Study On Short-Term Optimal Operation Of One-reservoir And Multi-cascade Hydropower System And Its Parallel Computing

As high quality clean energy, hydropower has significant advantage of peak load and frequency regulation, resulting in a rapid development in the past decade in China. In order to take full use of water head to produce more hydropower generation, one-reservoir and multi-cascade hydropower system becomes the typical development way of medium and small basins in China. In this way, there is a leading hydropower plant upstream with good regulation performance and large capacity, and downstream there are cascade hydropower plants with high water level, small capacity and poor regulation performance. For short-term optimal operation of hydropower systems, the exploitation way of one-reservoir and multi-cascade hydropower system can significantly increase the firm output and annual generation capacity according to the compensation regulation performance of leading hydropower plant. Simultaneously, good scheduling can make hydropower systems has a better peak load and frequency regulation and significantly improve the quality of power supply. Study on optimal operation of one-reservoir and multi-cascade hydropower system has a great significance to the safe, stable and efficient operation of the power grid in China. However, the optimal operation of cascade hydropower systems faced obstacles of large number of hydropower plants, strict hydraulic and power relationship between upstream and downstream plants as well as complicated water time delay of flow, resulting in high challenge. Combine with practical engineering, this paper made a deep research on ascertain of water time delay of flow, optimal operation with complex constraints and efficiency improvement of solving, the main contents include:(1) Water time delay of flow between upstream and downstream reservoirs is an important factor that affects water balance calculation and reliability of scheduling results of hydropower system short-term optimal operation. It discussed the influence of water time delay of water flow on short-term optimal operation of hydropower systems in detail, as well as conventional solution method of water time delay of flow. Then, Muskingum model is utilized to formulate water time delay of flow between upstream and downstream reservoirs, and least square method is used to estimate the model parameters. Finally, real flow data of Ansha and Gongchuan plants in Shaxi basin is used to formulate, analysis and verify, the results demonstrate the reliability and practicability of the method.(2) Short-term optimal operation of one-reservoir and multi-cascade hydropower system is to obtain optimal strategy of water supply and load distribution of each power plant in a relatively short time during water supply period to obtain maximum benefits. This scheme is helpful to maintain the coordination relationship between power grid and hydropower system as well as the safe and stable operation of power grid. Based on detail analyzed the hydraulic and electric relationship between cascade hydropower plants, cascade storage energy maximum model with total load is given and generation benefit maximum model based on time-of-use price are developed respectively, and POA is used to solve the presented models. Finally, simulation calculation is made using hydropower systems in Shaxi basin, the results verified the rationality of the model and effectiveness of the method.(3) Particle swarm optimization algorithm is a most intelligent optimization technique to solve the optimal operation problem of hydropower system. However, when the scale of hydropower system reaches a certain large degree, the amount of calculation and time consuming of using particle swarm optimization algorithm increase greatly that cannot meet requirement. Therefore, a multi-core parallel particle swarm optimization algorithm based on Fork/Join framework is developed to optimize optimal operation of hydropower system. The obtained result of hydropower system in Shaxi River demonstrates that the proposed parallel algorithm can significantly improve the computational efficiency as well as reduce the time consuming.Finally, summarized and prospected the parallel algorithm study of short-term optimal operation of one-reservoir and multi-cascade hydropower system.