CAJ | 학술논문

针对电网故障诊断中，警报信息存在的不确定性和不完整性，提出了一种粗糙集结合时间贝叶斯网络的故障诊断方法。本方法利用粗糙集理论的知识约简和处理不确定信息的能力，分层挖掘故障警报信息，对多个电网区域的决策表进行属性优选和诊断规则提取。再利用时间贝叶斯网络对各个区域并行推理，实现故障诊断。经过算例仿真，验证了该方法诊断快速准确、容错能力强、应用灵活。
침대전망고장진단중，경보신식존재적불학정성화불완정성，제출료일충조조집결합시간패협사망락적고장진단방법。본방법이용조조집이론적지식약간화처리불학정신식적능력，분층알굴고장경보신식，대다개전망구역적결책표진행속성우선화진단규칙제취。재이용시간패협사망락대각개구역병행추리，실현고장진단。경과산례방진，험증료해방법진단쾌속준학、용착능력강、응용령활。
To deal with the uncertainties and imperfections of alarm information after fault, this paper proposes a fault diagnosis method based on rough nets and temporal Bayesian networks for power networks. Using the ability of rough nets to reduce knowledge and process indeterminate information, the hierarchical mining of substation’s fault diagnosis knowledge is carried out. Then the fault decision table of divided power network areas is optimized, and diagnose rules are extracted. Then fault diagnosis is finished through parallel reasoning of Bayesian networks. The test proves that this method can diagnose the fault rapidly and accurately, and has strong fault tolerance and adaptability.