Regional charging pile optimal configuration method based on energy substitution rate

Abstract

The invention relates to a regional charging pile optimal configuration method based on an energy substitution rate, and the method comprises the steps: firstly providing an automobile ownership prediction method based on time trend prediction, and constructing a new energy automobile permeability prediction model in combination with factors affecting the permeability of a new energy automobile; new energy automobile permeability prediction and regional new energy automobile scale prediction are realized; an energy demand prediction model is established on the basis, the total energy consumption of the regional new energy vehicles is calculated, and meanwhile, the contribution of new energy vehicle growth to energy conservation and carbon reduction of road traffic is obtained according to the energy substitution rate; according to the total energy consumption analysis result and the power characteristics of the different types of charging piles, the coefficient characteristics are used to calculate the demands of the different types of charging piles, and compared with a traditional vehicle-pile ratio evaluation mode, the prediction result is more accurate, and the method has more guiding significance for developing charging pile construction planning; the method has the advantages that the permeability prediction model is constructed, the energy demand prediction model is established, and the prediction result is more accurate.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data analysis of new energy vehicles, and particularly relates to a method for optimizing the allocation of regional charging piles based on the energy substitution rate. Background Technique

[0002] With the gradual maturity of electric vehicle technology and the gradual reduction of costs, electric vehicles in the country have witnessed rapid growth. Along with this rapid growth, China attaches great importance to the construction of electric vehicle charging infrastructure. China has built the world's largest charging infrastructure system in terms of quantity, radiation area, and vehicle services, providing strong support for the rapid development of new energy vehicles. However, in many regions, there are still problems such as insufficient construction of public charging infrastructure, difficulties in installing and sharing charging facilities in residential communities, and prominent time-dependent supply-demand contradictions, which have restricted the release of the consumption potential of new energy vehicles in the region. Currently, there are relatively few indicators regarding the impact of the growth of electric vehicles on energy conservation and carbon reduction in road traffic. The main indicators considered are the electric vehicle penetration rate and the proportion of electric vehicles, where one reflects the proportion of electric vehicles in newly sold vehicles at the current stage, and the other reflects the proportion of new energy vehicles among the existing vehicles. However, in the context of energy conservation and carbon reduction in road traffic, these two indicators do not fully consider the energy consumption characteristics of various types of vehicles and cannot comprehensively reflect the value of the growth of new energy vehicles in terms of energy substitution and energy conservation and carbon reduction. Therefore, it is urgent to further study new indicators by combining the energy consumption characteristics of new energy vehicles and traditional vehicles, which can reflect the substitution of new energy vehicles for road traffic energy, so as to comprehensively evaluate the impact of the growth of new energy vehicles on energy conservation and emission reduction in road traffic. Currently, the influencing factors of the development trend of the electric vehicle penetration rate are attributed to aspects such as economic development level, policy subsidies, technology, charging speed, and intelligence, mainly focusing on the economic and social development level, policy subsidies, etc. However, with the economic and social development, the impact of factors such as policy subsidies on the growth of new energy vehicles is beginning to become smaller and smaller, while factors such as charging efficiency and intelligence level have an increasingly greater impact on the growth of the electric vehicle penetration rate. Currently, the research on the analysis of the electric vehicle penetration rate and the correlation between electric vehicles and charging piles generally adopts the method of a fixed vehicle-pile ratio to predict the demand for charging piles by new energy vehicles according to the general vehicle-pile ratio of different types of new energy vehicles. This method has two major problems: First, the demand for charging facilities by new energy vehicles is highly correlated with the types of new energy vehicles in the region and charging behavior habits. For example, most of the charging scenarios for private cars occur at private charging piles, and the demand for public charging piles is relatively small. Predicting the demand for charging piles by the method of a fixed vehicle-pile ratio is likely to result in either excessive or insufficient demand for local charging facilities. Therefore, it is necessary to conduct relevant research by combining the actual charging behavior and charging habits of new energy vehicles in the region to accurately evaluate the demand for charging facilities by new energy vehicles;Second, at the current stage, the types of charging facilities vary greatly, and the charging load ranges from relatively small private 7-kilowatt charging facilities to high-power 160-kilowatt DC charging piles. Different types of charging facilities are suitable for corresponding charging scenarios. Therefore, when conducting the analysis of the correlation between electric vehicles and charging piles, it is necessary to carry out research in combination with the specific charging scenarios of new energy vehicles, and obtain the specific charging load of charging facilities through the analysis and evaluation of the charging behavior of new energy vehicles. Therefore, it is very necessary to provide a method for optimizing the allocation of regional charging piles based on the energy substitution rate, which can construct a penetration rate prediction model, establish an energy demand prediction model, and make the prediction results more accurate. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a method for optimizing the allocation of regional charging piles based on the energy substitution rate, which can construct a penetration rate prediction model, establish an energy demand prediction model, and make the prediction results more accurate.

