一库多级式梯级库群短期优化调度及并行计算研究
发布时间：2018-07-28 16:23
【摘要】：作为优质清洁能源,水电具有显著的调峰调频优势,在我国近十年来得到了空前发展。为了最大限度地利用水头优势,增发水电电量,我国梯级流域特别是中小流域大多采用一库多级式开发方式,这些梯级电站群的特点是以调节性能较好(如多年、年、季调节)、库容较大的大型水电站做为龙头水电站,下游往往具有一级或多级高水头小库容、调节能力较差的水电群。梯级电站优化调度过程中,依靠龙头电站的补偿调节作用,一库多级式梯级可以在不增加工程规模的条件下显著增加整个梯级的保证出力和发电量,同时合理的调度方式可以使梯级电站具有更好的调峰调频效果,显著改善电源质量。研究一库多级式梯级电站群优化调度方式对电网安全、稳定、高效运行具有重要的意义。然而一库多级梯级短期优化调度面临计算电站数目较多、上下游电站间存在严格的水力电力联系及复杂的流量滞时关系等问题,使得计算求解异常复杂。结合实际工程,本文从梯级滞时关系推求、给定复杂约束的优化调度及通过并行提高求解效率等方面进行深入研究,主要内容包括：(1)上下游电站间流量滞时是影响梯级电站群短期优化调度中水量计算及调度结果合理性的重要因素。阐述了上下游电站滞时关系的常规推求方法,以马斯京根法为基础,以流量动态滞时为目标,构建了三段流量滞时关系曲线模拟模型,同时采用最小二乘法对模型参数进行求解,以沙溪流域安砂与贡川电站梯级的丰平枯实际资料进行计算、分析和验证,计算结果表明该方法的合理性和实用性。(2)一库多级式水电站群短期优化调度是研究调度期较短时间内逐时段各水电站最优蓄放水策略与负荷分配方案,以达到梯级总效益最大的目标。该方案有利于协调电网和梯级电站之间的关系,维持电网的安全、稳定运行。在充分分析一库多级式梯级电站之间水力电力联系的基础上,通过考虑峰谷电价因素,分别建立了给定负荷过程的梯级蓄能最大模型和基于分时电价的梯级发电效益最大模型,利用POA对模型进行求解。以沙溪流域梯级电站群为例进行了仿真计算,结果表明了模型的合理性及方法的有效性。(3)粒子群优化算法是求解梯级电站群优化调度问题的经典优化方法,但是在求解大规模电站群短期优化调度时,计算量较大导致计算耗时过长,无法满足实际需求。利用Fork/Join多核并行框架设计实现了一种基于粒子群算法的并行计算方法对一库多级式梯级电站群短期发电量最大模型进行优化求解。以沙溪梯级电站群为例进行了仿真计算,结果表明了多核并行粒子群算法可以显著提高计算效率,大幅度缩减计算耗时,为一库多级式梯级库群短期优化调度的求解提供了技术参考。最后对一库多级式梯级电站群短期优化调度并行算法研究进行了总结和展望。
[Abstract]:As a high quality clean energy, hydropower has the advantage of peak shaving and frequency modulation, and has been developed unprecedented in recent ten years. In order to maximize the advantages of water head and increase the power supply of hydropower, most of the cascade watersheds in China, especially in the middle and small watersheds, adopt the multi-stage development mode of one reservoir. The characteristics of these cascade power stations are that they have better regulating performance (for example, for many years). The large hydropower stations with larger reservoir capacity are the leading hydropower stations, and the lower reaches often have a small reservoir with one or more high water heads and a poor regulating capacity. In the process of optimal dispatching of cascade power stations, depending on the compensation and regulation function of the leading power stations, the multi-stage cascade of a reservoir can significantly increase the guaranteed output and power generation of the whole cascade without increasing the scale of the project. At the same time, the reasonable dispatching mode can make the cascade power station have better peak-modulation effect and improve the quality of power supply. It is of great significance to study the optimal dispatching mode of a multistage cascade power station group for the safe, stable and efficient operation of the power network. However, the number of power stations is large, the relationship between upstream and downstream power stations is strict and the time delay of flow is complex, which makes the calculation and solution very complicated. Combined with the actual engineering, this paper makes a deep research on the relationship between the cascade delay and time delay, the optimal scheduling of given complex constraints, and the parallel improvement of the efficiency of the solution, and so on. The main contents are as follows: (1) the lag time of flow between upstream and downstream power stations is an important factor that affects the calculation of water quantity and the rationality of dispatching results in the short-term optimal operation of cascade power stations. In this paper, the conventional calculation method of time-lag relationship of upstream and downstream power stations is described. Based on Muskinggen method and the target of flow dynamic time-lag, a simulation model of three-stage flow time-lag curve is established. At the same time, the model parameters are solved by the least square method, and the actual data of sand in Shaxi basin and Fengping and withered in Gongchuan Hydropower Station are calculated, analyzed and verified. The calculation results show that the method is reasonable and practical. (2) Short-term optimal dispatching of multi-stage hydropower station group in a reservoir is to study the optimal water storage and discharge strategy and load distribution scheme for each hydropower station in a short period of time in a relatively short period of time. In order to achieve the maximum overall benefit of the step goal. This scheme is helpful to coordinate the relationship between grid and cascade power station, and to maintain the safety and stability of power grid. On the basis of fully analyzing the relationship between hydraulic and electric power of a multistage cascade power station, the maximum model of cascade energy storage for a given load process and the maximum benefit model of cascade generation based on time-sharing price are established by considering the factors of peak and valley electricity price. The model is solved by POA. The simulation results show that the model is reasonable and the method is effective. (3) PSO is a classical optimization method to solve the optimal scheduling problem of cascade hydropower stations. However, in solving the short-term optimal dispatching of large scale power station group, the large amount of calculation results in the time-consuming calculation, which can not meet the actual demand. A parallel computing method based on particle swarm optimization (PSO) is designed and implemented by using Fork/Join multi-core parallel framework to solve the short-term maximum power generation model of a multistage cascade power station group. Taking Shaxi cascade hydropower station as an example, the simulation results show that the multi-core parallel particle swarm optimization algorithm can significantly improve the computational efficiency and greatly reduce the computation time. It provides a technical reference for solving the short-term optimal scheduling of a multi-level cascade library group. In the end, the parallel algorithm for short-term optimal dispatching of a multi-stage cascade hydropower station group is summarized and prospected.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TV697.12;TV737
