Research On Energy Consumption Prediction Of Equipment Manufacturing Enterprise Under The Background Of Big Data

IT has a great development in recent years, which constantly permeate all areas of economic, social and life, in this process of continuous innovation, the birth of a series of emerging technologies, such as mobile Internet, cloud computing and networking technology. Under continuous development and application of these innovative technologies, new applications of different patterns strongly promote the use of human information scope and forms of development, such as virtual services, social networking and collaboration to optimize manufacturing. This innovative approach to development based on the production of information and the Internet, so that the human society has entered a "third industrial revolution" period of development.Due to continuous innovation of information technology and new applications continue to show patterns, the amount of data the explosive worldwide expansion, IDC(International Data Corporation) in its research report released by the digital universe, which pointed out that the data following the 2011 global total after exceeding 1.8ZB, in 2020 the total amount of data in worldwide is expected to reach 35 ZB, the growth rate of approximately doubling every two years, following the new Moore’s Law. The amount of data growing process, characterized by its complex complexity, diversity, real-time, also become increasingly significant, "big data" era has arrived.Based on the review of relevant research results, based on, first, to the current research and application on large data were described and analyzed in detail, including the concept and meaning of big data, big data processing mode and basic processing and big data aspects of core technology.Secondly, with the current data and the actual background of the large case of CM’s field research, conducted a feasibility analysis of the status quo and energy equipment manufacturing companies to predict the next big data background and made a big data model based on analysis of data collected works and technical system architecture design.After that, the current equipment manufacturing enterprises manufacturing process energy consumption status were analyzed according to the relevant literature summarizes the energy equipment manufacturing enterprise manufacturing process consumes influencing factors as the type of energy after a predictive model of total factor input variables, and by CM conduct field research, access to raw data corresponding factors.Finally, based on support vector machine and equipment manufacturing companies to build energy forecasting model, using an improved three-step search method as a support vector machine parameter optimization method, the energy consumption of equipment manufacturing enterprises conducted regression analysis to predict and After different forecasting methods and parameter optimization methods are horizontal comparison, results show that support vector machine regression to predict the best method for predicting the effect of the improved three-step search method is best for its parametric search results.