[0004] The purpose of the present invention is achieved as follows: A method for optimizing the allocation of regional charging piles based on the energy substitution rate, the method includes the following steps:

[0005] Step 1: Based on the analysis of regional population influencing factors, combined with the analysis of vehicle ownership, predict the vehicle ownership in the target year through the method of time series data analysis;

[0006] Step 2: Through the analysis of the influencing factors of new energy vehicles, combined with the historical data of the penetration rate of new energy vehicles, predict the penetration rate of new energy vehicles and the scale of new energy vehicles in the target year;

[0007] Step 3: Calculate the total annual energy demand of new energy vehicles based on the electric vehicle ownership, annual average driving mileage, the proportion of electric mileage in the annual average driving mileage, and the vehicle's power consumption per 100 kilometers, and calculate the electric vehicle substitution rate index based on the energy demand to fully reflect the substitution effect of the growth of new energy vehicles on transportation energy;

[0008] Step 4: Based on the power of different types of charging piles and the average annual utilization time, combine the total energy demand data of private charging piles and public charging piles, and combine the existing total annual energy demand of new energy vehicles to calculate the demand scale of different types of charging piles;

[0009] Step 5: Further combine the service life and renewal requirements of the charging piles to calculate the new construction demand of the charging piles in different regions in the target year.

[0010] The prediction of vehicle ownership in Step 1 specifically includes the following steps:

[0011] Step 1.1: Collect the historical data of the electric vehicle ownership and population data in the region, and proceed to the next step to calculate the vehicle ownership per thousand people;

[0012] Step 1.2: Combine the historical electric vehicle ownership data, select the hyperbolic function for fitting to obtain the fitting function of the vehicle ownership per thousand people, and use this fitting function to calculate the vehicle ownership data year by year for the next 20 years: Let the independent variable x be the number of years, the dependent variable y be the electric vehicle ownership per thousand people, and the fitting function be: In the formula, x is the number of years; a, b, and c are the parameters to be fitted;

[0013] Step 1.3: Combine the historical population data and predict the population trend for the next 20 years through the trend extrapolation method;

[0014] Step 1.4: Calculate the vehicle data based on the predicted vehicle ownership per thousand people data and the predicted population data: Vehicle ownership in the horizontal year = Predicted result of vehicle ownership per thousand people * Predicted population size.

[0015] The prediction of the proportion of new energy vehicles in Step 2 specifically includes the following steps:

[0016] Step 2.1: Obtain the input variable data and the historical new energy vehicle sales data from the actual project;

[0017] Step 2.2: Calculate the mutual information of each input variable based on the mRMR algorithm and sort the input variables in descending order of mutual information;

[0018] Step 2.3: According to the sorting result of the input variable sequence obtained by the mRMR model, select 1 to m input variables in the sequence as the preselected input variable set, input it into the BP neural network prediction model for training, and obtain m optional BP neural networks;

[0019] Step 2.4: Calculate the deviation index RMSE for each of the m optional BP neural networks according to their output results. Taking the minimum deviation value RMSE as the evaluation index, for the m optional neural network models, taking the minimum deviation value RMSE as the goal, select the best mRMR - BP neural network prediction model, and the corresponding preselected input variable set is the optimal input variable set;

[0020] Step 2.5: After optimization by the mRMR algorithm, determine the topological structure of the mRMR - BP neural network prediction model and initialize the mRMR - BP neural network prediction model; Take the absolute value of the relative error of the model prediction result as the individual fitness value, and use the PSO algorithm to optimize the initial weights and thresholds of the mRMR - BP neural network prediction model;

[0021] Step 2.6: Train the model until the relative error between the predicted value and the actual value satisfies that the absolute value is less than or equal to 0.01, and then stop training.

[0022] The input variable data in Step 2.1 includes: population growth rate, GDP growth rate, government subsidies, cruising range, fuel price, battery life, vehicle-to-charging-pile ratio information data.

[0023] The calculation of the total annual energy demand of new energy vehicles in Step 3 is specifically as follows: According to the scale of new energy vehicles and the types of new energy vehicles, and taking into account the differences in charging methods for different types of vehicles, calculate the total energy consumption of new energy vehicles by type: In the formula, E is the alternative energy consumption of new energy vehicles; Q i is the number of the i-th type of new energy vehicle; D i is the driving distance of the i-th type of new energy vehicle; β i is the energy consumption per unit distance of the i-th type of new energy vehicle.

[0024] The calculation of the electric vehicle substitution rate index in Step 3 is specifically as follows: The new energy vehicle substitution rate is the ratio of the total amount of conventional energy substitution or savings formed by the energy used by new energy vehicles in the total energy consumption of road transportation. The calculation formula is: Among them, ε represents the new energy vehicle energy substitution rate; E e is the alternative energy consumption of new energy vehicles; E total is the total energy consumption of road transportation; Q ei is the number of the i-th type of new energy vehicle; D ei is the driving distance of the i-th type of new energy vehicle; β ei is the energy consumption per unit distance of the traditional vehicle replaced by the i-th type of new energy vehicle; Q oj is the number of the j-th type of conventional energy vehicle; D oj is the driving distance of the j-th type of conventional energy vehicle; β oj is the energy consumption per unit distance of the j-th type of traditional energy vehicle.

