Medium-and-long-term electricity hydrogen energy charging demand prediction method for electricity hydrogen production and charging integrated station based on diffusion model

Abstract

The invention discloses a medium-and-long-term electricity hydrogen energy charging demand prediction method for an electricity hydrogen production and charging integrated station based on a diffusion model, which aims at prediction of the inventory of regional new energy vehicles and annual total charging and hydrogenation demands, and considers the development condition of regional electric vehicles and related policy goals of the region. The medium-and-long-term electricity hydrogen energy charging demand prediction method for the electricity hydrogen production and charging integrated station based on the diffusion model is provided, and data support is provided for medium-and-long-term planning of the electricity hydrogen production and charging integrated station.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric hydrogen charging demand prediction, and specifically relates to a method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model. Background Art

[0002] As a clean and efficient secondary energy source, hydrogen energy plays an increasingly important role in the global energy transformation. The transportation field is currently a relatively mature field for hydrogen energy applications. As an important infrastructure for hydrogen energy transportation, hydrogen refueling stations are important hubs that connect upstream hydrogen production and transportation and downstream applications, and play a vital role. At present, 90% of the hydrogen refueling stations built in China basically adopt the off-site hydrogen supply mode, while the model of integrated hydrogen production and hydrogenation stations in the station reduces the intermediate storage and transportation links, saves the expensive hydrogen transportation costs, and can effectively solve the problem of high costs caused by hydrogen storage and transportation. It can realize the on-site production and use of hydrogen and self-sufficiency, which can effectively reduce the hydrogen source cost of vehicle hydrogen refueling stations and the safety problems caused by hydrogen transportation.

[0003] As a key link in the planning of electric-hydrogen integrated stations, the existing forecast of electric-hydrogen charging demand has some defects. The hydrogen fuel cell vehicle ownership prediction model often relies on many assumptions and subjective judgments, lacks sufficient data support and empirical testing, resulting in limited accuracy and reliability of the prediction results. At the same time, compared with the mature electric vehicle market, the hydrogen fuel cell vehicle market started late, and the accumulation of relevant data is insufficient, which limits the refinement of the model and the prediction accuracy. Although existing studies mainly analyze the economic and environmental benefits brought by the coupling of electric hydrogen energy and transportation from the perspective of operation and planning, there are relatively few studies on the specific application scenarios of hydrogen energy. In particular, the prediction of hydrogen fuel vehicle ownership and hydrogen refueling demand in specific regions, as well as the prediction technology of electric-hydrogen charging demand coupled with the two, are still relatively scarce.

[0004] Therefore, a reasonable forecast of electric hydrogen charging demand is crucial for the planning of charging and hydrogenation integrated stations. It helps to optimize the investment decision and facility layout of the integrated station, ensure the effectiveness of grid load management, and guide the reasonable selection of electric hydrogen integrated station technology and equipment. In addition, the forecast of electric hydrogen charging demand can also improve the operational efficiency of the integrated station, assist relevant government departments in formulating policies, respond to market changes in a timely manner, ensure the economic benefits of the electric hydrogen integrated station, avoid waste of resources, and promote the healthy development of the new energy vehicle industry, thereby providing data technology support for the long-term planning of the electric hydrogen integrated station. Summary of the invention

[0005] The purpose of the present invention is to design a method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model. The present invention is based on the historical relevant data and development status of regional electric vehicles and hydrogen fuel cell vehicles, combined with regional policy planning goals, and uses the selected optimal diffusion model to predict the future ownership of new energy vehicles in the region. At the same time, it integrates the development and changing trends of the penetration rates of different types of new energy vehicles to realize the prediction of the electric hydrogen charging demand of electric vehicles and hydrogen fuel cell vehicles in the region.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model, characterized in that: the method is as follows:

[0008] S1: Collect historical data on electric vehicles (EV) and hydrogen fuel cell vehicles (HFCV) in a certain area;

[0009] S2: Use several different diffusion models to fit the historical number of electric vehicles (EVs). The usual steps are to collect historical data related to the number of electric vehicles, and then preprocess the data, such as data cleaning, processing missing values, outliers, etc., and confirm the time interval of the data, such as annual or monthly data. Then use programming languages ​​and simulation platforms to estimate and set the parameters of each model, and finally apply each diffusion model to the data set, and appropriately adjust the parameters of the model to obtain the best fitting effect.

