Relaxation Techniques for Optimisation of Irrigation Systems as Flexible Energy Storage

Flexible energy storage systems are indispensable for the integration of renewable energy sources in modern power systems. We propose large irrigation systems as an innovative solution to provide energy storage services, while maintaining their irrigation role. This work focuses on the optimisation of these systems. To upgrade the system to provide these services, we formulated a sizing and scheduling optimisation as a MINLP problem. However, non-linearities and binary variables inherent in hydraulics systems negatively affect the convergence and execution time. We present and test relaxation techniques for the pumped storage hydropower optimisation problem and compare different solving algorithms in real case scenarios. Relaxations improve the convergence of the solution, reduce execution times up to one order of magnitude and have close to no effect on the objective value. The ability to find a solution still depends on the initialisation values, so further development is required.