Padula, S; (2015) Capacity expansion modelling to aid water supply investment decisions. Doctoral thesis , UCL (University College London).

Preview

Text SILVIA PADULA - PhD THESIS.pdf Download (5MB) | Preview

Abstract

Increasing population, economic development, and environmental changes imply that maintaining the water supply-demand balance will remain a top priority. Water resource systems may need to be expanded in order to respond to demand growth. Capacity expansion studies can be used to answer the question of what the optimal expansion size, timing and location of new infrastructure should be. This thesis develops and applies capacity expansion optimisation modelling approaches. We begin with the 'Economics of Balancing Supply and Demand’ (EBSD) planning framework used by the water industry since 2002 in England. The base model is formulated as a mixed integer linear programming optimisation model that selects the least cost annual schedule of supply and demand management options that meet forecasted demand over the planning horizon. Custom water saving profiles are allowed for demand management options. Multiple demand scenarios are considered to ensure the supply-demand balance is preserved under different demand conditions and that operating costs of selected options are accurately assessed. The base deterministic EBSD model is applied to the water companies of South East England (the WRSE area). Various extensions to the EBSD framework are then proposed and implemented. The model formulation is first expanded to incorporate a generic cost estimate for options not yet proposed in water company resources management plans. This allows to extend the WRSE network with new inter-company transfers for which costs are represented by a concave cost curve approximated by a piece-wise linear function. Considering additional interconnections allows evaluating the financial implications of further interconnectivity in the WRSE area. Next, an extension is proposed to improve the application of the stochastic version of the EBSD approach. The proposed method allows to identify the set of future capacity expansions that withstand uncertainty of supply and demand estimates and still achieve a required reliability. The method consists of an iterative process: at each iteration the EBSD optimisation model is run and the reliability of the solution set (supply-demand schedules) is tested under Monte Carlo simulation. Ad-hoc model constraints are introduced at each iteration to enable the EBSD model to exclude unreliable solutions identified at previous iterations. Next, the English price-cap regulatory process is represented within a modified EBSD model formulation. The model identifies future capacity expansions that maximise water company profit under constraints on the maximum price that can be charged to customers and on the allowed rate of return. The incentive schemes that the regulator uses to reward (or penalise) companies for out- (or under-) performance, are also represented. The goal is to help explain how the current regulatory system of incentives motivates water company investment decisions. Two further extensions are then presented in the appendixes. The first one allows the EBSD model formulation to be extended so that costs of activated schemes are accounted over the schemes’ useful life, beyond the typical 25-30 year planning horizon. This eliminates biased comparisons of schemes with different economic lifetimes. With the second extension, a diverse set of supply-demand schedules are generated, that solve the capacity expansion problem and are sufficiently ‘close’ (in terms of costs) to the least-cost solution. Generating multiple near-optimal solutions gives an idea of what alternative plans are available in addition to the leastcost one. This allows the consideration other un-modelled factors or strategic priorities in the decision making process.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item