Font Size: a A A

Multi-Agent Technology Research In Electric Power Decision Support System

Posted on:2008-04-23Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y WangFull Text:PDF
GTID:1102360212991372Subject:Systems analysis and integration
Abstract/Summary:PDF Full Text Request
Decision Support Systems (DSS) is an interactive software-based system intended to help decision makers compile useful information from raw data to identify and solve problems and make decisions. DSS begins with building model-driven DSS in the late 1960s, theory developments in the 1970s, and the implementation of financial planning systems and group DSS in the early and mid 80s. Data warehouses, OLAP and Business Intelligence evolved in the late 1980s and early 1990s.Nowdays intelligence decision support system (IDSS) is a new information system that joins artificial intelligence technology into decision support system.Agent is the independence intelligence software, which is a hot topic in the field of artificial intelligence. A multi-agent system consists of a number of agents, which interact with each other on behalf of each agent with different goals and motivations. To successfully interact, they will require the ability to cooperate, coordinate, and negotiate with each other.Electric power decision support system is widely used in electric power market, demand side management (DSM) and electric load forecasting. Although electric power DSS can be realized by object oriented programming technology, Agent Orientend Programming (AOP) isn't as in common use as OOP. The research on classical decision support system problem from Agent logic structure to Agent information system is deficiency. The research will be helpful in electric power decision support system. We need to use new Agent technology to realize the IDSS, maybe the technology will be widely used in advanced distributed electric power DSS as OOP in the future.The dissertation focuses on the theories and methods about software intelligence agent, multi-agent systems technology in intelligence DSS and in electric power DSS. The main contributions of this dissertation are as follows:Firstly, fuzzy reason logic on Agent is presented, by which paper deduced the procedure of agent getting knowledge. A feedback BDI-Agent (FBDI-Agent)model is put forward, which extends the classical BDI-Agent model.Secondly, a multi-agent information system of collection set according to information system is defined. Author researched into the procedure of generating rules during the multi-agent system and mined the rules in the instance using ID3 algorithm. The paper defines agent components relationship architecture. An agent-oriented programming has been implemented.Thirdly, three instances such as linear programming, multiple targets optimization and multiple linear regression analysis are selected from decision support system. Based on agent technology these instances can get results. Compared with their original results, the paper verified agent method is feasible in decision support system. Lastly, three instances such as electric price in electric power market framework, time of use (TOU) electric price and electric load forecasting are selected from decision support system. During the research on electric price in electric power market framework, multi-agent is included in building the framework and used to simulate the competition of electric power producer, electric power transmission and customer. During the research on time of use TOU electric price, Under the conditions of unchangeable electrical quantities, the optimal TOU electric price model based on multi-agent is used to make electric power consumers and electric power plants get economic benefit so as to flat the load curve. During the research on electric load forecasting, the electric load forecasting algorithm based on multi-agent is presented and is effective by results. Then an electric power load forecasting system is realized by multi-agent technology in several ways.The research in the paper is supported by Shanghai leading academic discipline Project: "electric power enterprise information and decision support system". The theory and techonolgy in the paper is used in this project.
Keywords/Search Tags:Artificial Inteligence, Decision Support, Multi -Agent, Electric Power
PDF Full Text Request
Related items