Electric Vehicle Travel Chain Model and Case Analysis Based on Time and Space Characteristics

Research on Charging-Discharging Operation Strategy for Electric Vehicles Based on Different Trip Patterns for Various City Types in China

“…Trips from each travel chain can therefore be divided into three categories based on the probability distribution function they follow: can vary, but they can be fitted with a certain distribution function. For trip type 1, the mean user departure time conforms to a gamma distribution, and the standard deviation conforms to a standard normal distribution [8,16]. Based on these distributions of the means and standard deviations, we solved the edge probability distribution of variable T to obtain the probability density formula (1) and cumulative distribution function (2) for the departure time of trip type 1.…”

“…Trip type 3 includes the second H-W trip in the complex work chain B and the E-H trip in the work-entertainment chain C. As the departure time for the second H-W trip in small cities is related to the "stay" time at home for lunch, and the departure time for the E-H trip in big cities is related to the stay time in the commercial area, we assume the stay time follows a random distribution [8], which is expressed as ∆t ~U(∆t min , ∆t max ). I this equation, ∆t min is the shortest residence time and ∆t max is the longest residence time.…”

“…Most studies [8][9][10] obtain EV trip patterns by conducting Monte Carlo sampling of the departure and arrival times of EVs based on normal distribution rules. Figure 5 shows the probability distribution of the time EVs reach the working area in the simple work chain A (H-W-H).…”

“…However, these studies did not consider the transfer of EVs in different spaces or its spatial distribution. Zhou et al [8] proposed an EV charging strategy that was based on information about user travel chains and traffic networks and considered the travel time of electric vehicles; the authors additionally derived a model predicting the temporal and spatial distribution of EV load under this strategy. Moghaddass et al [9] similarly used a mixed integer multi-objective optimization model to optimize an EV charging strategy that accounted for the power supply cost of the grid, load fluctuation, EV charging station benefits, user charging costs, and household electricity patterns.…”

“…Based on the probability density function parameters obtained by fitting data from previous studies [8,20], the gamma and standard normal distribution parameters for different type 1 trips were set as shown in Table A1:…”