SCADA Based Operator Support System for Power Plant Fault Diagnosis

摘要

Supervisory Control and Data Acquisition Systems (SCADA) play an important role in the monitoring and control of power generation plants. Whenever faults occur in any part of the power plant, critical alarms are generated by SCADA system. This paper discusses on the assessment of such faults using a set of rules encoded with human expert knowledge. Proposed system uses a search algorithm on SCADA history alarm database which is driven by the knowledge base rules. The expert system components are developed for fault diagnosis and for operator guidance. Rules and Meta rules for knowledge induction and backward chaining inference engine are used for fault diagnosis. The encoded knowledge helps in searching the SCADA alarm database to arrive at the root cause of the fault occurred. The decision support system provides step by step procedure to be followed by an operator to restore the system to normal operating conditions. The tedious manual processing involved in analysing SCADA history alarm database containing vast amount of information can be reduced by using the proposed operator support system. The explanation facility and structured methodology followed by the hybrid intelligent system makes fault diagnosis an effortless and efficient process. The system is implemented using Java Expert System Shell(JESS) in Eclipse platform.