The invention relates to a power monitoring system based on a big data fusion model. The system is divided into three layers: a field device layer, a communication management layer and a system management layer; the system adopts the CoxPH algorithm based on deep learning to analyze the received data. Extracting eigenvalues can complete various practical functions such as switch status monitoring, analog quantity acquisition, comprehensive protection monitoring, remote control, alarm management, curve bar graph analysis, report generation, statistical printing, user management, accident recall, etc. Power reliability, improved power management efficiency, improved power supply quality, and reduced operation and maintenance costs.