[0025] The calculation of the electric vehicle substitution rate index includes the following steps:

[0026] Step 3.1: Calculate the gasoline consumption substitution amount and diesel consumption substitution amount corresponding to the electric vehicle charging amount according to the mileage and the gasoline and diesel consumption ratios. The relevant calculation formulas are: gasoline consumption = electric vehicle mileage × gasoline consumption ratio × gasoline vehicle fuel consumption per 100 kilometers; diesel consumption = electric vehicle mileage × diesel consumption ratio × diesel vehicle fuel consumption per 100 kilometers; gasoline (diesel) consumption ratio = actual gasoline (diesel) consumption / total gasoline (diesel) consumption; electric vehicle mileage = electric vehicle charging amount / power consumption per 100 kilometers = electric vehicle alternative gasoline vehicle mileage + electric vehicle alternative diesel vehicle mileage;

[0027] Step 3.2: Calculate the energy substitution rate in the field of electric vehicles by calculating the proportion of gasoline and diesel consumption substitution in gasoline and diesel consumption.

[0028] Based on the energy substitution rate in the field of electric vehicles in Step 3.2, define the carbon emission reduction rate of new energy vehicles, specifically: According to the energy consumption data, calculate the total carbon emissions generated by each type of fuel consumption, and summarize the carbon emissions generated by each fuel consumption to obtain the total carbon emissions of the entire industry. The calculation formula is: Among them, Q is the total carbon emissions in the transportation field; H i is the total consumption of the i-th energy; δ i is the carbon emission factor of the i-th energy.

[0029] Step 4 uses two methods to predict and calculate the demand scale of different types of charging piles, including: the energy method prediction method and the charging pile demand method; among them, the energy method prediction method is specifically: Calculate the number of charging piles required based on the energy demand of electric vehicles. By predicting the total energy demand and the annual contribution energy of a single charging pile, calculate the total demand for charging piles in the planned year: In the formula, TER i is the total annual energy demand of the i-th new energy vehicle; N i is the total annual demand of the i-th charging pile; P i is the typical power of the corresponding type of charging pile; η is the annual demand coefficient of this type of charging pile; the charging pile demand method is specifically: N i =Q i ·λ i , in the formula, N i is the total annual demand of the i-th charging pile; Q i is the total number of new energy vehicles; λ i is the vehicle-to-charging pile ratio of new energy vehicles.

[0030] The newly built demand for charging piles in different regions in the horizontal year in Step 5 is specifically: The annual new demand for charging piles is based on the analysis of the charging pile demand in the planned year, combined with the number of new energy vehicle charging piles each year, considering the service life of new energy vehicle charging piles, and calculating the number of charging piles that need to be newly added each year. The calculation method is to superimpose the new demand for charging piles and the charging piles newly added k years ago as the charging piles that need to be newly added in the planned year. The calculation method is as follows: ΔN i =N i -N i-1 +(N i-k -N i-k-1 ), in the formula, ΔN i is the demand for newly added charging piles of type i in the planned year; N i is the charging pile demand in the i-th year; k is the service life of the charging pile; N i-kis the demand for charging piles in the (i - k)th year; N i-k-1 is the demand for charging piles in the year before the (i - k)th year.

[0031] Advantages of the present invention: The present invention provides an optimized allocation method for regional charging piles based on the energy substitution rate. In use, the present invention first provides a prediction method for the vehicle ownership based on time trend prediction, and then combines the influencing factors affecting the penetration rate of new energy vehicles, such as the price of new energy vehicles, average cruising range, battery life, and intelligence, etc., to construct a prediction model for the penetration rate of new energy vehicles, so as to realize the prediction of the penetration rate of new energy vehicles and the prediction of the scale of regional new energy vehicles; on this basis, the present invention establishes an energy demand prediction model according to factors such as the type of new energy vehicle, average cruising range, and energy consumption per unit mileage, calculates the total energy consumption of regional new energy vehicles, and at the same time analyzes the contribution of the growth of new energy vehicles to energy conservation and carbon reduction on road traffic according to the definition of the energy substitution rate; the present invention further calculates the demand for different types of charging piles according to the analysis results of the total energy consumption, as well as the power characteristics and utilization coefficient characteristics of different types of charging piles. The present invention has the advantages of constructing a penetration rate prediction model, establishing an energy demand prediction model, and more accurate prediction results. Description of the Drawings

[0032] Figure 1 is the overall flowchart of the present invention.

[0033] Figure 2 is the schematic diagram of the prediction method for the penetration rate of new energy vehicles of the present invention.

[0034] Figure 3 is the flowchart of the mRMR algorithm of the present invention. Detailed Embodiments

[0035] The following further describes the present invention with reference to the drawings.

[0036] Embodiment 1

[0037] The present invention first combines the development trend of the automotive industry, combines the historical vehicle ownership data of the region and the local population data to predict the vehicle ownership of the region; then combines the analysis of the influencing factors of the penetration rate of new energy vehicles in the region, analyzes the new energy vehicles in the region, and predicts the new energy vehicle substitution rate of the region; finally, combines the energy substitution rate to analyze the demand scale of the charging piles; the overall prediction process is as Figure 1 shown.

