Aiming at the problem that the conventional tram power consumption estimation model lacks quantitative analytical research on the significance of power consumption influencing elements, a tram power consumption estimation method on the strength of grey relational analytical research and multiple linear Multinomial logistic regression is proposed. Firstly, the influence elements of tram power consumption are made analytical research by tram dynamic model, and then the Grey relation analytical research method severs as calculating the relevant degree of these influence elements. Finally, the influence elements with higher correlation degree are chosen as the input variables of the model, and the tram power consumption estimation model is built based on the multiple linear Multinomial logistic regression. The experimental results indicate that the mean estimation error of the new model is 2.29%, which is smaller than the Multinomial logistic regression proposed in the reference. Therefore, the power consumption estimation model presented in this article can better reflect the trend of tram power consumption.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Prediction of Tram Energy Consumption Based on Modified Regression Model


    Contributors:
    Zhaolu, Su (author) / Liwei, Dong (author)


    Publication date :

    2019-11-01


    Size :

    132689 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Tram and trolleybus net traction energy consumption comparison

    Mwambeleko, Joachim Julius / Kulworawanichpong, Thanatchai / Greyson, Kenedy Aliila | IEEE | 2015


    ACoRP welcomes tram-tram trial

    British Library Online Contents | 2008


    TRAM

    LYALIN ALEKSANDR POLIKARPOVICH | European Patent Office | 2017

    Free access

    TRAM

    Nguyen, Tu | DataCite | 2024


    Model of tram line operation

    Bauer, Marek / Kandler, A. / Weiß, Hendrik | DataCite | 2006

    Free access