流域梯级水电站群及其互联电力系统联合优化运行

发布时间：2018-04-26 01:04

本文选题：梯级水电站群 + 发电调度　；参考：《华中科技大学》2014年博士论文

【摘要】：低碳经济模式下流域梯级水电站群及其互联电力系统中交织着各种物质流与信息流的映射关系,其联合优化运行受降雨、径流、用电、用水的不确定性以及各类电力能源发电特性和电网输电能力等因素影响,是一类大型的、多尺度、强耦合的非线性约束优化问题。传统的流域梯级水电站群联合调度模式由于存在长、中、短期调度难以衔接、水电站优化运行与火电站调度方式配合不紧、电力电量平衡相对孤立以及以经济效益为核心而很少考虑机组污染物排放等问题,已无法适应新时期流域梯级水电站群及其互联电力系统节能发电调度模式的实际工程需求,迫切需要研究新的理论与方法,以满足区域电力系统安全、经济运行和水电能源资源优化配置的需求。本文围绕流域梯级水电站群及其互联电力系统联合优化运行的若干关键问题,以梯级水电能源安全高效利用为目标,与系统科学理论、群体智能优化算法以及多目标进化优化技术相结合,对流域梯级水电站群及其互联电力系统联合优化运行的理论与方法开展系统、深入的研究,取得了一些具有基础理论方法和实际工程应用价值的研究成果。本文的主要研究内容和创新点包括： (1)针对传统优化方法难以解决的“维数灾”和容易陷入局部最优等问题,提出了一类基于自适应混沌人工蜂群算法的水库群优化调度方法。提出了改进的雇佣蜂算子、选择概率计算方法以及自适应改变率参数方法。该算法引入了基于logistic映射的混沌局部搜索策略,有效的提高了算法效率和寻优能力。将自适应混沌人工蜂群算法应用于标准测试函数和水电站群联合发电优化调度模型的求解中,结果表明本文提出的方法相比较传统方法能更有效的解决流域水电站群及其互联电力系统优化运行问题。 (2)围绕流域水电站群及其互联电力系统优化运行的时空变量多,约束条件复杂,优化目标多样等问题,本研究将人工蜂群算法引入多目标进化优化理论架构,提出基于Pareto优化的自适应多目标人工蜂群算法。算法采用小生境技术的外部群体空间保优策略,并提出基于逐步优化算法的局部搜索策略。将本文提出的自适应多目标人工蜂群算法进行函数测试,并应用于三峡梯级水电站长期发电优化调度,以及复杂水火电力系统多目标优化调度。结果表明本文提出的方法能有效解决调度目标冲突问题,一次性获得满足约束条件的Pareto优化解集。与文献中列出的方法进行比较表明,本文提出的方法能获得更广的Pareto优化前沿和更好的收敛效果。 (3)针对复杂水火电能源系统优化运行中水力电力联系嵌套约束问题,本文提出了一种多尺度循环修正约束处理方法。该方法将等式约束中的违反量采用多尺度循环修正方法分配到各个时段对应的变量中,有效实现了算法进化过程中不可行解向可行解的转化。将本文提出的约束处理方法应用于复杂水火电能源及其互联电力系统优化运行的模型求解,结果显示此方法能有效获得可行解。 (4)为解决径流随机性对优化调度的影响问题,提出了三峡入库流量的随机径流模拟方法,并采用机会约束方法研究长期优化调度建模问题。此外,为解决传统的流域梯级水电站群联合调度模式存在不同调度期,优化调度难以衔接的问题,提出了一种变长调度期嵌套优化方法。该方法将长期优化调度结果作为中期优化调度的输入条件,采用基于蒙特卡洛的人工蜂群算法进行问题的求解。将此方法应用于葛洲坝-三峡梯级水电站的中长期嵌套优化模型的求解中,结果表明本文提出的方法能获得在不同置信度下的精细化年内长期优化调度结果并有效描述约束违反风险与优化效益的对应关系,为指导实际生产运行提出了有意义的理论指导。
[Abstract]:In the low carbon economy model, the cascade hydropower stations and their interconnected power systems are interwoven with the mapped relationship between the material flow and the information flow. The combined optimization of the cascade hydropower stations is influenced by the factors such as rainfall, runoff, power consumption, water use uncertainty, power generation characteristics and power transmission capacity of various kinds of power. It is a large, multi scale and strong coupling. The combined scheduling model of the traditional cascade hydropower stations is difficult to link up because of the long, medium and short term scheduling. The optimal operation of the hydropower station is not fit with the dispatching mode of the thermal power station, the power and electricity balance is relatively isolated and the economic benefit is the core, and the pollutant emission is seldom considered. It is unable to adapt to the actual engineering requirements of the cascade hydropower stations and the energy saving power generation dispatching mode of the interconnected power systems in the new period. The new theories and methods are urgently needed to meet the needs of the regional power system security, the economic operation and the optimal allocation of the hydropower resources. The key problems of the system combined optimization operation are aimed at the efficient utilization of the cascade hydropower energy security and the system science theory, the swarm intelligence optimization algorithm and the multi-objective evolutionary optimization technology to carry out a systematic and in-depth study on the theory and method of the combined optimization operation of the cascade hydropower stations and their interconnected power systems in the basin. Some research results with basic theoretical methods and practical engineering applications have been obtained. The main contents and innovations of this paper include:
(1) aiming at the problem of "dimension disaster" which is difficult to solve in the traditional optimization method and easy to fall into the local optimal problem, a class of optimal scheduling method of reservoir group based on adaptive chaotic artificial bee colony algorithm is proposed. The improved employment bee operator, the selection probability calculation method and the self adaptable change rate parameter method are proposed. The algorithm is introduced based on L. The chaotic local search strategy of ogistic maps effectively improves the efficiency and optimization of the algorithm. The adaptive chaotic artificial bee colony algorithm is applied to the solution of the standard test function and the optimal scheduling model of the joint power generation of hydropower stations. The results show that the method proposed in this paper can be more effective in solving the hydropower station. The problem of optimal operation of groups and their interconnected power systems.
(2) in this study, the artificial bee colony algorithm is introduced into the multi-objective evolutionary optimization theory framework, and the adaptive multi target artificial bee colony algorithm based on Pareto optimization is proposed. The algorithm adopts the exterior of the niche technology. A local search strategy for group space optimization is proposed and a local search strategy based on progressive optimization algorithm is proposed. The adaptive multi-objective artificial bee colony algorithm proposed in this paper is used to perform function testing, and is applied to the optimization and scheduling of long-term power generation in the Three Gorges cascade hydropower stations, as well as the multi target optimal scheduling of complex water and fire power systems. The results show that the proposed method can be used in this paper. In order to solve the problem of scheduling target conflict effectively, the Pareto optimal solution set to satisfy the constraint conditions is obtained at one time. The comparison with the methods listed in the literature shows that the proposed method can obtain a wider range of Pareto optimization frontiers and better convergence effect.
(3) in this paper, a multi scale cyclic modified constraint processing method is proposed to solve the nested constraint problem in the optimization operation of the complex hydroelectric energy system. This method assigns the violation of the equality constraint to the corresponding variables in each period by using the multiscale cyclic correction method, and effectively implements the algorithm evolution process. The constraint processing method proposed in this paper is applied to the model solution of the optimal operation of complex hydro electric energy and interconnected power system. The results show that this method can effectively obtain a feasible solution.
(4) in order to solve the problem of the effect of runoff randomness on optimal scheduling, a stochastic runoff simulation method of the Three Gorges reservoir flow is proposed, and the opportunity constraint method is used to study the long-term optimal scheduling modeling. In addition, in order to solve the problem of different scheduling period for the traditional cascade hydropower station group joint scheduling model, the problem of optimal scheduling is difficult to link up. A nested optimization method of variable length scheduling period is proposed. This method uses the long-term optimal scheduling results as the input condition of the mid-term optimization scheduling, and uses the Monte Carlo based artificial bee colony algorithm to solve the problem. This method is applied to the solution of the middle and long term nested optimization model of the Gezhouba Dam Three Gorges cascade hydropower station. The method proposed in this paper can obtain the long-term optimal scheduling results under different confidence, and effectively describe the corresponding relationship between the risk of constraint violation and the optimization benefit, and put forward meaningful theoretical guidance for guiding the actual production operation.

【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TV737;TM732;TP18

【参考文献】