[0010] S3: Based on the fitting results of different diffusion models in S2, the fitting effects of different models are measured based on the mean absolute error, and the diffusion model with the best fitting effect is selected based on the cumulative error of the fitting;

[0011] S4: Based on the current status of electric vehicle (EV) development and the policy objectives of the region, determine the relevant quantity data of electric vehicles and hydrogen fuel cell vehicles. According to the historical trend changes of electric vehicle ownership in the region and the regional ownership of electric vehicles and hydrogen fuel cell vehicles planned in the relevant policy objectives, determine the new energy vehicle ownership data for the specific planning year, and apply the data to the optimal diffusion model for operation; that is, according to the relevant documents of regional policies, there are statements related to the planned number of electric vehicles and hydrogen fuel cell vehicles designed in the planning objectives for the specific year, the ownership should reach the expected amount, and it should enter a stable period in the specified year. In this way, the development status of new energy vehicles, that is, the ownership planning target at the corresponding time node, is constructed, and the optimal diffusion model is selected. The diffusion model contains various operating parameters, such as n (potential market demand), j (growth rate), k (inflection point) of the Gompertz model, etc. Therefore, it is necessary to set the model's operating parameters according to the constructed development scenario so that the situation corresponding to the model fitting is consistent with the scenario.

[0012] S5: Use the optimal diffusion model to predict the number of electric vehicles (EV) and hydrogen fuel cell vehicles (HFCV) in the forecast year;

[0013] S6: Based on different penetration rates, the ownership of each type of electric vehicle EV and hydrogen fuel cell vehicle HFCV in the forecast year is predicted respectively. According to the above-mentioned total ownership forecast of each type of new energy vehicle in the forecast year and the regional automobile market policy planning target for the penetration rate of each type of new energy vehicle, the ownership penetration rate of each type of new energy vehicle in the forecast year is reasonably set. Based on the setting of different market penetration rates, the operating parameters of the optimal prediction model of different types of new energy vehicles are constructed, and then the ownership forecast based on different penetration rates is carried out; the setting of the new energy vehicle ownership penetration rate here is based on the market share change trend of various types of electric vehicles and hydrogen fuel cell vehicles mentioned in the policy planning target of the region, such as the private car share and bus share of electric vehicles mentioned in the planning target. Therefore, the parameter setting should meet the development proportion trend of various types of new energy vehicles in the policy target and be consistent with the market application proportion of various types of new energy vehicles in the region in the future.

[0014] S7: Based on the number of electric vehicles (EVs) of each type in the above steps, the total annual charging demand for EVs in the region in the predicted year is predicted. Based on the statistical annual charging demand for electric vehicles of different models, it is assumed that the annual charging demand for electric vehicles of each model remains unchanged over a certain time scale. Based on the predicted results of the number of electric vehicles of different types, the total annual charging demand for EVs in the region in the predicted year is cumulatively calculated.

[0015] S8: Based on the number of HFCVs of various types in the above steps, the total annual hydrogen refueling demand for HFCVs in the region in the predicted year is predicted. Based on the statistical annual hydrogen refueling demand for individual hydrogen fuel cell vehicles of different models, it is assumed that the annual hydrogen refueling demand for each model of hydrogen fuel cell vehicle remains unchanged over a certain time scale. Based on the predicted results of the number of different types of hydrogen fuel cell vehicles, the total annual hydrogen refueling demand for HFCVs in the region in the predicted year is calculated cumulatively.

[0016] Furthermore, the historical data of electric vehicles (EV) and hydrogen fuel cell vehicles (HFCV) include historical ownership data of different types of EVs and HFCVs in the region, annual average charging demand data, policy target data for new energy vehicles, and market penetration planning data for various types of vehicles.

