriented modularity index versus minimizing the number of conceptual cuts (i.e., not accounting for devices). The second step is the actual optimal DMA design, returning the positions of flow meters and closed valves at the conceptual cuts. Since closed valves change the system’s hydraulic paths, the implementation of DMAs could allow reductions in pressure and leakages through the WDN. Optimal DMA design is therefore achieved by solving a three-objective optimization minimizing the number of flow meters and total unsupplied customer demand while maximizing the reduction of background leakages; thus, pressure-driven modeling is mandatory. The effectiveness and flexibility of the procedure is demonstrated using the Apulian and Exnet networks. Results shows that DMA implementation can allow leakage reduction in systems with excess hydraulic capacity.

The economic and social costs of pipe bursts in water distribution networks (WDNs) are very significant. Water managers need reliable replacement plans for critical pipes, balancing investment with expected benefits in a risk-based management scenario. Thus, a robust and feasible decision support tool for water system rehabilitation is required. This kind of tool should incorporate (i) a model to forecast pipe failures and (ii) a strategy to solve a multi-objective optimisation problem trading investment vs. benefits. The former requires the collection of company asset data and the statistical modelling of pipe bursts. In this article, the burst modelling is performed by the evolutionary polynomial regression technique, providing a symbolic model for predicting pipe bursts. The benefits of burst reduction achieved by mains rehabilitation are evaluated by a multi-objective optimisation model over a short-term planning horizon (taken to be one year in this study). The multi-objective strategy is embedded in a genetic algorithm search methodology. The procedure identifies different subsets of pipes scheduled for rehabilitation, ranging from no-replacement (i.e., no reduction of the predicted number of bursts) to the complete replacement scheme (i.e. maximum reduction of the predicted number of bursts), trading cost of rehabilitation against achieved benefits. The result of the strategy is a Pareto (trade-off) front, which by itself does not provide any prioritisation of pipes for replacement. Thus, the article introduces a further processing step by which pipes are prioritised for rehabilitation based on the number of times each belongs to a solution on the Pareto front. By considering costs and such priority rating of each main, an improved investments/benefit diagram is constructed. The procedure is tested on a real-world UK WDN.