Towards intelligent decision making in power plant operation

Abstract

In a Power station, with multiple units and variety of control systems, management of alarms is the biggest challenge. The operator of each unit will be able to act upon the alarm depending upon the priority of the alarm message. There are two kinds of alarms. One process related alarms on which operator needs to act upon the disturbance. Second the equipment or system related alarm on which the operator needs to inform expertise maintenance person to act upon. But the appropriate corrective action depends upon the process experts who analyses the combination of alarms over time. The so called experts are different for each process, system and equipment and also they are distributed over different areas and locations. NTPC Simhadri is coastal based STPS in India having a capacity of 4x500MW units. This paper discusses about, how to capture expert's knowledge and integrate with the process alarm system for an expert advice and solution. This paper explains about designing and development of a common Expert alarm system integrating various platforms of Data acquisition systems of all four units. The expert's knowledge in the form of decision tree is used as the acquisition of structured knowledge in the form of concepts is introduced to build a knowledge base. This system enables the operator to take preventive measures in time, avoiding costly outages, in turn increasing the plant availability and efficiency. This system is user configurable, designed for continuous improvement and also available to any user in office LAN. The proposed system has been successfully implemented on visualization application environment and is written in Microsoft Visual Basic.