Hydrogen consumption load development prediction method based on LEAP model

Abstract

The invention discloses a hydrogen consumption load development prediction method based on an LEAP model. The method comprises the following steps: S1, establishing an LEAP model; a plurality of hydrogen using departments are arranged in the LEAP model; s2, acquiring energy data of a plurality of hydrogen use departments through a data acquisition system; s3, establishing an influence factor prediction model, and obtaining an influence factor prediction result based on the influence factor prediction model; and S4, substituting the energy data and the influence factor prediction result into the LEAP model, and solving to obtain hydrogen load prediction results of a plurality of hydrogen use departments. On the basis of the LEAP model, the global terminal hydrogen energy demand is divided into a plurality of hydrogen use departments, and data processing and correction are performed on hydrogen use load conditions of different scenes and time dimensions through data input of each dimension so as to obtain a more accurate prediction result, so that the new energy utilization efficiency can be enhanced, and the energy consumption can be reduced. And a direction and a basis are provided for future energy system planning.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing for management or prediction purposes, and in particular to a method for predicting the development of hydrogen load based on a LEAP model. Background Art

[0002] As the global demand for energy continues to increase, hydrogen energy, as an important future energy source, has received widespread attention.

[0003] From a global perspective, the key core technologies of the entire hydrogen energy industry chain are becoming mature, the construction of hydrogen energy infrastructure has been significantly accelerated, and a regional hydrogen energy supply network is taking shape. my country's hydrogen energy industry is in its early stages of development, and the industry's development form and development path need to be gradually improved. Therefore, it is necessary to make corresponding adjustments or settings based on the development forecast results of hydrogen load. However, the existing hydrogen load development forecasting methods have many problems in data collection and forecasting, which limits the accuracy and practicality of the forecasting model.

[0004] At present, load forecasting often uses gray model method, time series method, support vector machine method and optimization algorithm that combines and innovates various algorithms. These algorithms have high prediction accuracy and have been widely used in power systems, natural gas and other fields. For hydrogen load forecasting, there have been certain research results at home and abroad. Among them, some research results are aimed at hydrogen supply chain systems, considering the growth trend of hydrogen demand in long-term network design models. For example, Chinese patent CN113743680B discloses a method, device and storage medium for predicting hydrogen energy demand: obtain the first demand output by the prediction model for hydrogen refueling stations in the target area, determine the transaction record set of hydrogen energy vehicles in the target area within a preset period of time, and determine the actual demand based on the first demand and the transaction record set. Send the actual demand to the hydrogen production plant so that the hydrogen production plant prepares hydrogen energy according to the actual demand to avoid excessive or insufficient preparation. Although this technical solution predicts the use of hydrogen energy for a specific field scenario, the prediction accuracy of this method is limited and it cannot cope with hydrogen load prediction at a larger level. In addition, some studies have conducted qualitative analysis on the internal and external factors of hydrogen demand to predict the hydrogen demand of various industries under different economic development speeds. However, with the rapid development of hydrogen energy, the use of hydrogen scenarios has gradually increased. The macro-forecast of hydrogen load is difficult to meet the demand for hydrogen energy in various industries, and the forecast results are also difficult to provide more reliable technical support in the comprehensive evaluation at the macro level.

[0005] In summary, although there are a large number of technical solutions available for long-term hydrogen load forecasting methods for specific scenarios, how to accurately predict the hydrogen load under different scenarios is a key issue that needs to be solved urgently. Summary of the invention

[0006] The present invention provides a method for predicting the development of hydrogen load based on the LEAP model to solve the above problems.

[0007] The present invention adopts the following technical solutions to solve the above technical problems:

[0008] The present invention provides a method for predicting the development of hydrogen load based on the LEAP model, the steps comprising:

[0009] S1. Establishing a LEAP model; wherein the LEAP model is provided with multiple hydrogen-using departments;

[0010] S2. Collecting energy data of the multiple hydrogen-using departments through a data acquisition system;

[0011] S3, establishing an influencing factor prediction model, and obtaining an influencing factor prediction result based on the influencing factor prediction model;

[0012] S4. Substitute the energy data and the prediction results of the influencing factors into the LEAP model, and solve to obtain the hydrogen load prediction results of the multiple hydrogen-using departments.

