基于多Agent的梯级泵站优化调度系统研究

发布时间：2018-03-14 18:44

本文选题：梯级泵站优化调度　切入点：Agent　出处：《兰州理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：梯级泵站被广泛应用到区域内及区域间的调水工程中,但是,由于梯级泵站在运行过程中各个泵站的流量、扬程密切相关,加之各泵站机组台数不一,输水路线长,沿途区间分水多等原因,使得梯级泵站在运行管理上存在较大困难,在实际运行中往往由于决策不当,造成能源大量浪费,效益低下,这一问题也被许多专家所重视,对梯级泵站系统调度进行优化,实现节能的前景非常可观。在原有的梯级泵站优化调度的理论基础上,将多Agent理论引入梯级泵站的优化调度中,将Agent的自主性、协调性等特点应用到该系统中,建立了一个基于多Agent的梯级泵站优化调度系统,以提高泵站的运行效率,降低运行能耗。主要研究内容有： (1)对梯级泵站的基础理论和国内外的研究现状进行了介绍,并根据梯级泵站调度过程中存在的一些问题,提出了本文的主要研究内容、整体思路及创新点。 (2)对梯级泵站优化调度的方法、基本思想以及多Agent理论做了详细的介绍,进而建立了基于多Agent的梯级泵站优化调度系统模型。该系统由泵站多Agent子系统、算法多Agent子系统以及一个管理Agent和一个通信Agent构成,管理Agent和通信Agent用来协调两个多Agent子系统共同完成系统的任务。 (3)通过分析梯级泵站优化调度的特点及多Agent系统的通信协调的类型特点等,确定了本系统的协调模型及通信语言。对泵站子系统与算法子系统两个子系统之间的通信协调以及各子系统内部的通信协调进行了介绍,确定了系统完成任务的详细步骤。 (4)将多Agent理论与遗传算法相结合,建立算法多Agent子系统,采用并行微种群遗传算法对数据进行处理。并对梯级泵站历史数据进行优化处理,通过结果对比,本系统可有效降低系统运行时的能耗,相应的提高了梯级泵站的经济效益。
[Abstract]:Cascade pumping stations are widely used in regional and inter-regional water transfer projects. However, due to the close relationship between the discharge and lift of each pumping station during the operation of the cascade pumping station, the number of units in each pumping station is different, and the water conveyance line is long. There are many problems in the operation and management of cascade pumping stations due to the large water distribution along the road, which is often due to improper decision making in actual operation, resulting in a large amount of energy waste and low efficiency. This problem has also been paid attention to by many experts. Optimization of cascade pumping station system scheduling, energy-saving prospects are very impressive. On the basis of the original theory of optimal dispatching of cascade pumping stations, the multiple Agent theory is introduced into the optimal dispatching of cascade pumping stations. The autonomy and coordination of Agent are applied to the system. A cascade pumping station optimal dispatching system based on multiple Agent is established in order to improve the operation efficiency of the pumping station and reduce the running energy consumption. The main research contents are as follows:. This paper introduces the basic theory of cascade pumping station and the present research situation at home and abroad. According to some problems existing in the dispatching process of cascade pumping station, it puts forward the main research contents, overall ideas and innovation points of this paper. 2) the method, basic idea and multi-#en0# theory of optimal dispatching of cascade pumping station are introduced in detail, and then the optimal dispatching system model of cascade pumping station based on multiple Agent is established. The system is composed of multi-#en2# subsystem of pumping station. The algorithm is composed of multiple Agent subsystem, a management Agent and a communication Agent. The management Agent and the communication Agent are used to coordinate the two multi-#en5# subsystems to complete the task of the system. 3) by analyzing the characteristics of optimal dispatching of cascade pumping stations and the type of communication coordination of multi-#en0# system, etc. The coordination model and communication language of the system are determined. The communication coordination between the two subsystems of pump station subsystem and algorithm subsystem and the communication coordination within each subsystem are introduced, and the detailed steps to complete the task of the system are determined. 4) combining multiple Agent theory with genetic algorithm, the multi-algorithm Agent subsystem is established, and the parallel micropopulation genetic algorithm is used to process the data, and the historical data of cascade pumping station are optimized, and the results are compared. The system can effectively reduce the energy consumption while the system is running, and improve the economic benefits of the cascade pumping station.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TV675

【参考文献】