Scenario Based Cost Optimization of Water Distribution Networks Powered by Grid-Connected Photovoltaic Systems

TL;DR

This paper introduces a scenario-based predictive control approach for optimizing water distribution networks powered by grid-connected photovoltaic systems, significantly reducing electrical costs and increasing PV energy contribution.

Contribution

It develops a stochastic scenario optimization framework with a day-ahead PV prediction method, improving cost efficiency in water networks with renewable energy integration.

Findings

PVs supplied 66.95% of the energy in the tested network.

The stochastic optimization outperformed deterministic methods.

Significant reduction in electrical costs achieved.

Abstract

The paper presents a predictive control method for the water distribution networks (WDNs) powered by photovoltaics (PVs) and the electrical grid. This builds on the controller introduced in a previous study and is designed to reduce the economic costs associated with operating the WDN. To account for the uncertainty of the system, the problem is solved in a scenario optimization framework, where multiple scenarios are sampled from the uncertain variables related to PV power production. To accomplish this, a day-ahead PV power prediction method with a stochastic model is employed. The method is tested on a high-fidelity model of a WDN of a Danish town and the results demonstrate a substantial reduction in electrical costs through the integration of PVs, with PVs supplying $66.95\%$ of the required energy. The study also compares the effectiveness of the stochastic optimization method with a deterministic optimization approach.