Modeling Of Data Correlation And Application Based On Gray System Theory

Grey system is a system containing both known information and unknown information. Grey system theory as a new method of system science theory, has its own unique advantages. The main task of the gray system theory is to propose a new modeling ideas and methods for a system not very clear and lack of overall information, from the perspective of control theory, and by analyzing the relationship between various factors and the amount of measure, use "gray data mapping" approach to deal with random values and found the law to make the system's development from ignorance to knowledge, known little to more, and the gray of system decrease, the whiteness gradually increase until the change law of the system understood. Grey system has been applied in many areas of social engineering since its birth.Grey incidence relational analysis is an important part of the grey system theory, but also the Grey system prediction, analysis and an important part of decision-making. The method proposed by some ways to seek the subsystem relations between the geometric values and the system provide a clear trend of the dynamic process of analysis. With the extensive application of grey system theory and communication, Grey relational analysis is also cause for concern.This article describes the concept of grey relational analysis and the application of the methods. This research work mainly in the following two aspects:First of all, considering the model's own shortcomings in theory and some of the limitations in practice, we improve the relative correlation of gray model according to the characteristics of the grey relational grade. The method of calculating the mean of sequence difference is adopted to reduce the impact of singular value, thus overcoming the defects that the existing system is vulnerable to the impact of singular value. The effect of resolution coefficient in the calculating of grey relational grade is discussed, also how the value of resolution coefficient is quantified and analyzed. One practical case was presented, which illustrates that this method not only increases the distribution range of the grey relational grade effectively, but also improves correlation resolution, making correlation analysis is more realistic.Secondly, we take the country's electricity demand as an example, consider the gross domestic product, the consumer level, fixed-asset investment, population, fiscal revenues and expenditures for science and technology as the 6 major factors affect to the demand of electricity, to build the indicators'model for the development of electricity. Use the three Grey association analyses as the method. Choose six national data from 1995 to 2006 to do the system's analysis to the development of China's electricity demand and its impact factors, and data analysis and comparison have done of its results. This method is simple and the result is reliable.