Abstract：Distribution network public transformer is a key equipment in the power system, and its quantity is very large. Due to power demand fluctuations and other conditions,public transformer’s load is in the great changes. In order to?evaluate each public transformer more accurately,this paper introduces the big data analysis platform based on Hadoop CDH,then proposes a method of public transformer load condition evaluation based on electricity big data.The method determines the fuzzy judgment matrix by historical statistics data,calculates weight by this matrix and overcomes the defects of subjective factors when using comprehensive evaluation method.After evaluating the condition of heavy overload,light load and load fluctuation,it calculates the weighted sum of each condition,and finally determines the load status of each public transformer.The process is clear, and the calculation result is accurate and has an important supporting decision-making role in the field of precision investment,operation,maintenance and post-investment assessment.