Integrating wind, photovoltaic, and large hydropower during the reservoir refilling period

Hydropower facilities are an ideal solution to complement the intermittent production of energy from wind and solar photovoltaic facilities in electric power systems. However, adding this task to the multiple diverse duties of a reservoir (e.g., flood mitigation, water supply, and power generation) poses a challenge related to pursuing multiple and sometimes conflicting objectives. This study proposes an approach for integrating hydro, wind, and photovoltaic power during a reservoir's refill period. Specifically, this approach simultaneously minimizes the fluctuation in the combined power output of these three resources and maximizes their combined power generation while adhering to the target reservoir's water levels. The proposed approach uses a multiobjective optimization model that prescribes a day-ahead optimal hourly operation for a hydropower facility in terms of spilled water, water stored in the reservoir, and water used for power generation, while meeting a daily target to refill the reservoir. The prescribed scheduling is then used as the input into a model that simulates the actual operations of the power system. This study focuses on a hydro-wind-photovoltaic system located in southwestern China, where the peak power generating capacity of the hydropower facility is ten percent larger than the combined installed capacity of the wind and solar power. The results show that by using the proposed model, the hydropower facility effectively smooths the fluctuations in the combined power output caused by variable wind and photovoltaic power and concurrently meets the reservoir replenishing targets under dry, moderate, or wet hydrologic scenarios. Furthermore, the trade-offs between power generation maximization and power fluctuation reduction were found to depend on two conditions: whether the reservoir is full, and whether the turbine is generating electricity at its maximum capacity. The hydro-wind-photovoltaic integration is more cost-effective when the reservoir is not full and the turbines are not generating electricity at their maximum capacity. When the reservoir is full, hydropower still has the ability to balance the wind and photovoltaic power without curtailment but tends to result in water spillage (22–402 m3/s) and reductions in electricity generation (0.1–11.4 GWh per day). The proposed method for scheduling operations allows hydropower facilities to complement wind and photovoltaic power output, while meeting the target water levels during the refill period.

Duke Scholars

Author Dalia Patino-Echeverri Environmental Sciences and Policy