Interval-valued Fuzzy Inference And Its Application In Assessment Of Island Resources

Fuzzy reasoning has been widely utilized in information processing,data mining,decision making,image processing,expert systems and so on.Interval-valued fuzzy sets with more degrees of freedom are more suitable for data mining and decision making environments.In order to better apply the fuzzy inference system in practice,it is a great significance to systematically investigate the interval-valued fuzzy inference mechanism and the approximation mechanism.In recent years,with the acceleration of ZhouShan Islands New Area's process,the activities of island development in ZhouShan have gradually intensified.Therefore,a scientific assessment of island resources will be a scientific basis for strengthening ZhouShan Island protection,carrying out island planning and strengthening island management.Due to the lack of data or the uncertainty and ambiguity of available data,and the lack of awareness of many mechanisms,it becomes a challenging task to assess the complex systems of island resources.Taking into account the numerous uncertainties and human behavioral factors,it has an incomparable advantage than the classical method with the interval fuzzy set theory in the corresponding evaluation model.For this reason,this thesis mainly studies on the fuzzy inference mechanism and its application in evaluation of island resources.The results obtained in the following:1.The QIP solutions of FMP and FMT for R-?S-?QL-?f-and g-implications are firstly presented.And then the robustness of certain specific fuzzy conjunction QIP method is studied.Moreover,we discuss the universal approximation properties of multiple-input and single-output(MISO)fuzzy systems with QIP method for R-?S-and QL-implications.These results reveal that the QIP method possesses better performance in fuzzy rule-based system.2.We firstly extend the concept of admissible order on the set of intervals and study some properties of admissible order vector space.Employing an admissible order and interval OWA operator,the MOILP problem is transformed into a weighted sum scalarization optimization problem.Furthermore,we reduce the MOILP problem into solving several real-valued linear programs and obtain its solution.In addition,the W-efficient solutions and axial solutions of the MOILP problem are introduced to account for decision makers' preferences as partial information of the MOILP problem.3.With the relevant results about multi-objective linear programming,taking Zhoushan Island as some examples,three practical problems are assessed according to the information provided by the decision makers.These results consistent with the actual phenomena.This implies that it is feasible in practical application to assess the island resources with interval-valued multi-objective linear programming.