[0017] Furthermore, the different diffusion models in S2 are Gompertz, Logistic and Bass. Each model has a different mathematical form and reasonable operating parameters need to be set respectively.

[0018] The Gompertz model was originally used to describe how human mortality changes with age, and was later extended to other fields, such as biological growth, tumor growth, market growth, etc. The Gompertz curve shows an S-shaped growth, which starts slowly, accelerates in the middle, and then gradually slows down until it approaches saturation.

[0019] The logistic model was originally used to describe population growth under limited resource conditions. It takes into account the impact of resource limitations on growth and has been used in other fields. The logistic model also presents an S-shaped growth curve, but compared with the Gompertz model, the growth rate of the logistic model decreases more slowly when approaching saturation.

[0020] The Bass model is mainly used in the study of marketing and new product diffusion. It predicts the acceptance and market penetration of new products by considering the innovativeness and imitativeness of different consumer groups. The Bass model can predict the market acceptance process of new products, from slow growth in the early stage to rapid growth, and then to the saturation stage in the later stage, showing an S-shaped growth curve.

[0021] As a durable good, the number of cars in use shows a low growth rate when the per capita income level is relatively low, but it will gradually accelerate as the income level increases. As the number of cars in use increases, the satisfaction brought by buying a car gradually weakens. In addition, the use cycle of durable goods is long. Therefore, people's consumption of cars will gradually enter a period of deceleration growth after reaching a certain level. At this time, although the ownership rate is still growing, the growth rate is constantly slowing down. The number of cars in use will go through three stages of accelerated growth, decelerated growth and saturation. From the above analysis, it can be seen that the general trend of car ownership meets the curve of the above diffusion model, so we choose the above diffusion model for overall prediction.

[0022] Based on the above-mentioned collection and collation of historical data on electric vehicles in this area, appropriate parameters of different diffusion models are set to effectively fit the historical data.

[0023] The Gompertz model can be used to describe the following phenomenon, that is, the growth is slow at first, then gradually accelerates, reaches a certain level, and finally slows down and approaches the asymptote; its mathematical model is expressed as:

[0024]

[0025] In the formula, t represents time, N(t) represents the number of vehicles in use, n represents potential market demand, j represents growth rate, and k represents the inflection point.

[0026] The Logistic model is a mathematical description of the S-curve and is one of the typical diffusion models. Its mathematical model is expressed as:

[0027]

[0028] In the formula, a represents the slope factor, b represents the growth rate, and c represents the potential market demand.

[0029] The Bass model is often used to predict the market share of innovative products and technologies. Its mathematical model is expressed as:

[0030]

[0031] In the formula, p is the innovation coefficient, q is the imitation coefficient, and m is the potential market demand.

[0032] According to the fitting results of the above three different models, and based on the mean absolute error, the fitting effects of different models are measured, and the optimal model is selected for subsequent predictions based on the cumulative error of the fitting.

[0033] The mean absolute error (MAE) is used to reflect the deviation between the predicted number and the actual number. The smaller the MAE value, the better the fitting effect. Its mathematical model is expressed as:

[0034]

[0035] Where E represents the mean absolute error, T represents the total number of predictions, and N'(t) represents the actual value at time t.

[0036] Furthermore, the Gompertz model has the smallest mean absolute error and is the optimal diffusion model;

[0037] Furthermore, S4 needs to comprehensively consider the development status of EV and HFCV in the region and related planning objectives, analyze and compare the similarities and differences in the development of the two types of vehicles, the construction of related supporting technologies, and the layout of the industry, make reasonable plans, and construct development scenarios for EV and HFCV.

[0038] Furthermore, the various types of electric vehicles EV in S7 include buses, trucks, taxis, and private cars.

