System modeling of bidirectional charging system
VISPIRON SYSTEMS, has created a MATLAB tool for an OEM that can realistically simulate bidirectional charging processes of an electric vehicle with a wide range of input and output parameters at high resolution. Bidirectional charging offers great technical potential for a variety of use cases, but the best use cases of these need to be evaluated simulatively in a model.
1. Initial situation - potential of bidirectional charging
The initial situation at the OEM was that an assessment should first be made of the potential offered by bidirectional charging for electric vehicles (EVs). In bidirectional charging, the EV is to be charged and also discharged in a controlled manner in order to execute predefined use cases. This includes, for example, increasing PV self-consumption by charging the EV during the day and discharging it at night. Based on the identified potential, an assessment should later be made as to whether and to what extent bidirectional charging makes sense for EVs and end customers.
2. Challenge - Estimation of damage
A major challenge for the customer was to estimate whether bidirectional charging could cause damage to the EV’s electronics or memory. After all, the service life of the energy storage unit in the EV should be the top priority in the evaluation. How this estimation should be done was initially unclear. The findings from a previously conducted estimation using Excel calculations turned out to be too inaccurate to generate reliable results.
3. Procedure - Simulation model
The primary aim of the procedure was therefore to enable a reliable estimate to be made. Based on this initial situation, the VISPIRON programmers decided to create a simulation model in Matlab that uses real measurement data. For example, second-resolved solar irradiance measurement data was used to determine potential PV generation. Furthermore, second-by-second measurement data of the electricity consumption of more than 70 households for one year were included in the simulation model, and more than 30 million data points of the grid frequency in the German power grid were used. In addition to the extensive data basis, both the EV incl. real driving profiles and its DC charging station were meticulously modeled in order to be able to generate (customer) behavior in the simulation that is as real as possible. Using this data, the simulation model was able to map power flows both in the energy storage system and in the household of users for an entire year in second-by-second resolution.
4. Conclusion - A far-reaching understanding of the system
Due to the extensive system understanding of the VISPIRON project staff, it was possible to develop a comprehensive simulation model including linked databases. Exact modeling now makes it possible to make precise estimates for products whose series maturity is years in the future. In this way, measures can be taken at an early stage that maximize both revenue potential for customers and the service life of all components used.