[0038] As Figures 1-3 shown, an optimized allocation method for regional charging piles based on the energy substitution rate, the method includes the following steps:

[0039] Step 1: Based on the analysis of influencing factors such as regional population, combined with the analysis of the vehicle ownership, predict the vehicle ownership in the target year through the method of time series data analysis;

[0040] In this embodiment, the vehicle ownership prediction method is specifically as follows: The development of the automotive market is divided into three stages: gestation period, popularization period, and saturation period. The vehicle ownership per unit population in most countries in the world conforms to this trend. To predict the vehicle ownership, the present invention intends to fit the vehicle ownership trend by combining with the hyperbolic function, and on this basis, predict the vehicle ownership in different target years. The main calculation steps for vehicle ownership prediction are as follows:

[0041] The first step: Complete the collection of basic data, collect the historical vehicle ownership data and population data of electric vehicles in the region, and calculate the vehicle ownership per thousand people through the following method;

[0042] The second step: Combine the historical vehicle ownership data of electric vehicles, select the hyperbolic function for fitting, obtain the fitting function of the vehicle ownership per thousand people, and use this fitting function to calculate the vehicle ownership data year by year for the next 20 years: Let the independent variable x be the number of years (set 2000 as 0, 2001 as 1, and so on), and the dependent variable y be the vehicle ownership per thousand people of electric vehicles. The fitting function is: In the formula, x is the number of years. Starting from 2000, for the convenience of fitting, x = 1 in 2000, x = 2 in 2001, and so on. This fitting is up to 2045 at most, that is, x = 45; take the vehicle ownership per thousand people, unit: vehicles per thousand people; a, b, c are the parameters to be fitted;

[0043] The third step: Combine the historical population data, and predict the population trend for the next 20 years through the method of trend extrapolation;

[0044] The fourth step: According to the predicted data of the vehicle ownership per thousand people and the predicted population data, calculate the vehicle data: Vehicle ownership in the target year = Predicted result of vehicle ownership per thousand people * Predicted population scale.

[0045] Step 2: Through the analysis of influencing factors of new energy vehicles such as endurance and battery life, combined with the historical data of the new energy vehicle penetration rate, predict the penetration rate of new energy vehicles and the scale of new energy vehicles in the target year;

[0046] In this embodiment, the prediction of the proportion of new energy vehicles is specifically as follows: The main goal of this step is to predict the penetration rate of new energy vehicles, so as to analyze and predict the number of new energy vehicles year by year.

[0047] Analysis of influencing factors: Study the main influencing factors affecting the growth rate of electric vehicles and the corresponding energy substitution rate from dimensions such as economy, population, vehicle ownership, and electric vehicle technology (endurance mileage, charging rate), and economy (local subsidy policies, electricity price subsidies, time-of-use electricity prices); Generally, the development scale of electric vehicles is mainly affected by the following factors: 1) Energy demand and energy usage habits for travel, which are affected by macro factors such as economic development and population size; 2) Maturity and economy of key technologies such as the endurance mileage, charging time, service life, and safety of electric vehicles; 3) Completeness of charging and swapping infrastructure such as the number, layout, and charging convenience of charging piles; 4) Policy mechanisms such as government planning goals, environmental constraints, and preferential policies; 5) Influence of intelligent levels such as autonomous driving. Based on the analysis of influencing factors, the prediction process of the penetration rate of new energy vehicles is as Figure 2 shown.

[0048] The main calculation steps are as follows:

[0049] Step 1: Obtain input variable data (information such as population growth rate, GDP growth rate, government subsidies, endurance mileage, fuel price, battery life, vehicle-to-pile ratio, etc.) and annual sales data of new energy vehicles from actual projects;

[0050] Step 2: Calculate the mutual information of each input variable based on the mRMR algorithm, and sort the input variables according to the mutual information from high to low; Among them, the mRMR algorithm is used to select the most representative and least redundant feature subset from numerous features; Its core idea is that when selecting features, it is necessary to ensure that the selected features have a high correlation with the target variable and at the same time minimize the redundancy between features; In this way, the number of input features can be reduced, the complexity of the model can be lowered, and at the same time, the prediction accuracy and generalization ability of the model can be improved; Its flow chart is as Figure 3 shown.

[0051] Step 3: According to the sorting result of the input variable sequence obtained by the mRMR model, select 1 to m input variables in the sequence as the preselected input variable set, and input it into the BP neural network prediction model for training, and thus obtain m optional BP neural networks;

[0052] Step 4: For the m optional BP neural networks, calculate their deviation index RMSE according to their output results respectively. Taking the minimum deviation value RMSE as the evaluation index, for the m optional neural network models, taking the minimum deviation value RMSE as the goal, select the best mRMR-BP neural network prediction model, and the corresponding preselected input variable set is the optimal input variable set;

[0053] Step 5: After optimization by the mRMR algorithm, determine the topological structure of the mRMR-BP neural network prediction model and initialize the mRMR-BP neural network prediction model; use the absolute value of the relative error of the model prediction result as the individual fitness value, and use the PSO algorithm to optimize the initial weights and thresholds of the mRMR-BP neural network prediction model;

[0054] Step 6: Train the model until the absolute value of the relative error between the predicted value and the actual value is less than or equal to 0.01, then stop training.

