Pre-Dispatch of Load in Thermoelectric Power Plants Considering Maintenance Management Using Fuzzy Logic

Abstract

This paper presents a new method for load pre-dispatch considering the technical conditions of engines in thermoelectric power plants by combining several maintenance and diagnostic techniques and using computational intelligence. A diagnosis of the technical conditions of the engines is performed using a lubricant analysis, vibration analysis, and thermography. With these data from a statistical analysis, it is possible to predict when an engine will fail and to consider this phenomenon in the load pre-dispatch. To increase the engine reliability and power supply, a maintenance management program is developed using MANAGEMENT tools, applying only 4 total productive maintenance pillars and combining them with predictive maintenance and diagnostics, thus reducing failures in plant equipment. Some results achieved after this implementation are as follows: a reduction in the annual cost of maintenance, a reduction in the corrective maintenance, an increase in the mean time between failures, and a decrease in the mean time to repair in all areas. In addition, the pre-dispatch ensures that the demanded power is met with a high degree of reliability and quality, and at minimal cost.