Large-scale electric vehicle charging load prediction based on Monte Carlo method (Python & Matlab implementation)

content

1 Introduction to Monte Carlo Simulation Method

2 Calculation method for large-scale electric vehicle charging load prediction

0 Overview

For the research in this paper, according to the distribution functions and setting parameters of the influencing factors of electric vehicles for different purposes, the Monte Carlo method is used to randomly sample the probability distribution parameters of the daily mileage and initial charging time of electric vehicles for various purposes, and calculate the initial load. Electric state and charging time, and then predict the charging load curve of various types of electric vehicles, and finally obtain the total charging load curve by superimposing the charging load curves of electric vehicles for various purposes.

1 Introduction to Monte Carlo Simulation Method

The Monte Carlo method in computer simulation is also called random sampling technique or statistical testing method. The most important feature of this method is that it is a method based on probability and statistics theory. With the development of science and technology and the invention of electronic computers, Monte Carlo method has been widely used in various fields because of its advantages in describing the characteristics of physical development and the process of physical experiments.

2 Calculation method for large-scale electric vehicle charging load prediction

According to the research on the influencing factors of charging of electric vehicles for various purposes, the basic parameters are obtained, including the distribution of initial charging capacity and the distribution of initial time during the charging period, as shown in Table 1.

This table gives the parameters required for establishing the charging load forecast calculation for electric vehicles of various purposes:

Electric buses are generally charged regularly, and the daily mileage and initial charging time are uniformly distributed;

Electric taxis are generally charged quickly, the daily mileage is uniformly distributed, and the initial charging time is subject to a normal random distribution;

The electric private car is supplied with electric energy in two ways: conventional charging and fast charging. The daily mileage is subject to exponential distribution, and the initial charging time is subject to normal random distribution. In conventional charging and fast charging, the charging power of different types of electric vehicles is different. The charging power of electric buses is approximately five times that of electric private cars, and the charging power of electric taxis is approximately twice that of electric private cars.

After the principle and model research, it is assumed that the electric vehicles of various purposes are in the state of disorderly charging. The specific calculation method is as follows:

(1) According to the predicted ownership of electric vehicles for various purposes in Beijing, the market size of electric vehicles is determined as shown in Table 2.

(2) According to the distribution function and setting parameters of the influencing factors of electric vehicles for different purposes, as shown in Table 1, the Monte Carlo simulation method is used for simulation, and the date type, the initial charging time of the electric vehicle and the daily mileage are randomly selected.
(3) Calculate the initial state of charge of the vehicle and the time required for charging. The remaining power of the electric vehicle power battery is directly related to the charging time of the electric vehicle. For the charging time of an electric vehicle, the article uses the state of charge (SOC) of the electric vehicle power battery to calculate. Assuming that the power consumption of the battery is proportional to $d$ the driving distance, it is the traveled distance and $d_{m}$ the maximum cruising range in electric mode. It is assumed here that the electric vehicles of each different purpose consume the same amount of electricity per kilometer, and the remaining electricity at the end of the last trip is calculated by the following formula:

In the formula, SOC1 is the state of charge of the battery when charging is completed, and SOC2 is the state of charge of the battery when charging was completed last time.

To calculate the time T required for charging an electric vehicle, it can be obtained by calculating the battery capacity C, the initial state of charge SOC and the charging power Р.
The specific formula is as follows:

(4) Calculate the total charging load of an electric vehicle for a certain purpose at the i-th charging load calculation point. In this paper, the daily calculation is 1440 minutes, and the charging load is calculated every 15 minutes, with a total of 96 charging load calculation points. The prediction calculation method of the charging load of an electric vehicle for a certain purpose is as follows:

Among them $t_{1}$ , it represents the time when the nth electric vehicle of a certain purpose ends charging, the time when $t_{2}$ the nth electric vehicle of a certain purpose starts to be charged, and T represents the time required for the charging of the nth electric vehicle of a certain purpose. The total charging load of an electric vehicle for a certain purpose at the i-th charging load calculation point $P_{n,i}$ can be calculated by the following method

$P_{t_{i}}$ Represents the charging load of the n-th electric vehicle at the i-th charging load calculation point, and N represents the number of electric vehicles for a certain purpose.
(5) The total electric vehicle charging load is obtained by superimposing the charging loads of electric vehicles for various purposes. The calculation method of the total electric vehicle charging load at the i-th charging load calculation point is shown in formula (4-6):

In the formula, Nc, Nt, Nb represent the number of electric private cars, electric taxis, and electric buses charged at time i , respectively; $P_ {nc, i}$ , , represent the charging loads of electric private cars, electric taxis, and electric buses at time i , respectively size. $P_{nt,i}$ $P_{nb,i}$

3 complete code

Full Resource: Scaled Electric Vehicle Charging Load Forecasting Based on Monte Carlo Method (implemented in Python & Matlab)