Validation & Velification

“Build the right model or Build the model right.”

During the development of the time and carbon constrained models, validation and verification was utilised as an iterative process to determine the functionality of the code before producing our final simulation results. During the process of verification, it was determined that the model cannot successfully process more than 1024 electric vehicles in a singal timestep without failing. For this reason, the code was only used to produce simulated demand for a smaller population of electric vehicles in order to prevent this condition from ocurring. 

​

These steps above were conducted during the development of a simulation model with the ultimate goal of producing an accurate and credible model.

Time Constrained Model Validation

As presented previously, the model is considered success to perform time constraint if the EVs morning peak charging time was shifted to specified time window which is between 12pm-04pm. As we can see on the slides the unconstrained demand showed as blue line and time constrained as orange line both represent charging demand over the one-week period. As we focus on the highlighted black box, it indicates the area of the morning peak demand shifted to the designated time window, it is shown as the offset point of starting time constrained demand peak. Despite of that, the total charging demand output value is very close with the implementation of time constrained code. Unconstrained model given demand of 5716 MWh, while Time Constrained 5641 MWh.

Carbon Constrained Model Validation

Meanwhile, the model is considered to be carbon constrained, if the EVs charging patterns were shifted to meet a specific period of the day when the Carbon intensity is at its lowest point during the day. It is shown in the graph that the blue line represents carbon intensity for a week period, and the orange line is the electricity demand over the same period. The data presented in the graph are from Scotland carbon constrained model. We can notice that for each carbon’s minimal point is meet by a peak of electricity demand, and in the zoom-in view of the second day of the week the Gray dotted line is tangent to the minimum value and that It is when the demands peak, meaning most of the 10k EVs are charged at this period. Of course, this creates a strain on the grid but as far as the code goes it is performing as required.