[0039] Furthermore, the penetration rate of different types of cars in S6 refers to the proportion of a certain type of car ownership in a region to the total number of cars in the region within a certain period of time. It reflects the popularity and market acceptance of this type of car in the regional market. Its mathematical model is expressed as:

[0040]

[0041] Where n irepresents the number of a certain type of car in the area, N represents the total number of all types of cars in the area, and α i Represents the penetration rate of a certain type of car in the area.

[0042] The following beneficial effects can be obtained through the above technical solution:

[0043] The present invention aims at predicting the regional new energy vehicle ownership and the annual total charging and hydrogenation demand, taking into account the regional electric vehicle development situation and the relevant policy objectives of the region, and proposes a medium- and long-term electric hydrogen charging demand prediction method for electric hydrogen production and filling integrated stations based on a diffusion model, providing data support for the medium- and long-term planning of electric hydrogen production and filling integrated stations.

[0044] Based on the historical data and development status of regional electric vehicles and hydrogen fuel cell vehicles, combined with regional policy planning objectives, the selected optimal diffusion model is used to predict the future ownership of new energy vehicles in the region. At the same time, the development and changing trends of the penetration rates of different types of new energy vehicles are integrated to realize the prediction of the electric and hydrogen charging demand of electric vehicles and hydrogen fuel cell vehicles in the region, providing technical support for the planning of electric and hydrogen integrated stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Flowchart of the forecast method for the number of new energy vehicles and charging and hydrogenation demand;

[0046] Figure 2 The results of different diffusion models fitting the electric vehicle ownership from 2014 to 2022;

[0047] Figure 3 The forecast results of hydrogen fuel cell vehicle ownership under the optimistic scenario;

[0048] Figure 4 Prediction results of different types of electric vehicles ownership;

[0049] Figure 5 The forecast results of annual charging demand for different types of electric vehicles;

[0050] Figure 6 Prediction results of different types of hydrogen fuel cell vehicles;

[0051] Figure 7 Forecast results of annual hydrogen refueling demand for different types of hydrogen fuel cell vehicles. DETAILED DESCRIPTION

[0052] The present invention will be further described below in conjunction with the accompanying drawings:

[0053] Example: According to Figure 1-7As shown in the figure, a method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model is provided. In practical application, the method process is as follows:

[0054] (1) Collect historical data on EV and HFCV in a certain area;

[0055] This paper collects data on the number of electric vehicles and hydrogen fuel cell vehicles in Shanghai from 2014 to 2022. The data comes from the Shanghai Statistics Bureau and the Shanghai Transportation Network. As shown in Table 1, the number of electric vehicles has grown steadily from 2014 to 2019. As the electric vehicle industry matures in the Shanghai area, the number of electric vehicles will grow rapidly from 2020 to 2022. On the other hand, the HFCV industry in Shanghai is still in its infancy, and the number of HFCVs has grown slowly from 2017 to 2022.

[0056] Table 1 Number of electric vehicles and hydrogen fuel cell vehicles in Shanghai

[0057]

[0058] (2) Use three different models, Gompertz, Logistic and Bass, to fit the EV historical maintenance volume;

[0059] The Gompertz model can be used to describe the following phenomenon, that is, the growth is slow at first, then gradually accelerates, reaches a certain level, and finally slows down and approaches the asymptote; its mathematical model is expressed as:

[0060]

[0061] In the formula, t represents time, N(t) represents the number of vehicles in use, n represents potential market demand, j represents growth rate, and k represents the inflection point.

[0062] The Logistic model is a mathematical description of the S-curve and is one of the typical diffusion models. Its mathematical model is expressed as:

[0063]

[0064] In the formula, a represents the slope factor, b represents the growth rate, and c represents the potential market demand.

[0065] The Bass model is often used to predict the market share of innovative products and technologies. Its mathematical model is expressed as:

[0066]

[0067] In the formula, p is the innovation coefficient, q is the imitation coefficient, and m is the potential market demand.

[0068] Table 2 Different model setting parameters and prediction error results

[0069]

[0070] According to the results of fitting different diffusion models to the electric vehicle ownership from 2014 to 2022, the Gompertz model has the smallest cumulative fitting error and the best fitting effect for the historical ownership of electric vehicles.