[0055] Step 3: Calculate the total energy demand of new energy vehicles each year based on the number of electric vehicles, annual average mileage, the proportion of electric mileage in annual average mileage, and vehicle power consumption per 100 kilometers. Calculate indicators such as the electric vehicle substitution rate based on the energy demand to fully reflect the role of new energy vehicle growth in replacing transportation energy.

[0056] Step 4: Based on the power of different types of charging piles, the average annual utilization time (in hours), the total energy demand of private and public charging piles, and other data are combined with the existing annual total energy demand of new energy vehicles to calculate the demand scale of different types of charging piles. In addition, a fixed car-to-pile ratio can be used to calculate the demand for different types of charging piles in a horizontal year based on the number and type of new energy vehicles;

[0057] Step 5: Further combine the service life and renewal requirements of charging piles to calculate the demand for new charging pile construction in different regions in different years.

[0058] The present invention is a method for optimizing the configuration of regional charging piles based on energy substitution rate. In use, the present invention first provides a method for predicting the number of cars in use based on time trend prediction, and then combines the factors affecting the penetration rate of new energy vehicles, such as new energy vehicle price, average cruising range, battery life, and intelligence, to construct a new energy vehicle penetration rate prediction model to achieve new energy vehicle penetration rate prediction and regional new energy vehicle scale prediction; on this basis, the present invention establishes an energy demand prediction model according to factors such as new energy vehicle type, average cruising range, and unit mileage energy consumption, calculates the total energy consumption of regional new energy vehicles, and at the same time, according to the definition of energy substitution rate, analyzes the contribution of new energy vehicle growth to road traffic energy conservation and carbon reduction; the present invention further calculates the demand for different types of charging piles based on the total energy consumption analysis results, as well as the power characteristics and usage coefficient characteristics of different types of charging piles; compared with the traditional method of evaluating using the vehicle-to-pile ratio, the prediction results provided by the present invention are more accurate and have more guiding significance for carrying out charging pile construction planning; the present invention has the advantages of constructing a penetration rate prediction model, establishing an energy demand prediction model, and having more accurate prediction results.

[0059] Example 2

[0060] As Figures 1-3 shown, a method for optimizing the allocation of regional charging piles based on the energy substitution rate, the method comprising the following steps:

[0061] Step 1: Based on the analysis of influencing factors such as regional population, combined with the analysis of the vehicle ownership, predict the vehicle ownership in the horizontal year through the method of time series data analysis;

[0062] Step 2: Through the analysis of influencing factors affecting new energy vehicles such as endurance and battery life, combined with the historical data of the penetration rate of new energy vehicles, predict the penetration rate of new energy vehicles and the scale of new energy vehicles in the horizontal year;

[0063] Step 3: Calculate the total annual energy demand of new energy vehicles based on parameters such as the electric vehicle ownership, the average annual driving mileage, the proportion of electric mileage in the average annual driving mileage, and the electricity consumption per 100 kilometers of the vehicle, and calculate indicators such as the electric vehicle substitution rate based on the energy demand, fully reflecting the substitution effect of the growth of new energy vehicles on transportation energy;

[0064] In this embodiment, the analysis of the total energy consumption of new energy vehicles and the analysis of the energy substitution rate are specifically as follows:

[0065] ① Calculation of the total annual energy demand of new energy vehicles: According to the scale of new energy vehicles and the types of new energy vehicles, while considering the differences in charging methods for different types of vehicles, such as private cars generally charging at private piles and public piles generally charging at home; calculate the total energy consumption of different types of new energy vehicles according to the following formula: In the formula, E is the alternative energy consumption of new energy vehicles; Q i is the number of the i-th type of new energy vehicle (here the types of new energy vehicles include private cars, taxis, buses, special operation vehicles, etc.); D i is the driving distance of the i-th type of new energy vehicle; β i is the energy consumption per unit distance of the i-th type of new energy vehicle.

[0066] ② Calculation of the electric vehicle substitution rate index: Considering that different types of vehicles have large differences in energy consumption, the traditional index of the penetration rate of new energy vehicles cannot fully reflect the substitution effect of the growth of new energy vehicles on road traffic energy consumption. In order to accurately reflect the value of the rapid growth of new energy vehicles for energy substitution and road traffic energy conservation and carbon reduction, the present invention adds the concept of the new energy vehicle substitution rate, and defines the new energy vehicle substitution rate as the ratio of the total amount of conventional energy substitution or savings formed by the energy used by new energy vehicles in the total energy consumption of road traffic, and the calculation method is as follows: Among them, ε represents the new energy vehicle energy substitution rate; E eReplace the energy consumption of new energy vehicles; E total Total energy consumption of road transportation; Q ei The number of the i-th type of new energy vehicle (here, the types of new energy vehicles include private cars, taxis, buses, special operation vehicles, etc.); D ei The driving distance of the i-th type of new energy vehicle; β ei The energy consumption per unit distance of the traditional vehicles (gasoline and diesel) replaced by the i-th type of new energy vehicle; Q oj The number of the j-th type of conventional energy vehicle; D oj The driving distance of the j-th type of conventional energy vehicle; β oj The energy consumption per unit distance of the j-th type of traditional energy vehicle.