[0013] Optionally, also include:

[0014] S5. Classifying the hydrogen load prediction results based on the multiple hydrogen-using departments and obtaining classified prediction results;

[0015] S6. Calculating energy efficiency based on the classification prediction results;

[0016] The energy benefits include economic benefits, new energy consumption benefits and carbon reduction benefits.

[0017] Optionally, in the LEAP model, the multiple hydrogen-using departments include an industrial department, a power department, a transportation department and a heating department.

[0018] Optionally, each hydrogen-using department is further divided into multiple sub-departments, and each sub-department has a corresponding energy type.

[0019] Optionally, the energy types include at least hydrogen energy.

[0020] Optionally, the plurality of hydrogen-using departments are each provided with corresponding activity levels and energy efficiency parameters;

[0021] The activity level corresponding to the industrial sector is industrial output value, and the energy efficiency parameter is energy consumption per unit output value;

[0022] The activity level corresponding to the power sector is power generation, and the energy efficiency parameter is energy consumption per unit of power generation;

[0023] The activity levels corresponding to the transportation sector are vehicle ownership and vehicle mileage, and the energy efficiency parameter is energy consumption per 100 kilometers;

[0024] The activity level corresponding to the heating department is the area of ​​the heating region and the permanent population of the region, and the energy efficiency parameters are the energy consumption per unit area of ​​heating and the per capita heating energy consumption.

[0025] Optionally, the S2 further includes: performing data preprocessing on the energy data;

[0026] The data preprocessing includes data cleaning and data noise reduction.

[0027] Optionally, the influencing factor prediction model is a two-stage hybrid prediction model, and the two-stage hybrid prediction model establishment process further includes:

[0028] S31. Establish a first-order model based on multi-head attention mechanism and one-dimensional convolutional neural network;

[0029] S32. Establish a second-order linear regression model based on influencing factors.

[0030] Optionally, the influencing factors include constraint conditions and cost conditions;

[0031] The constraints include hydrogen energy utilization rate, hydrogen energy industry investment and carbon emissions;

[0032] The forecast results of the influencing factors include expected hydrogen energy utilization rate, expected hydrogen energy industry investment, expected carbon emissions and expected cost conditions.

[0033] Optionally, in S4, the LEAP model includes a baseline scenario and an auxiliary scenario;

[0034] Substituting the energy data into the baseline scenario;

[0035] The prediction result of the influencing factor is substituted into the auxiliary scenario.

[0036] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0037] Based on the LEAP model, the present invention divides the global terminal hydrogen energy demand into multiple hydrogen-using departments. Through the data input of various dimensions, the hydrogen load conditions in different scenarios and time dimensions are processed and corrected, and more accurate prediction results are obtained, which is conducive to improving the utilization efficiency of new energy and provides direction and basis for future energy system planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 is a flow chart of a method in a specific embodiment of the present invention;

[0040] Figure 2 is a structural diagram of a LEAP model in a specific embodiment of the present invention;

[0041] Figure 3 It is a LEAP model calculation roadmap in a specific embodiment of the present invention;

[0042] Figure 4 It is a display diagram of the prediction results of the industrial field module of Qinghai Province in a specific embodiment of the present invention;

[0043] Figure 5 It is a display diagram of the prediction results of the power field module in Qinghai Province in a specific embodiment of the present invention;

[0044] Figure 6 It is a display diagram of the prediction results of the Qinghai Province transportation field module in a specific embodiment of the present invention;

[0045] Figure 7 It is a display diagram of the prediction results of the heating field module in Qinghai Province in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0048] It should be noted that the methods used in the present invention are all conventional methods unless otherwise specified; the raw materials and devices used are all conventional commercially available products unless otherwise specified.

[0049] This embodiment provides a method for predicting the development of hydrogen load based on the LEAP model. Figure 1 As shown, the main steps include:

[0050] S1. Establish the LEAP model; the full name of the LEAP model is the Long-Rage Energy Alternatives Planning System, which is an energy policy analysis and climate change assessment model. It is established using the domain activity analysis method and is a bottom-up energy-environment accounting tool based on scenario analysis. The LEAP model includes energy supply, energy processing and conversion, and terminal energy demand. It can be mainly used in energy planning or forecasting, energy balance sheets and environmental pollutant inventories, greenhouse gas emission reduction analysis, and comprehensive resource utilization planning. Because the LEAP model has a flexible and diverse structure, it can flexibly construct models and data structures according to the characteristics of the research object, the availability of data, and the purpose and type of analysis, and analyze energy consumption and greenhouse gas emissions in different scenarios.