[0071] (3) According to the above results, based on the cumulative error of fitting, select the diffusion model with the best fitting effect;

[0072] According to the above prediction results, it can be seen from Table 2 that the Gompertz model has the smallest average absolute error and fits the changing trend of the historical ownership of electric vehicles very well. It has the best prediction effect. Therefore, the Gompertz model is selected for the prediction simulation of the future ownership of subsequent electric vehicles and hydrogen fuel cell vehicles.

[0073] (4) Construct development scenarios for EVs and HFCVs based on the current status of electric vehicle development and Shanghai’s policy objectives;

[0074] That is, according to the relevant documents of Shanghai regional policies, there are planning targets mentioned in the planning targets, such as how many electric vehicles and hydrogen fuel cell vehicles should be planned in a certain year, how many should be owned, and when should it enter a stable period. In this way, the development scenario of new energy vehicles is constructed, that is, the planning target of the ownership at the corresponding time node. Taking into account the development status of electric vehicles in Shanghai, combined with the goals proposed in the "Shanghai Fuel Cell Vehicle Development Plan" and the "Shanghai Hydrogen Energy Industry Development Medium- and Long-Term Plan", the future ownership trend of new energy vehicles is set under an optimistic scenario.

[0075] In the optimistic scenario, based on the goals set out in the Shanghai Fuel Cell Vehicle Development Plan, it is expected that the number of HFCVs in Shanghai will reach 30,000 by 2025, combining this planning target with the historical number of HFCVs from 2017 to 2022.

[0076] (5) Use the optimal diffusion model to predict the number of EVs in the forecast year;

[0077] It can be seen from the forecast results that under the optimistic scenario, the number of electric vehicles in 2025-2030 will grow rapidly, and the overall number of electric vehicles will increase significantly. By 2035, the number of EVs in use will reach about 5.9 million.

[0078] (6) Use the optimal diffusion model to predict the number of HFCVs in the predicted year;

[0079] Under the optimistic scenario, the number of HFCVs could reach 390,000 by 2035, exceed 1 million by 2040, and reach 7.52 million by 2050. Therefore, compared with electric vehicles, more policy support is needed to achieve the forecast target for HFCVs in Shanghai.

[0080] (7) Forecast the number of EVs of each type in the forecast year based on different penetration rates;

[0081] According to relevant literature, taking into account the market penetration rates of different types of EVs, under an optimistic scenario, the penetration rates of buses in 2025, 2035 and 2050 will be 3%, 3% and 2% respectively, the penetration rates of trucks will be 2%, 2% and 1% respectively, the penetration rates of taxis will be 10%, 5% and 1% respectively, and the penetration rates of private cars will be 85%, 90% and 95% respectively.

[0082] According to the forecast results, among the four types of vehicles, the future ownership of private cars will grow fastest, from nearly 1.4 million in 2025 to nearly 4 million in 2030, and may exceed 5.5 million in 2035. Compared with private cars, the other three types of vehicles, buses, trucks, and taxis, grow more slowly, with only a small increase in ownership. The total ownership of the three types may reach about 500,000 by 2035.

[0083] (8) Forecast the number of HFCVs of various types in the forecast year based on different penetration rates;

[0084] Common HFCVs include buses, trucks, taxis and private cars. The supporting facilities for Shanghai's hydrogen fuel cell vehicle industry are not yet complete, and the main models are buses and trucks. According to the "Shanghai Hydrogen Energy Industry Development Medium- and Long-Term Plan", it is expected that by 2035, Shanghai will complete the layout of HFCV supporting industries. By then, the number of private cars will increase rapidly. According to the literature, taking into account the penetration rate of different types of HFCVs, under the optimistic scenario, the penetration rate of buses in 2025, 2035 and 2050 will be 40%, 10% and 1% respectively, the penetration rate of trucks will be 50%, 10% and 1% respectively, the penetration rate of taxis will be 5%, 30% and 8% respectively, and the penetration rate of private cars will be 5%, 50% and 90% respectively.