[0067] This indicator can fully reflect the substitution effect of the increase in the proportion of new energy vehicles on traffic energy consumption, as well as the corresponding support for energy conservation and carbon reduction in road traffic; this indicator can comprehensively reflect the proportion of renewable energy in road traffic energy consumption.

[0068] The specific calculation method of this indicator is as follows:

[0069] The first step: Calculate the substitution amount of gasoline consumption and diesel consumption corresponding to the charging amount of electric vehicles according to the mileage and the proportion of gasoline and diesel consumption. The relevant calculation formulas are as follows: Gasoline consumption = Electric vehicle mileage × Gasoline consumption proportion × Gasoline vehicle fuel consumption per 100 kilometers; Diesel consumption = Electric vehicle mileage × Diesel consumption proportion × Diesel vehicle fuel consumption per 100 kilometers; Gas (Diesel) consumption proportion = Gas (Diesel) actual consumption / Gas (Diesel) total consumption; Electric vehicle mileage = Electric vehicle charging amount / Power consumption per 100 kilometers

[0070] = Electric vehicle's mileage replacing gasoline vehicles + Electric vehicle's mileage replacing diesel vehicles;

[0071] The second step: Calculate the energy substitution rate in the field of electric vehicles by calculating the proportion of gasoline and diesel consumption substitution amount in gasoline and diesel consumption.

[0072] Considering that only considering this single indicator of energy substitution rate is not sufficient to comprehensively reflect the carbon emission reduction contribution of new energy vehicles to transportation energy; through investigations, taking Zhejiang Province as an example, roughly calculated using coal-fired power, 1 kWh of electricity generates about 0.8 kg of carbon dioxide, and an electric vehicle can run an average of 5 kilometers per 1 kWh of electricity; the carbon emission per kilometer is about 0.8 / 5 = 0.16 kg. Taking the household commuting vehicles, which account for the main use of vehicles, as an example, without calculating other pollutants, the carbon emission per kilometer of medium-sized gasoline-consuming cars is only between 0.27 and 0.3 kg; in addition, the proportion of coal-fired power generation in Zhejiang Province in 2019 was 62%, and the proportion of renewable energy generation was about 1 / 3.

[0073] In order to further consider the carbon emission reduction effect of new energy vehicles, the present invention defines the concept of carbon emission reduction rate of new energy vehicles. First, according to the energy consumption data released by the authority, the total carbon emissions generated by each type of fuel consumption are calculated. The carbon emissions generated by each type of fuel consumption can be summarized to obtain the carbon emissions of the entire industry. The calculation formula is shown as follows: Where Q is the total carbon emissions in the transportation sector; H i is the total consumption of the i-th energy; δ i is the carbon emission factor of the i-th energy source.

[0074] Step 4: Based on the power of different types of charging piles, the average annual utilization time (in hours), the total energy demand of private and public charging piles, and other data are combined with the existing annual total energy demand of new energy vehicles to calculate the demand scale of different types of charging piles. In addition, a fixed car-to-pile ratio can be used to calculate the demand for different types of charging piles in a horizontal year based on the number and type of new energy vehicles;

[0075] In this embodiment, the vehicle-pile association analysis and charging pile demand prediction are specifically as follows: the present invention adopts two different methods to predict the demand for charging piles.

[0076] ① Energy method prediction method: The energy method calculates the required number of charging piles based on the energy demand of electric vehicles. The energy demand is determined by the number of electric vehicles, the average daily mileage, the power consumption per 100 kilometers and other parameters. By predicting the total energy demand and the annual energy contribution of a single charging pile, the total demand for charging piles in the planning level year is calculated. The calculation formula is as follows: In the formula, TER i is the total annual energy demand of the i-th new energy vehicle; N i is the total annual demand for the i-th type of charging pile; P i is the typical power of the corresponding type of charging pile. According to the current technical development stage of charging piles, slow charging piles are generally 7kW-30kW; fast charging piles are 60-120kW; super charging piles are above 240kW; η is the annual demand coefficient of this type of charging pile, which reflects the proportion of the charging pile usage period to the annual period; to simplify the processing period, private cars are generally considered to be charged at private piles, and other types of new energy vehicles are charged at public piles.

[0077] ② Charging pile demand forecasting method: N i =Q i ·λ i , where N i is the total annual demand for the i-th type of charging pile; Q i is the total number of new energy vehicles; i The ratio of cars to charging piles for new energy vehicles.

[0078] Step 5: Further combine the service life and update requirements of charging piles to calculate the new construction requirements of charging piles in different regions in the horizontal year.