[0051] In the LEAP model structure of this embodiment, multiple hydrogen-using departments are provided; Figure 2 As shown, in the LEAP model, the baseline scenario is constructed, and its hydrogen-using sectors include the industrial sector, the power sector, the transportation sector, and the heating sector. In order to fully build the scenario, each sector is refined, so that each hydrogen-using sector is also divided into multiple sub-sectors, and each sub-sector has a corresponding energy type. Multiple hydrogen-using sectors are set with corresponding activity levels and energy efficiency parameters. Based on the premise of hydrogen energy analysis, it is also required that the corresponding energy type of each sector includes at least hydrogen energy.

[0052] Among them, the activity level corresponding to the industrial sector is industrial output value, and the energy efficiency parameter is energy consumption per unit of output value; further, the industrial field is divided into key industries and general industries according to industry characteristics, and the key industries are subdivided into the steel industry and the chemical industry. The activity level is represented by industrial output value, and the energy consumption level is represented by energy per unit of output value. The energy types are electricity, heat, natural gas, diesel and hydrogen.

[0053] The corresponding activity level of the power sector is power generation, and the energy efficiency parameter is energy consumption per unit of power generation; further, the power sector is divided into hydrogen energy storage, hydrogen-fired units, and hydrogen distributed power stations according to the type of hydrogen-containing units. The activity level is characterized by power generation (discharge), and the energy consumption level is characterized by energy consumption per unit of power generation. The main energy types are coal, natural gas, and hydrogen.

[0054] The corresponding activity level of the transportation sector is the number of vehicles in use and vehicle mileage, and the energy efficiency parameter is energy consumption per 100 kilometers. Furthermore, the transportation sector is classified into four categories according to the types of transportation vehicles related to hydrogen energy: buses, large coaches, heavy trucks and cruise ships. The activity level is represented by the number of vehicles in use and mileage, and the energy consumption level is represented by energy consumption per 100 kilometers. The main energy types are electricity, gasoline and hydrogen.

[0055] The activity level corresponding to the heating sector is the area of ​​the heating area and the permanent population of the region, and the energy efficiency parameters are the energy consumption per unit area of ​​heating and the per capita heating energy consumption; further, the heating field is divided into public area heating and residential heating according to the nature of the heating area. Among them, the public building part only considers commercial areas, the activity level is represented by the building area, the energy consumption level is represented by the energy consumption per unit building area, and the main energy types are electricity and heat; for residential heating, considering that the research area is a provincial area, the urban and non-urban populations can also be subdivided, the activity level is represented by the permanent population, the energy consumption level is represented by the per capita energy consumption, and the main energy types are electricity, natural gas and hydrogen.

[0056] S2. The current data collection methods in the industrial field mainly rely on manual records and regular reporting, which have problems such as incomplete data collection, insufficient accuracy and lack of details; and the existing data collection methods are difficult to capture the key data of each link in the production process, resulting in a lack of depth and details in the data, and unable to fully reflect the changes in hydrogen load in the industrial production process. For example: in the power field, the lack of unified data standards and data islands are the main problems; in the transportation field, the problem of single data type is particularly prominent. The existing traffic data mainly focuses on the number of vehicles and mileage, and ignores the recording of other important factors (such as vehicle type, fuel consumption, etc.). The main problems in the heating field are incomplete data collection, low data update frequency and backward data processing methods. The existence of the above problems makes it difficult to fully reflect the changes in hydrogen load in the production process of various fields, affecting the accuracy of the prediction model.

