Abstract
The regulations for internal combustion vehicles, CO2 emission or NOx emission or noise and so on, are strengthened, .Therefore EV (electric vehicle)'s market is expanding. The amount of EV get more, the amount of electric get more and the impact for grid that are voltage fluctuation and frequency fluctuation is concerned. The short driving range is also problem for usability. It is important to inform driver not only SOC but possibility to arrive at destination without charge. So predicting to traffic condition is important. As the basic technology, the prediction the vehicles’ state that is drive or stay is important to solve two problems. In this research, Algorithm for predicting vehicle fleet's condition is developed. The data for study and test is obtained by person-trip survey conducted by Ministry of Land, Infrastructure and Transport. The location was divided into3 areas. And the state was stay or drive from an area to an area. The algorithm is based on left to right Markov-model. Future state probability is predicted using the latest observed state and state transition probability. As the result, prediction error is 3 % as parking. The prediction error of stay is less than the prediction error of drive. The frequency of stay is more than the frequency of drive, and robustness of stay for outliers is stronger. Therefore study data and test data are separated into week day and holiday, prediction error is about 1 %.