[0085] Table 3 Gompertz model parameters of different types of HFCV

[0086]

[0087] According to the forecast results, among the four types of vehicles, the number of private cars and taxis will grow faster in the future, from only 7,000 in 2025 to nearly 20,000 in 2030, and may exceed 170,000 in 2035. Compared with private cars and taxis, the number of the other two types of vehicles, buses and trucks, will grow more slowly, and the number of both is expected to exceed 40,000 in 2035.

[0088] (9) Based on the number of EVs of various types, forecast the total annual charging demand for EVs in the region in the forecast year;

[0089] According to the statistics of the annual charging demand of different types of EVs in the literature, assuming that the annual charging demand of each type of EV remains unchanged, the forecast results of the annual total charging demand of different types of electric vehicles are obtained. From the forecast results, it can be seen that the annual total charging demand of private cars and buses is growing rapidly, while the annual total charging demand of trucks and taxis is growing at a slower rate, but there is still a significant increase. By 2035, the annual total charging demand of electric vehicles can reach 26.26TWh.

[0090] (10) Based on the inventory of various types of HFCVs, the total annual hydrogenation demand for HFCVs in the region in the forecast year is predicted.

[0091] According to the statistics of the annual hydrogenation demand of different types of HFCV in the literature, assuming that the annual hydrogenation demand of each type of HFCV remains unchanged, the forecast results of the annual total hydrogenation demand of different types of HFCV under the optimistic scenario are obtained. It can be seen from the forecast results that under the optimistic scenario, although the future ownership of buses and trucks will grow slowly, due to the characteristics of their large hydrogenation demand, the annual total hydrogenation demand of the two models will still maintain a rapid growth trend. By 2035, the annual total hydrogenation demand of the two may reach 300 kilotons. However, due to the small demand for hydrogenation of taxis and private cars, even if the ownership grows very rapidly, the annual total hydrogenation demand will still only account for a small proportion. By 2035, the annual total hydrogenation demand of taxis and private cars will be about 22 kilotons. At the same time, it can be seen that by 2035, the annual total hydrogenation demand of hydrogen fuel cell vehicles can reach 320 kilotons.

[0092] The above are all preferred embodiments of the present invention. For ordinary technicians in this technical field, without departing from the principle of the present invention, various equivalent modifications to the present invention belong to the protection scope of the claims attached to this application.

Claims

1. A method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model, characterized in that: The method is as follows: S1: Collect historical data on electric vehicles (EV) and hydrogen fuel cell vehicles (HFCV) in a certain area; S2: Use several different diffusion models to fit the historical ownership of electric vehicles (EVs); S3: Based on the fitting results of different diffusion models in S2, the fitting effects of different models are measured based on the mean absolute error, and the diffusion model with the best fitting effect is selected based on the cumulative error of the fitting; S4: Combine the current status of electric vehicle (EV) development and the policy objectives of the region to determine the relevant quantity data of electric vehicles and hydrogen fuel cell vehicles. According to the historical trend of electric vehicle ownership in the region and the regional ownership of electric vehicles and hydrogen fuel cell vehicles planned in the relevant policy objectives, determine the new energy vehicle ownership data for the specific planning year, and apply the data to the optimal diffusion model for operation; S5: Use the optimal diffusion model to predict the number of electric vehicles (EV) and hydrogen fuel cell vehicles (HFCV) in the forecast year; S6: Forecast the ownership of each type of electric vehicle EV and hydrogen fuel cell vehicle HFCV in the forecast year based on different penetration rates. According to the above forecast of the total ownership of each type of new energy vehicle in the forecast year and the regional automobile market policy for the penetration rate planning data of each type of new energy vehicle, set the ownership penetration rate of each type of new energy vehicle in the forecast year. Based on the setting of different market penetration rates, construct the operating parameters of the optimal prediction model of different types of new energy vehicles, and then conduct the ownership prediction based on different penetration rates. S7: Based on the number of electric vehicles (EVs) of each type in the above steps, the total annual charging demand for EVs in the region in the predicted year is predicted. Based on the statistical annual charging demand for electric vehicles of different models, it is assumed that the annual charging demand for electric vehicles of each model remains unchanged over a certain time scale. Based on the predicted results of the number of electric vehicles of different types, the total annual charging demand for EVs in the region in the predicted year is cumulatively calculated. S8: Based on the number of HFCVs of various types in the above steps, the total annual hydrogen refueling demand for HFCVs in the region in the predicted year is predicted. Based on the statistical annual hydrogen refueling demand for individual hydrogen fuel cell vehicles of different models, it is assumed that the annual hydrogen refueling demand for each model of hydrogen fuel cell vehicle remains unchanged over a certain time scale. Based on the predicted results of the number of different types of hydrogen fuel cell vehicles, the total annual hydrogen refueling demand for HFCVs in the region in the predicted year is calculated cumulatively.