[0057] On this basis, this embodiment collects energy data of multiple hydrogen-using departments through a data collection system; optionally, this embodiment uses an intelligent integrated data collection system (Intelligent Integrated Data Collection System, referred to as: IIDCS) to collect the data required by each department. Among them, IIDCS is a highly integrated and intelligent data collection platform, which is specially used to collect and analyze data in the four major fields of industry, electricity, transportation and heating. The system uses advanced sensor networks and Internet of Things devices to achieve real-time data collection and efficient processing. The following are the steps to collect energy data using IIDCS:

[0058] S21. System deployment: Identify relevant major data sources, including sensors, remote sensing equipment, Internet platforms, social media, etc. Install sensor networks and IoT devices in industrial production facilities, power systems, traffic management centers, and heating systems. Among them, the sensor network includes temperature sensors, flow sensors, pressure sensors, etc., which are used to monitor various key parameters in real time. IoT devices include smart meters, vehicle terminals, and heating metering equipment to ensure comprehensive coverage and real-time upload of data in various fields.

[0059] S22. Data collection: In the industrial field, sensor networks monitor production data and energy consumption data in real time, and collect information including the operating status of production lines of various types of hydrogen-consuming factories, equipment energy consumption, and raw material usage. In the power sector, IoT devices collect power generation data, power transmission and distribution data, and power usage in real time, including detailed data such as power generation, load curves, peak and trough power consumption. In the transportation sector, vehicle terminals and traffic management systems collect data such as the number of vehicles, mileage, fuel consumption, etc. in real time, covering multiple means of transportation such as public transportation, private cars, and freight vehicles. In the heating sector, heating metering equipment monitors the operating status and energy consumption of the heating system in real time, including the heating area, heat transmission efficiency, and user heat usage data.

[0060] An adaptive data collection strategy is adopted to adaptively adjust the data collection frequency and method according to the characteristics of the data source and the data quality. It also includes data preprocessing of the acquired primary energy data, including data cleaning and data noise reduction. For example, for highly variable data sources, the collection frequency is increased; for data sources with large noise, multiple sampling and data fusion technology are used to improve data quality.

[0061] S23. Data transmission and storage: All collected data is transmitted to the central database through the IoT network to achieve unified storage and management of data. Data transmission uses encryption technology to ensure data security and privacy protection.

[0062] S3. Since the LEAP model needs to calculate the hydrogen load based on the development of various fields in the current year, it is necessary to make detailed predictions on the development of various sectors in the future. In addition, the hydrogen industry is significantly affected by various factors during its development stage, so it is necessary to use the factors to correct the prediction results. Therefore, this embodiment establishes an influencing factor prediction model, and obtains the prediction results based on the influencing factor prediction model;

[0063] Specifically, the influencing factor prediction model of this embodiment is a two-stage hybrid prediction model, and the two-stage hybrid prediction model establishment process also includes:

[0064] S31. Establish a first-order model based on multi-head attention mechanism and one-dimensional convolutional neural network; in the prediction stage (first stage), use one-dimensional convolutional neural network (1DCNN) and add multi-head attention mechanism (Multi-headattention) to make the prediction results closer to the real results. Traditional CNN is mainly used to process two-dimensional data such as images, while 1DCNN is specially designed to process sequence data such as time series. Through convolution operation, 1DCNN can extract features at different time scales in the time dimension, which helps to better capture patterns and trends in time series data. In addition, by introducing the attention mechanism, the model can learn the importance weights of different time steps, so as to more effectively utilize the information in the sequence data.

[0065] Therefore, the specific steps are as follows:

[0066] S311, Hydrogen load prediction based on multi-head attention mechanism and 1DCNN; The energy data set collected in S2 is standardized, and the formula is:

[0067]

[0068] In the formula, X ij is the element in the i-th row and j-th column of the original data set X; is the element in the i-th row and j-th column after standardization; max(X ij )、min(X ij ) are used to find the maximum and minimum values ​​of the elements in the column vector of X respectively.

[0069] S312, one-dimensional convolution layer; through multiple convolution kernels W k Perform convolution operation to extract local features. The formula is:

[0070]

[0071] In the formula, is the output of the convolutional layer, b k For paranoid items.

[0072] S313, multi-head attention mechanism. Calculate the weights of multiple attention heads using the formula:

[0073]

[0074] In the formula, is the attention weight; is the attention score; w (h) 、b (h) is the weight parameter.

[0075] Therefore, the formula for the attention weighted sum c is:

[0076]

[0077] Where H is the number of attention heads.

