Study On The Fuzzy Cluster Method Of Microtrip Based On The Passenger Car Dirving Cycles

With the increasing of the number of passenger car and the performance demand of passenger car, It is required that the passenger car factory must manufacture all kind of passenger car to satisfy all kind of people's need. Each area has its own traffic condition, so the passenger car driving cycles is different. The data of traffic needs to be analyzed and combined to reflect the passenger car driving cycles of each area.Passenger car driving cycles is mainly applied on the evaluation of the passenger car .Making accurate evaluation on the fuel economy and emission level of the passenger car or not depends on whether the driving cycles can reflect the real condition of the passenger car under certain traffic condition. And whether the evaluation on the fuel economy and emission of the passenger car is reasonable or not will directly affect the fuel economy and emission optimization of the passenger car. Through the analyzing and processing of the experiment data, the clustering microtrip database is generated in this paper, which contains the basic driving data, such as velocity, acceleration, driving time, and the traffic condition, etc.It is the foundation for designing the passenger car driving cycle under different driving condition conveniently and quickly. This paper takes the experiment data of passenger car driving on the real road as study object, the statistical theory is used to establish the original microtrip database and clustering microtrip database which explain the real driving condition of the passenger cars, and the combine and design of driving cycles is also used to validate the validity of the clustering of microtrip database. The following content is studied in this paper.(1) Establishing the microtrip original database. This section mainly introduces the analysis processing of the experiment data which is obtained from the driving experiments in the city of Changchun, the transition of the format of the data and the establishment of the original microtrip database. First of all, the unit of the original experiment data is transited to standard unit and the format is transited the format which the program can be disposed. Secondly, according to threshold of idle speed, the passenger car experiment data is divided into passenger car driving microtrip, which contains the velocity, acceleration, driving time, ect. And the microtrip is combined with the traffic condition data (such as the date of passenger car driving, area, sections of road, cycles, etc) to form a microtrip unit. Putting all microtrip units into a database to form the microtrip original database, each cell shows a microtrip unit.(2) Establishing the principal composition microtrip of original database. This section mainly introduces the analysis method and the application of principal composition. The microtrip original database is used to draw out 15 indexes with reflect the passenger car driving performance (such as driving distance, average velocity, acceleration, standardize deviation of velocity and acceleration, etc). And the 15 indexes are combined with all microtrips to construct a microtrip index matrix. Secondly, analyze and standardized the matrix, based on the correlative matrix of the standard matrix, the eigenvalues of the standardized matrix and the contribute rate are obtained. And combine the standardized eigenvector of the bigger eigenvalues with the indexes of each microtrip in the microtrip index matrix to form several principal compositions which can reflect the character information of passenger car driving and correlative with the indexes. And the principal compositions are not correlative with each other.(3) Establishing clustering microtrip database. This section mainly introduces the method of fuzzy cluster and the establishment of the clustering microtrip database. Analyzing the principal compositions, the fuzzy cluster theory is used to fuzzy cluster and fuzzy optimizede cluster. According to the estimate index of validity, the principal composition clustering reflecting different driving character and clustering center are created, each clustering center describes one kind of driving condition. When estimating the validity, although the proportions of each estimate index and the numbers of clustering is different, the clustering can reflect the driving condition correctly. According to relationship of principal composition and original microtrip database, the clustering principal composition database is translated into microtrip clustering database. In the database, each line stands for a clustering, every microtrip describes the same driving condition in a clustering, but the character of every microtrip coming from different clustering is different.(4) Establishing the Matlab Graphical User Interface. This section establishes the Graphical User Interface (GUI) of the above sections, for the display of input data, input index and last result.