This paper proposes a new data mining model of energy consumption for regional industry, which is used? to solve the problem that there is no effective method on data analysis in regional energy monitor. First, it analyses the characteristics of regional energy consumption in different industries in k-means, clusters a single industry and finds out the high, middle, low energy consumption industrial enterprises. Then, multidimensional association rules are used to find the correlation of processes, equipments and energy efficiency to guide the energy conservation in regional energy monitor. The algorithm has been improved on the basis of apriori algorithm to find the frequent predicate set of specific product's processes and equipment. The analysis of the example shows the model is effective and feasible which provides a new way for energy saving of regional industry.