本文编号：2150869
[Abstract]:As a high quality clean energy, hydropower has the advantage of peak shaving and frequency modulation, and has been developed unprecedented in recent ten years. In order to maximize the advantages of water head and increase the power supply of hydropower, most of the cascade watersheds in China, especially in the middle and small watersheds, adopt the multi-stage development mode of one reservoir. The characteristics of these cascade power stations are that they have better regulating performance (for example, for many years). The large hydropower stations with larger reservoir capacity are the leading hydropower stations, and the lower reaches often have a small reservoir with one or more high water heads and a poor regulating capacity. In the process of optimal dispatching of cascade power stations, depending on the compensation and regulation function of the leading power stations, the multi-stage cascade of a reservoir can significantly increase the guaranteed output and power generation of the whole cascade without increasing the scale of the project. At the same time, the reasonable dispatching mode can make the cascade power station have better peak-modulation effect and improve the quality of power supply. It is of great significance to study the optimal dispatching mode of a multistage cascade power station group for the safe, stable and efficient operation of the power network. However, the number of power stations is large, the relationship between upstream and downstream power stations is strict and the time delay of flow is complex, which makes the calculation and solution very complicated. Combined with the actual engineering, this paper makes a deep research on the relationship between the cascade delay and time delay, the optimal scheduling of given complex constraints, and the parallel improvement of the efficiency of the solution, and so on. The main contents are as follows: (1) the lag time of flow between upstream and downstream power stations is an important factor that affects the calculation of water quantity and the rationality of dispatching results in the short-term optimal operation of cascade power stations. In this paper, the conventional calculation method of time-lag relationship of upstream and downstream power stations is described. Based on Muskinggen method and the target of flow dynamic time-lag, a simulation model of three-stage flow time-lag curve is established. At the same time, the model parameters are solved by the least square method, and the actual data of sand in Shaxi basin and Fengping and withered in Gongchuan Hydropower Station are calculated, analyzed and verified. The calculation results show that the method is reasonable and practical. (2) Short-term optimal dispatching of multi-stage hydropower station group in a reservoir is to study the optimal water storage and discharge strategy and load distribution scheme for each hydropower station in a short period of time in a relatively short period of time. In order to achieve the maximum overall benefit of the step goal. This scheme is helpful to coordinate the relationship between grid and cascade power station, and to maintain the safety and stability of power grid. On the basis of fully analyzing the relationship between hydraulic and electric power of a multistage cascade power station, the maximum model of cascade energy storage for a given load process and the maximum benefit model of cascade generation based on time-sharing price are established by considering the factors of peak and valley electricity price. The model is solved by POA. The simulation results show that the model is reasonable and the method is effective. (3) PSO is a classical optimization method to solve the optimal scheduling problem of cascade hydropower stations. However, in solving the short-term optimal dispatching of large scale power station group, the large amount of calculation results in the time-consuming calculation, which can not meet the actual demand. A parallel computing method based on particle swarm optimization (PSO) is designed and implemented by using Fork/Join multi-core parallel framework to solve the short-term maximum power generation model of a multistage cascade power station group. Taking Shaxi cascade hydropower station as an example, the simulation results show that the multi-core parallel particle swarm optimization algorithm can significantly improve the computational efficiency and greatly reduce the computation time. It provides a technical reference for solving the short-term optimal scheduling of a multi-level cascade library group. In the end, the parallel algorithm for short-term optimal dispatching of a multi-stage cascade hydropower station group is summarized and prospected.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TV697.12;TV737
【引证文献】
相关期刊论文 前1条
1 叶其革;张岚;樊冬梅;;智能变电站多层数据交换调度优化方法研究[J];自动化与仪器仪表;2017年08期
，本文编号：2150869
本文链接：https://www.wllwen.com/kejilunwen/shuiwenshuili/2150869.html