[0078] S314, fully connected layer. The attention result is input into the fully connected layer for prediction. The formula is:

[0079]

[0080] In the formula, is the initial prediction result; W f is the weight matrix; b f is the bias term.

[0081] S32. In order to input the influencing factors into the model in mathematical form, it is necessary to select an appropriate method to quantify them. In this embodiment, the influencing factors mainly include various constraints and cost conditions. Among them, the constraints are mandatory requirements for the future development of hydrogen energy, such as hydrogen energy utilization rate, hydrogen energy industry investment and carbon emissions.

[0082] The following are examples of quantifying different types of influencers:

[0083] The first constraint condition is to convert the target and actual completion of the proportion of hydrogen energy products in the total products of the industry into time series data. For example, in the industrial field, the proportion of "hydrogen-to-ammonia" in the total ammonia products is selected; in the power field, the proportion of hydrogen-fired turbine installed capacity in the total installed capacity is selected, etc. It represents the proportion of hydrogen energy products in field k in year t.

[0084] The second constraint is to convert the investment amount in the hydrogen energy industry in the economic development plan into time series data. Specifically, the annual investment amount in the hydrogen energy industry is used. For example, in the transportation field, the investment amount in hydrogen fuel cell vehicles and hydrogen refueling stations is selected. represents the amount of government investment in field k in year t.

[0085] Cost conditions, the amount of subsidies for the hydrogen energy industry in the industry subsidy plan is converted into time series data. Specifically, the total amount of subsidies each year is used. For example, the amount of subsidies for hydrogen power generation in the power sector and the amount of subsidies for purchasing hydrogen vehicles in the transportation sector are selected. Represents the amount of hydrogen energy industry subsidies in field k in year t.

[0086] The third constraint condition is to convert the carbon emission standards in the four fields into time series data. For example, in the power sector, the carbon quota for power generation is selected, and in the heating sector, the emissions per unit area are selected. represents the national carbon quota for power generation in sector k in year t.

[0087] Thus, a second-order model of linear regression is established based on the influencing factors. In the correction stage (second stage), the linear regression model is used to correct the preliminary prediction results considering the above influencing factors.

[0088] Combining the above influencing factors, the influencing factor vector p that needs to be considered in field k is t,k for:

[0089]

[0090] Train a linear regression model, the mathematical form is:

[0091]

[0092] In the formula, is the final prediction result of the tth year in field k; β0 is the intercept term; β1 is the coefficient of the initial prediction result. ω is the coefficient of the influencing factor ω; Ω is the set of influencing factors, Ω={re,tz,bt,pf}.

[0093] Use the relevant data of recent years to build a training set, fit the model parameters, and obtain β0, β1, β ω The value of .

[0094] Get the final forecast results of the development of the four fields. Input the initial forecast results of the year to be forecasted and the influencing factor vector The forecast results corrected according to the influencing factors are:

[0095]

[0096] S4. Substitute the energy data and influencing factor prediction results into the LEAP model. Figure 3 The route shown is solved to obtain the hydrogen load forecast results of multiple hydrogen-consuming departments.

[0097] Among them, the LEAP model includes the baseline scenario and auxiliary scenario built earlier;

[0098] The energy data is substituted into the baseline scenario accordingly; while the prediction results of the aforementioned influencing factors are substituted into the auxiliary scenario.

[0099] Specifically, take Qinghai Province as an example:

[0100] S41. Data collection: Detailed data on the resource endowment, relevant plans, economic development, industrial conditions, etc. of the study area are collected by consulting national statistical yearbooks, consulting statistical yearbooks of the study area, collecting development plans of the study area in recent years, investigating the research results and related forecast results of major domestic and foreign industry research institutions, and visiting management departments and enterprises.

[0101] S42, data input: input the energy data obtained in the above steps into the constructed LEAP model framework in sequence. In the industrial field, input the annual output value, growth rate, unit output value energy consumption of each sub-industry, etc.; in the transportation field, input the number of vehicles, growth rate, passenger volume, mileage, etc.; in the power field, input the installed capacity, depreciation period, energy conversion efficiency, etc. of hydrogen energy components; in the heating field, input the heating area, population size, population growth rate, etc.

[0102] S43, scenario design: Based on the baseline scenario, design auxiliary scenarios, such as acceleration scenarios, carbon reduction scenarios, and other different scenarios. Consider the impact of possible alternative results on possible future events, and design different economic assumptions and development rate assumptions under different scenarios.