[0079] In this embodiment, the prediction of the new demand for charging piles in the planning year is specifically as follows: The annual new demand for charging piles is based on the analysis of the charging pile demand in the planning horizontal year, combined with the number of new energy vehicle charging piles each year, considering the service life of new energy vehicle charging piles, and calculating the number of new charging piles required each year. The calculation method is to superimpose the new charging pile demand and the new charging piles added k years ago as the new charging piles required in the planning year. The calculation method is as follows: ΔN i = N i - N i-1 +(N i-k - N i-k-1 ), where ΔN i is the new demand for charging piles of type i in the planning year; N i is the demand for charging piles in the i-th year; k is the service life of the charging pile; N i-k is the demand for charging piles in the (i - k)-th year; N i-k-1 is the demand for charging piles in the year before the (i - k)-th year.

[0080] The present invention is a method for optimizing the allocation of regional charging piles based on the energy substitution rate. In use, the present invention first provides a method for predicting the vehicle ownership based on time trend prediction, and then combines the influencing factors affecting the penetration rate of new energy vehicles, such as the price of new energy vehicles, average cruising range, battery life, and intelligence, to construct a prediction model for the penetration rate of new energy vehicles, realizing the prediction of the penetration rate of new energy vehicles and the prediction of the scale of regional new energy vehicles; on this basis, the present invention establishes an energy demand prediction model according to factors such as the type of new energy vehicle, average cruising range, and energy consumption per unit mileage, calculates the total energy consumption of regional new energy vehicles, and at the same time analyzes the contribution of the growth of new energy vehicles to energy conservation and carbon reduction on road traffic according to the definition of the energy substitution rate; the present invention further calculates the demand for different types of charging piles according to the analysis results of the total energy consumption, as well as the power characteristics and usage coefficient characteristics of different types of charging piles; the present invention can support the prediction of the development scale of regional new energy vehicles and the prediction of charging pile demand, thereby supporting the development of charging pile development plans and the formulation of corresponding distribution network adaptability measures to serve the rapid growth of regional new energy vehicles; the present invention has the advantages of constructing a penetration rate prediction model, establishing an energy demand prediction model, and more accurate prediction results.

Claims

1. A regional charging pile optimization configuration method based on the energy substitution rate, characterized in that: The method includes the following steps: Step 1: Based on the analysis of regional population influencing factors and combined with the analysis of the vehicle ownership, predict the vehicle ownership in the target year through time series data analysis; Step 2: Through the analysis of the influencing factors of new energy vehicles and combined with the historical data of the penetration rate of new energy vehicles, predict the penetration rate of new energy vehicles and the scale of new energy vehicles in the target year; Step 3: Calculate the total annual energy demand of new energy vehicles based on the electric vehicle ownership, annual average driving mileage, the proportion of electric mileage in the annual average driving mileage, and the vehicle power consumption per 100 kilometers, and calculate the electric vehicle replacement rate index based on the energy demand to fully reflect the substitution effect of the growth of new energy vehicles on transportation energy; Step 4: Based on the power of different types of charging piles and the average annual utilization time, calculate the total energy demand data of private charging piles and public charging piles, and combine with the existing total annual energy demand of new energy vehicles to calculate the demand scale of different types of charging piles; Step 5: Further combine the service life and renewal requirements of the charging piles to calculate the new construction demand of the charging piles in different regions in the target year.

2. The regional charging pile optimization configuration method based on the energy substitution rate according to claim 1, wherein: The prediction of vehicle ownership in Step 1 specifically includes the following steps: Step 1.1: Collect the historical data of the electric vehicle ownership and population data in the region, and proceed to the next step to calculate the vehicle ownership per thousand people; Step 1.2: Combined with the data of electric vehicle ownership in previous years, a hyperbolic function is selected for fitting to obtain the fitting function of the number of cars per thousand people, and the fitting function is used to calculate the annual car ownership data for the next 20 years: Let the independent variable x be the number of years, the dependent variable y be the number of electric vehicles per thousand people, and the fitting function is: In the formula, x is the age; a, b, c are the parameters to be fitted; Step 1.3: Combine the historical population data and predict the population trend in the next 20 years through the trend extrapolation method; Step 1.4: According to the predicted data of the vehicle ownership per thousand people and the predicted population data, calculate the vehicle data: Vehicle ownership in the target year = Predicted result of vehicle ownership per thousand people * Predicted population scale.

3. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 1, characterized in that: The prediction of the proportion of new energy vehicles in Step 2 specifically includes the following steps: Step 2.1: Obtain the input variable data and the historical sales data of new energy vehicles from actual projects; Step 2.2: Calculate the mutual information of each input variable based on the mRMR algorithm, and sort the input variables in descending order of mutual information; Step 2.3: According to the sorting result of the input variable sequence obtained by the mRMR model, select 1 to m input variables in the sequence as the preselected input variable set, input it into the BP neural network prediction model for training, and obtain m optional BP neural networks; Step 2.4: Calculate the deviation index RMSE for each of the m optional BP neural networks according to their output results. Taking the minimum deviation value RMSE as the evaluation index, for the m optional neural network models, taking the minimum deviation value RMSE as the goal, select the best mRMR-BP neural network prediction model, and the corresponding preselected input variable set is the optimal input variable set; Step 2.5: After optimization by the mRMR algorithm, determine the topological structure of the mRMR-BP neural network prediction model and initialize the mRMR-BP neural network prediction model; taking the absolute value of the relative error of the model prediction result as the individual fitness value, use the PSO algorithm to optimize the initial weights and thresholds of the mRMR-BP neural network prediction model; Step 2.6: Train the model until the prediction value and the actual value satisfy that the absolute value of the relative error is less than or equal to 0.01, and then stop the training.

4. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 3, wherein: The input variable data in the above Step 2.1 includes: population growth rate, GDP growth rate, government subsidy, cruising range, fuel price, battery life, vehicle-pile ratio information data.

5. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 1, wherein: The calculation of the total annual energy demand of new energy vehicles in step 3 is specifically as follows: Based on the scale and type of new energy vehicles, and considering the differences in charging methods among different types of vehicles, the total energy consumption of new energy vehicles by type is calculated: In the formula, E is the alternative energy consumption of new energy vehicles; Q i is the number of the i-th type of new energy vehicle; D i is the driving distance of the i-th type of new energy vehicle; β i is the energy consumption per unit distance of the i-th type of new energy vehicle.

6. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 5, characterized in that: The calculation of the electric vehicle substitution rate index in step 3 is specifically as follows: The new energy vehicle substitution rate is the ratio of the total amount of conventional energy substitution or savings formed by the energy used by new energy vehicles in the total energy consumption of road transportation. The calculation formula is: Among them, ε represents the new energy vehicle energy substitution rate; E e is the new energy vehicle substitution energy consumption; E total is the total energy consumption of road transportation; Q ei is the number of the i-th new energy vehicle; D ei is the driving distance of the i-th new energy vehicle; β ei is the unit distance energy consumption of the traditional vehicle replaced by the i-th new energy vehicle; Q oj is the number of the j-th conventional energy vehicle; D oj is the driving distance of the j-th conventional energy vehicle; β oj is the unit distance energy consumption of the j-th traditional energy vehicle.

7. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 6, characterized in that: The calculation of the electric vehicle substitution rate index includes the following steps: Step 3.1: Calculate the gasoline consumption substitution amount and diesel consumption substitution amount corresponding to the electric vehicle charging amount according to the mileage and the gasoline and diesel consumption ratios. The relevant calculation formulas are: gasoline consumption substitution amount = electric vehicle mileage × gasoline consumption ratio × gasoline vehicle fuel consumption per 100 kilometers; diesel consumption substitution amount = electric vehicle mileage × diesel consumption ratio × diesel vehicle fuel consumption per 100 kilometers; gasoline (diesel) consumption substitution ratio = gasoline (diesel) actual consumption / gasoline (diesel) total consumption; Step 3.2: Calculate the energy substitution rate in the electric vehicle field by calculating the proportion of the gasoline and diesel consumption substitution amounts in the gasoline and diesel consumption.

8. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 7, wherein: In step 3.2, based on the energy substitution rate in the field of electric vehicles, the carbon emission reduction rate of new energy vehicles is defined as follows: According to the energy consumption data, calculate the total carbon emissions generated by each type of fuel consumption, and summarize the carbon emissions generated by each fuel consumption to obtain the total carbon emissions of the entire industry. The calculation formula is: where Q is the total carbon emissions in the transportation field; H i is the total consumption of the i-th energy; δ i is the carbon emission factor of the i-th energy.

9. The method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 1, characterized in that: In step 4, two methods are adopted to predict and calculate the demand scale of different types of charging piles, including: the energy method prediction method and the charging pile demand method; among them, the energy method prediction method is specifically: calculating the required number of charging piles based on the energy demand of electric vehicles, and calculating the total demand of charging piles in the planned horizontal year by predicting the total energy demand and the annual contribution energy of a single charging pile: In the formula, TER i is the total annual energy demand of the i-th new energy vehicle; N i is the total annual demand of the i-th charging pile; P i is the typical power of the corresponding type of charging pile; η is the annual demand coefficient of this type of charging pile; the charging pile demand method is specifically: N i =Q i ·λ i In the formula, N i is the total annual demand of the i-th charging pile; Q i is the total number of new energy vehicles; λ i is the vehicle-to-charging pile ratio of new energy vehicles.

10. A method for optimizing the allocation of regional charging piles based on the energy substitution rate according to claim 1, characterized in that: The new construction demand for charging piles in the horizontal year sub-regions in Step 5 is specifically as follows: The annual new demand for charging piles is calculated based on the analysis of the charging pile demand in the planned horizontal year, combined with the number of new energy vehicle charging piles each year, considering the service life of new energy vehicle charging piles. The calculation method for the number of new charging piles required each year is to add the new charging pile demand and the charging piles added k years ago as the new charging piles required in the planned year. The calculation method is as follows: ΔN i = N i - N i-1 +(N i-k - N i-k-1 ), where, ΔN i is the demand for new charging piles of type i in the planned year; N i is the demand for charging piles in the i-th year; k is the service life of the charging piles; N i-k is the demand for charging piles in the (i - k)-th year; N i-k-1 is the demand for charging piles in the year before the (i - k)-th year.

Images