The Dissipative Structure Theory And Study Of Spatiotemporal Data Processing

Geographic information system (GIS) that has been developed and improved for half a century is widely used in various fields of social economy. With the innovation and wide application of GIS, and the rapid development and integration of relevant information collection systems, GIS will integrate various kinds of information in geospace (even in a wider rage) and have gigantic database. For the purpose thereof, people should resolve two problems: for one thing, how to save, organize and manage these data; for another, how to make full use of these data to serve production, life and scientific research. GIS will be transformed from database to analytic type.At present, the rate of utilization of GIS data is low, and research on GIS lags behind the research on spatial data structure and the technology of spatial database. With the development, evolution and improvement of GIS, GIS database is not only applied in enquiry, checking, drawing or ordinary administration affairs, but more importantly in extracting information of data, analyzing mechanism and regularities of things trough analyzing and processing of GIS data. Therefore, GIS will not only become an important tool in policymaking but also a data resource and processing platform for scientific research.Dissipative Structure Theory can be summed up as follows: a kind nonlinear open system which is far from equilibrium, with constant flux of energy and matter, which produces nonequilibrium phase change through fluctuation when a particular parameter reaches certain threshold, and changes into ordered spatial state from chaotic state, and has certain functions. The basic principle of Dissipative Structure Theory is widely applied in various systems including physical, chemical and social economic systems, and is the summary of the law of changes of these systems.Dissipative Structure Theory, a science of self-organizing system, is, to a certain extent, a technology to trace systematic information (spatiotemporal data). Matter is unity of quality and quantity, and people reveal the internal mechanism and development lawof matter through analysis of quantity. The combination of Dissipative Structure Theory and GIS provides a possibility of analyzing mechanism of matter (system), upon which the Spatiotemporal Data Processing method proposed in this dissertation is based.At present, the spatial analyst method is adopted, which indeed helps to process GIS data analysis to a certain extent, but GIS is still not utilized to the fullest extent. Conventional GIS spatial analyst only adopts ordinary topological processing, transformation or "Elementary Deduction" on data layer. The relatively complicated processing method could only generally and superficially analyze spatiotemporal data and cannot reveal mechanism of matter. Conventional statistic analysis of signaling system analyzes the objective as dependent variable (several independent variable components), which though covers mechanism analysis to a certain extent, is incapable of analyzing complex systems, and incapable expressing characteristics of special distribution.This dissertation summarizes the basic contents and ideology of the Dissipative Structure Theory, and demonstrates the mathematical model and the construction process of dissipative structure by describing in a mathematical way and deducing certain phenomenon of self-organization.Then the thesis makes a further research and conclusion on mathematical theories and method concerning spatiotemporal data processing. The relationship between stochastic process and partial differential equation has proved the rationality of the diffusing equation in nature, which is the theoretical foundation of establishing diffusing equation model of continuum model reaction. Finite difference method is the most effective and common numerical method to solve the parabolic partial differential equation and an effective numerical computation method to GIS spatiotemporal data processing due to the similarity between its discrete data formation and GIS data stru...