2. According to the method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 1, it is characterized by: The historical data of electric vehicles (EVs) and hydrogen fuel cell vehicles (HFCVs) include the historical ownership data of different types of EVs and HFCVs in the region, the annual average charging demand data, the policy target data of new energy vehicles, and the market penetration planning data of various types of vehicles.

3. According to the method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 1, it is characterized by: The different diffusion models in S2 are Gompertz, Logistic and Bass. Each model has a different mathematical form and requires reasonable operating parameters to be set separately.

4. According to claim 3, a method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model is characterized by: The Gompertz curve shows an S-shaped growth, which starts slowly, accelerates in the middle, and then gradually slows down until it approaches saturation.

5. According to claim 3, a method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model is characterized by: The Logistic model also presents an S-shaped growth curve, but compared with the Gompertz model, the growth rate of the Logistic model decreases more slowly when approaching saturation.

6. According to claim 3, a method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model is characterized by: The Bass model can predict the market acceptance process of new products, from slow growth in the early stage to rapid growth and then to the saturation stage in the later stage, presenting an S-shaped growth curve.

7. A method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 3 or 4, characterized in that: The Gompertz model has the smallest mean absolute error and is the optimal diffusion model; The Gompertz model can be used to describe the following phenomenon: growth is slow at first, then gradually accelerates, reaches a certain level, and finally slows down and approaches an asymptote; Its mathematical model is expressed as: In the formula, t represents time, N(t) represents the number of vehicles in use, n represents potential market demand, j represents growth rate, and k represents the inflection point.

8. The method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 3 or 5, characterized in that: The Logistic model is a mathematical description of the S-curve and is one of the typical diffusion models. Its mathematical model is expressed as: In the formula, a represents the slope factor, b represents the growth rate, and c represents the potential market demand.

9. A method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 3 or 6, characterized in that: The Bass model is often used to predict the market share of innovative products and technologies. Its mathematical model is expressed as: In the formula, p is the innovation coefficient, q is the imitation coefficient, and m is the potential market demand.

10. The method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 1 is characterized by: In S4, it is necessary to comprehensively consider the development status of EV and HFCV in the region and related planning objectives, analyze and compare the similarities and differences in the development of the two types of vehicles, the construction of related supporting technologies and the layout of the industry, make reasonable plans, and construct development scenarios for EV and HFCV.

11. The method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 1 is characterized by: The various types of electric vehicles EV in S7 include buses, trucks, taxis, and private cars.

12. The method for predicting the medium- and long-term electric hydrogen charging demand of an electric hydrogen production and filling integrated station based on a diffusion model according to claim 1 is characterized by: The penetration rate of different types of cars in S6 refers to the proportion of a certain type of car in the region to the total number of cars in the region within a certain period of time. It reflects the popularity and market acceptance of this type of car in the regional market. Its mathematical model is expressed as: Where n i represents the number of a certain type of car in the area, N represents the total number of all types of cars in the area, and α i Represents the penetration rate of a certain type of car in the area.

Images