Font Size: a A A

Research On Dynamic Diagnosis Of Power System

Posted on:2006-04-05Degree:DoctorType:Dissertation
Country:ChinaCandidate:W P WangFull Text:PDF
GTID:1102360155461343Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
Dynamic diagnosis of power system is studied on the basis of current fault diagnosis methods to meet the need of power system. The dynamic diagnosis model based on hybrid system, the system structure based on multiagent, the knowledge-based diagnosis method, the fault analysis and early-warning method based on OPA (ORNL-PSerc-Alaska) /CASCADE model and hybrid diagnosis based on hybrid power system model are discussed to deal with complicated faults of power system with hierarchical dispatching mechanism. The main contributions of this dissertation are as follows: 1) The literatures about fault diagnosis of power system are investigated and summarized; the advantages and disadvantages of the main diagnosis methods are analyzed; and dynamic diagnosis of power system and dynamic analysis strategies are proposed according to the problems. Dynamic diagnosis of power system can be used to trace and diagnose faults online during their evolvement. Dynamic diagnosis is based on hybrid system and multiagent, and follows dynamic analysis strategies. 2) Considering power system is a hybrid system in nature and operates in a hierarchical mode, hybrid power system model based on Controlled General Hybrid Dynamical Systems (CGHDS) and multilayer diagnosis system based on multiagent are established. Hybrid system model can describe the continuous and discrete characteristics of the system and the relationship between them. The mutual operation and cooperation among Agents help solve the large scale and complicated problems in dynamic diagnosis system. 3) It is a main problem of knowledge-based diagnosis method to integrate shallow knowledge and deep knowledge because the knowledge of fault diagnosis is diversified. Formal expression of rule is established based on symbolic logic firstly. The knowledge-based diagnosis method, relevant theorems and rule indexing approach adaptive for fast diagnosis and knowledge evolvement online are established. Some examples are given to show the process of knowledge evolvement and the way to make use of rule index. 4) OPA/CASCADE model represents the nature of a type of cascading faults and has potential for online use. OPA model is modified to simulate the process of faults conveniently. And the iterative procedure for modified OPA model is also given. Simulating calculations are performed on RTS-96 in Matlab using Matpower, and the results of different redispatch approaches are compared. The results show this method is feasible for early-warning of the events which...
Keywords/Search Tags:Electric Power System, Blackout, Complicated Fault, Dynamic Fault Diagnosis, Hybrid System, Multiagent, OPA/CASCADE, Knowledge-based Diagnosis, Hybrid Diagnosis
PDF Full Text Request
Related items