Proposal of Usage Facility Data for Fire Risk Estimation on the Thermal Power Plants

In this paper, we propose data classification and analysis methods to estimate fire risk using facility data of thermal power plants. The data classification and analysis methods presented in this paper can provide indirect experience to data analysts who do not have domain knowledge about power plant fires and can also offer good inspiration for data analysts who need to understand power plant facilities. Since power plant facilities are designed according to operating conditions, generated data is modeled in three states, steady state, transient state, and anomaly state, considering that the data maintains a constant state if the operating conditions do not change. To estimate the fire risk using facility data, the risk level was expected for the narrowest area possible, and classification criteria by zone and facility were presented to satisfy the conditions that can be applied to the operating fire protection system. Experimental analysis was conducted on the data classified by the proposed method for 500 and 100 MW thermal power plants.