[4P1-OS-17a] OS-17

[4P1-OS-17a-02] Demand-Supply Planning Optimization with Market Transactions and Diverse Power Sources

An Approach That Achieves Practical Computation Time and Over 1 Billion Yen in Annual Revenue Improvement

〇Yoshiharu Kohji¹, Ryosuke Umeda¹, Keita Takahashi¹ (1. GRID inc.)

Keywords:Mathematical Optimization, Supply and Demand Planning, Electricity trading market, Market Trading, Power Transmission and Distribution Planning

In this study, we address the demand-supply planning problem in the electric power sector using mathematical optimization, and report on a case of its application to an weekly planning process. Our method creates multiple scenarios based on forecasts of weather, demand, and electricity prices, and computes the startup and shutdown schedules of generators for each scenario.Within our company’s supply area, not only do we incorporate our own thermal, hydro, pumped-storage, and nuclear power generation, but we also integrally model solar power and purchased electricity from external companies. Each power plant is modeled in detail to reflect the operational constraints required for real scheduling.We adopt a two-stage approach to solve this problem. In the first stage, we determine the most economically rational operational plan for each power plant and power source, taking electricity prices into account. In the second stage, we adjust each generator’s output to balance overall supply and demand while also factoring in sales to external regions.Thanks to this two-stage approach, the overall model is greatly simplified, enabling the derivation of generator startup and shutdown plans within computation times that satisfy practical requirements. In an actual deployment, applying this optimization method has achieved annual revenue improvements of over one billion yen