[Paper Review] Management strategies for run-of-river hydropower plants -an optimal switching approach

Niklas L. P. Lundström, Marcus Olofsson+1 more|arXiv (Cornell University)|Sep 22, 2020

Water resources management and optimization42 references4 citations

TL;DR

This paper proposes an optimal switching framework for run-of-river hydropower plants using stochastic differential equations to model river flow and partial differential equations to solve variational inequalities for production planning. The method generates fully automatic, cost-aware operational strategies that achieve near-optimal performance—within 2% and 5% of theoretical optimum—while accounting for switching costs and forecasted flow variations.

ABSTRACT

The mathematical theory for optimal switching is by now relatively well developed, but the number of concrete applications of this theoretical framework remains few. In this paper, we bridge parts of this gap by applying optimal switching theory to a set of production planning problems related to hydropower plants. In particular, we study two different cases involving small run-of-river hydropower plants and show how optimal switching can be used to create fully automatic production schemes in these cases, with non-zero cost of switching between different states of production. Along the way of deriving these schemes, we also create a model for the random flow of water based on stochastic differential equations and fit this model to historical data. This stochastic flow model, which should be of independent interest, mimics the long term seasonal behaviour of the flow while still allowing for stochastic fluctuations and can incorporate a given forecast to damp the impact of such fluctuations in near time. We benchmark the performance of our model using actual flow data from a small river in Sweden and find that our production scheme lies close to the optimal, within 2 % and 5 %, respectively, in a long term investigation of the two plants considered.

Motivation & Objective

• To develop a mathematically rigorous, automatic production planning strategy for run-of-river hydropower plants with non-zero switching costs.
• To model stochastic river flow dynamics that capture seasonal trends and short-term fluctuations, incorporating real-time forecasts.
• To apply optimal switching theory—previously underutilized in energy systems—to real hydropower operations with empirical validation.
• To benchmark the proposed strategy against naïve heuristics and assess performance under varying forecast horizons and switching costs.
• To demonstrate the scalability and practical feasibility of the method for real-world deployment in small hydropower plants.

Proposed method

• Formulates a stochastic differential equation (SDE) model for river flow that captures long-term seasonality and short-term stochastic fluctuations.
• Calibrates the SDE model to historical flow data from a Swedish river, enabling realistic simulation of flow dynamics.
• Integrates forecast data into the SDE framework to reduce the impact of short-term flow uncertainty on decision-making.
• Models the hydropower plant’s operational states (e.g., on/off) as a switching control problem governed by a variational inequality.
• Solves the resulting high-dimensional PDE system numerically using a finite difference scheme to derive optimal switching thresholds.
• Implements a numerical algorithm that computes optimal switching strategies in minutes on standard hardware, enabling real-time or near-real-time application.

Experimental results

Research questions

• RQ1 Can optimal switching theory be effectively applied to real-world run-of-river hydropower production planning with non-negligible switching costs?
• RQ2 How does the performance of the optimal switching strategy compare to a simple on/off heuristic strategy under varying flow conditions and forecast horizons?
• RQ3 To what extent can a stochastic flow model incorporating forecasts improve the robustness and accuracy of hydropower production decisions?
• RQ4 How sensitive is the optimal strategy to changes in switching cost relative to electricity price (C/D ratio)?
• RQ5 Can the proposed framework be extended to more complex hydropower systems, such as those with storage or pumped-storage capabilities?

Key findings

• The proposed optimal switching strategy achieves performance within 2% and 5% of the theoretical optimum in long-term simulations for two distinct run-of-river plants.
• The method significantly outperforms a naïve on/off strategy, especially as the number of production modes increases, due to better handling of switching costs.
• Incorporating forecasts improves performance, though longer forecasts can occasionally underperform shorter ones due to increased uncertainty in rapid flow changes.
• The model remains robust even when the electricity price is held constant; extension to time-varying or stochastic prices is theoretically feasible with no structural limitations.
• The computational cost is low—strategy computation takes only a few minutes on a standard laptop—making it suitable for real-time deployment.
• The framework is generalizable to higher-dimensional problems, though computational complexity increases with the number of stochastic sources, suggesting a need for alternative methods (e.g., Monte Carlo) in high-dimensional settings.