Investigation on Characteristic Parameters Identification and Evolution of Supercapacitor Energy Storage System From Sparse and Fragmented Monitoring Data

Supercapacitors with advantages of high-power density, fast charging speed and long cycle life, have very promising application prospects in many fields such as transportation and energy storage. Usually, the onboard operation profiles will be recorded and sent to remote monitoring terminal for further study. As the working condition of onboard supercapacitor is complex and fluctuating, it requires high sampling frequency, accuracy, and continuous recording. However, the remote monitoring data usually has the problems of low sampling frequency and fragmented recording, making it difficult to extract the characteristic parameters by traditional parameter identification methods. To solve the problem, this paper makes an extensive investigation on the long-term remote monitoring data of a supercapacitor tram and proposes a set of data processing method that can extract the characteristic parameters of supercapacitors from the sparse and fragmented data. Firstly, a group of proper data segments considering thermal stability of supercapacitor system was selected as effective data from the fragmented daily monitoring data. Secondly, the parameters which can be used to study the aging indicators are extracted by interior point method. With this method, the simulated voltage error is within 0.3% with a sampling frequency of 0.1 Hz. Through analyzing 3.5 years of remote monitoring data, it is found that the characteristic parameters exhibit the feature of seasonal fluctuations, which is highly related to temperature. To further extract the aging trend, a linear fitting model which eliminates the effect of seasonal fluctuations is proposed which can be used for analyzing the evolution characteristic of the studied supercapacitor system.