Kevin Lansey

Reliability/availability analysis of water distribution systems considering adaptive pump operation

Water distribution systems (WDS) are designed for delivering high quality water to consumers at desired pressures. Significant works have been conducted to measure and efficiently evaluate the performance and behavior of WDS. Although several literatures presented some methodologies for appraising WDS reliability/availability, few studies look at the impact of pump and tank operations while assessing WDS performance. In practice water utilities would take actions to respond emergency, such as pipe breaks, summer peak demand by turning on more pumps in short-term or maintaining a high water level in storage tank in long-term to meet the system’s pressure requirement. Therefore, to represent this reality, pump and tank operations must be addressed in WDS reliability analysis and potentially optimal design. Herein a practical WDS reliability analysis methodology is presented. The model takes into account pump operation schemes to quantify WDS reliability/availability that is defined as the probability that water distribution system can supply its consumers’ demand over a certain period. Monte Carlo stochastic simulation is conducted to hypothetically generate nodal demands and component failure. EPANET is employed as a hydraulic solver to estimate system pressure. Applications to mid-sized sample network show the adaptive pump operations as responding to pipe breaks improve system resiliency with trading in additional moderate pumping cost.

SUSTAINABILITY INDICATORS FOR LONG-TERM WATER SUPPLY: CASE STUDIES OF TUCSON ACTIVE MANAGEMENT AREA

Meeting growing water demand is a significant concern in Arizona. Potential future water shortage caused by climate changes and droughts in the Colorado River basin have raised the awareness to save the water whenever it is possible to support sustainable water supply in the future. In order to manage the finite groundwater resources in Arizona, five areas relying on mined groundwater were identified and designated as Active Management Areas (AMAs). The primary management goal of the AMAs is so-called “safe-yield”, which is accomplished when no more groundwater is being withdrawn than is being annually recharged. In general, groundwater overdraft is used as an indicator to measure sustainability, which is simply the difference between the annual amount of groundwater withdrawn and the natural and artificial recharge in the basin. Although a water budget measure provides a broad overview of a water usages and supplies in a geographically delineated area, it is clear that more specific indicators are needed to help those who are engaged in evaluating the alternative water management plans with respect to sustainability of supplies. This study focuses on developing water resource sustainability indicators. Then a wide range of water conditions and management scenarios will be evaluated using the developed measures based on the data and projection of Tucson Active Management Area (TAMA).

Development of a Cost Function for Residential Subdivisions through Genetic Algorithms

The study of urbanization, and the water distribution networks that supply water to urban areas, has led to new approaches for evaluating the distribution systems that serve residential subdivisions. The present study aims to expand the evaluation method commonly applied by using Genetic Algorithms to add the actual hydraulic constraints required by Tucson Water in Arizona to an optimization model. Furthermore, an alternative calculation method utilizing a heuristic pre-optimization tool coupled with a greedy algorithm is compared to the genetic algorithm results. The improved model is capable of achieving a near optimal solution, one that is comparable to the genetic algorithm results and should minimize the cost of constructing and operating a water distribution network. Preliminary results show that population density has little effect on the total cost and that area is the driving factor in cost. In addition, the slope increases the rate at which these 2 parameters increase the cost, making high density areas much more cost effective with respect to the operation of water distribution systems. Finally, the main assumption, which considers residential subdivisions as rectangular networks, is explored by comparing the generated networks against their realistic counterparts. Results showed that the realistic networks cost more than the generated networks.

Enhancing artificial neural networks applied to the optimal design of water distribution systems

Achieving an optimal design for a typical water distribution system (WDS) essentially involves determining which combination of pipes and arrangements will produce the most efficient and economical network. Solving the problem is a complex process, one well suited to computationally intensive heuristic methods. Including water quality constraints can pose a special challenge due to the demanding, extended-period simulations involved. Employing artificial neural networks (ANNs) can reduce the amount of computation time needed. ANNs can in fact approximate disinfectant concentrations in a fraction of the time required by a conventional water quality model. This study presents a methodology for improving the accuracy of ANNs applied to the optimal design of a WDS by means of a probabilistic approach based on the fast finding of a network similar to the optimal WDS. This work also presents a methodology to find such a network. ANNs trained with the probabilistic dataset generated using the proposed approach were shown to be more accurate than their counterparts trained with a random dataset.

Post-optimization heuristics complementing the design of real water distribution systems

Adaptive search methods are often used to design urban water distribution networks when the number of pipes in the network is insignificant. For complex, realworld networks, however, such methods are computationally demanding and they have difficulty finding near-global optima. To identify a solution as close to the global optimum (and in which no pipe can be reduced without violating pressure constraints), requires a high-speed computer potentially running for a long time and also probably some good fortune. This work presents a methodology for refining the solutions found by adaptive search algorithms used in the design of large waterdistribution networks. The approach employs two heuristics to search for an optimal combination of pipes that, after a reduction of their diameters, will maximize cost savings while continuing to meet design constraints. The post-optimization approach presented here is shown to be an efficient complement to heuristic search algorithms used in the design of real-world networks.

Bad data processing for water distribution system demand estimation

Previous studies by the authors have shown that nodal demands in a water distribution system (WDS) can be estimated in real-time using pipe flow data collected by supervisory control and data acquisition (SCADA) system. Estimated demands can be used for optimal operation of system to support pressure and water quality. It is not unusual the data from SCADA systems contain gross errors due to system failure and/or meter malfunctions. The estimator is sensitive to these erroneous measurements and the estimates based on the bad measurements are not reliable for system operation; therefore bad data should be filtered prior to demand estimation. However, system failure and meter malfunctions are random phenomena and hard to identify. This study presents a series of statistical methods to detect bad data, identify their locations, and correct the data values. The proposed methods are based on a linear measurement model that linearly relates state variables (nodal demands) to the field measurements (pipe flow rates). The scheme is applied prior to a demand estimation to eliminate the effects of erroneous data on the demand estimates. The proposed method is applied to a hypothetical simple network using synthetically generated data sets, such as error-free data, Gaussian-noisy data, fire flow data, and noisy data containing one or more contaminated measurements. Application to a simple hypothetical network using synthetically generated data shows that the method can be successfully used as a pre-processing for single and multiple non-interacting bad data for reliable demand estimation.

SEQUENTIAL ESTIMATION OF DEMAND AND ROUGHNESS IN WATER DISTRIBUTION SYSTEM

Pipe roughness and water consumption at a demand node are the most uncertain input variables in a simulation model because they are not typically directly measurable. Instead, information provided from field measurements are used to estimate them indirect way. Parameter estimation is the process of adjusting model parameters so that the simulation model represents the real system adequately by fitting the model output to the field data. To provide more accurate estimates and account for all associated uncertainties, the two variables, i.e., demand and roughness, must be estimated simultaneously. This study proposes a two-step sequential method for dual estimation of demand and roughness coefficient based on a weighted least squares (WLS) scheme using field measurements of pipe flow rates and nodal pressure heads under multiple demand loading conditions. The algorithm is applied to a simple hypothetical system using synthetically generated field data. The proposed two-step sequential model provides accurate estimates with little effort in terms of simulation time.