| With the rapid development of economy and the continuous improvement of living standards,the discharge of pollutants such as industrial wastewater and domestic sewage in the offshore area increases consecutively,leading to the increasing risk of seawater pollution in the offshore area.It is urgent to assess the risk of marine pollution scientifically and rationally.However,the incompleteness of seawater evaluation index and human subjectivity bring great challenges to risk assessment of seawater pollution.Although the classical fuzzy set theory has been used in risk assessment,it corresponds to the exact value on the unit interval,which is insufficient when dealing with missing data and human subjectivity.In comparison,the interval-valued fuzzy sets can deal with uncertainty and fuzziness better.In addition,aggregation function is not only a tool of information fusion but also widely used in fuzzy logic.Considered the advantages of interval-valued fuzzy sets and aggregation function,this dissertation combine the interval-valued fuzzy sets with aggregation function to establish a model of seawater pollution risk assessment based on the interval-valued implications,and discuss the sup-A reasoning(ACRI)method and quintuple implication principle(QIP)method based on IFMP and IFMT problems.The main work is as follows:1.Form the viewpoint of quasilinear space,we investigate the linear independence of interval vectors and dimension of quasilinear space<IRn,+,·>composed by interval vectors.Concretely speaking,regarded the interval as conjunctive set,the linear independence of interval vectors is firstly discussed.And the algorithm to check the linear independence of interval vectors is provided.Further,we study dimension of quasilinear space<IRn,+,·>.On the other hand,regarded the interval as disjunctive set,we discuss the weak,strong,tolerable and controllable linear independence of interval vectors,respectively.The algorithms to check the weak,strong,tolerable and controllable linear independence of interval vectors are given,too.2.To solve the interval-valued fuzzy modus ponens(IFMP)and interval-valued fuzzy modus tollens(IFMT)problems,the interval-valued sup-A compositional rule of inference(ACRI)and quintuple implication principle(QIP)based on the interval-valued residual implication generated by aggregation function are presented respectively.In order to develop these methods,we firstly study the property of interval-valued negation,residual implication generated by interval-valued aggregation function and(A,N)-implication.The solutions of ACRI and QIP methods for IFMP and IFMT problems are then provided.Finally,it is revealed that ACRI and QIP methods for IFMP and IFMT problems are recoverable.3.According to the data of dissolved oxygen and active phosphate from the coastal of Dalian City in 2020,a risk assessment model of seawater pollution based on the intervalvalued implication generated by aggregation function is established.Firstly,the K-means clustering method and fuzzy statistics method are used to establish the corresponding fuzzy sets,and the interval-valued fuzzy sets is constructed by combining the weak ignorance function.Secondly,according to the experience of experts,the fuzzy rule base is established,and the input interval-valued fuzzy sets are represented as the output interval-valued fuzzy set according to the inference rules.Further,according to the parameters of 7 mg/L、active phosphate concentration of 0.013 μmol P/L、0.019 μmol·P/L and 0.022 μmol·P/L,we obtain four interval-valued fuzzy sets for the risk of seawater pollution as the inference rules.The output interval-valued fuzzy sets are then obtained using the ACRI methods based on the interval-valued implication as mentioned before.Finally,an exact value is obtained by using the type reduction method,which represents the risk index of seawater pollution.In the reasoning process,we use five kinds of aggregation operators to compare and analyze the risk level of seawater pollution. |
PDF Full Text Request