Optimization study on competence of power plant using gas/steam fluid material parameters by machine learning techniques

Abstract

The power plant significance is to envisage the complete load electric power production of a permanent load on a power supply to increase the yield from the accessible megawatts hours (MW hrs). This power plant productivity may rely upon conservation variables like pressure, temperature and humidity. Here, the preliminary way to increase overall efficiency can be precise by the combination of two thermodynamic cycles powered by gas and steam that reduces fuel costs also. In this proposal, machine learning techniques such as principal component analysis for reducing dimensions in the dataset where data points are plotted and K-Means, agglomerative for clustering method to predict the cluster for each data point, finally calculating the cluster center also. By statistical analysis, statistics of complete dataset can be done through features such as ambient pressure, relative humidity, ambient variable temperature, exhaust vacuum, power output. The foremost aspire of power plant is to accomplish overall efficiency carried through each and every components comes under power plant. These tremendous precise forecasts produce an upgrade production inventory that overestimates effectiveness and productivity of power station.