[0103] S44, data calculation: call the built-in data analysis model of LEAP to calculate the hydrogen load forecast results. Through the result page, you can view the growth of hydrogen load over time under different scenarios, the distribution of hydrogen load in various fields under different scenarios, etc. Through the energy balance interface, view the energy balance table and analyze the flow conversion of hydrogen energy, electricity and other energies.

[0104] Accordingly, by collating the data results, we can draw hydrogen load development charts, hydrogen load charts by field, etc., so as to summarize the hydrogen load development forecast under different scenarios, obtain more intuitive results, and facilitate development forecast analysis.

[0105] On the basis of the above steps, in order to further analyze the prediction results, this embodiment also includes:

[0106] S5. Based on multiple hydrogen-using sectors, the hydrogen load forecast results are divided and the classified forecast results are obtained; wherein the division method is based on the classification of four sectors (four fields), and the total hydrogen load includes the hydrogen load in the four fields of industry, electricity, transportation, and heating. The formula is:

[0107]

[0108] In the formula, is the total hydrogen load in year t; They are the hydrogen load in the four fields of industry, electricity, transportation and heating in the tth year.

[0109] Therefore, the forecast results of each field (department) are sorted out in turn and presented in the form of a model, specifically:

[0110] 5.1 Industrial field: The industrial field mainly considers the use of hydrogen in the chemical industry and the steel industry. The model is:

[0111]

[0112] In the formula, The hydrogen load for industrial use; are the hydrogen loads of the chemical industry and the steel industry in year t respectively; is the hydrogen load of other industries in the industrial sector in year t.

[0113] The specific calculation model for hydrogen load in the chemical industry is:

[0114]

[0115] In the formula, The hydrogen demand for producing green aviation kerosene in year t; is the hydrogen demand for producing ammonia in year t; is the hydrogen demand for producing methanol in year t; is the hydrogen demand for producing dimethyl ester in year t.

[0116] The specific calculation model for hydrogen load in the steel industry is:

[0117]

[0118] In the formula, a iron is the consumption coefficient, indicating that 1 ton of iron consumes a iron Ton of hydrogen, generally 0.09375; The demand for reduced iron.

[0119] The specific calculation model for hydrogen load in the industrial field is:

[0120]

[0121] In the formula, a indu is the proportionality coefficient, which takes different values ​​in different scenarios.

[0122] 5.2 Power sector: The power sector mainly considers hydrogen used in hydrogen hybrid engines and hydrogen fuel cells for power generation. The model is:

[0123]

[0124] In the formula, Indicates the hydrogen load used in the power sector; is the hydrogen demand for hydrogen hybrid engine power generation in year t; is the hydrogen demand for hydrogen fuel cell power generation in year t; is the hydrogen demand for other applications in the power sector in year t.

[0125] Among them, the models of other applications in the power field are:

[0126]

[0127] In the formula, a elec is the proportionality coefficient, which takes different values ​​in different scenarios.

[0128] 5.3 Transportation: The transportation sector mainly considers the hydrogen demand for land transportation and water transportation. The model is:

[0129]

[0130] In the formula, Indicates the hydrogen load used in the transportation sector; is the hydrogen demand for land transportation in year t, mainly including hydrogen buses, hydrogen public transport, and hydrogen heavy trucks; is the hydrogen demand for water transportation in year t, mainly including hydrogen cruise ships; is the hydrogen demand for other applications in the transportation sector in year t.

[0131] The specific calculation model for hydrogen load in land transportation is:

[0132]

[0133] In the formula, The annual average hydrogen consumption of each hydrogen bus, hydrogen public transport and hydrogen heavy truck respectively; They are the numbers of hydrogen buses, hydrogen public transport and hydrogen heavy trucks in year t respectively.

[0134] The specific calculation model for hydrogen load of water transportation is:

[0135]

[0136] In the formula, is the average annual hydrogen consumption of hydrogen cruise ships in year t; The number of hydrogen-powered cruise ships.

[0137] The specific calculation model for other hydrogen loads in the transportation field is:

[0138]

[0139] In the formula, a trans is the proportionality coefficient, which takes different values ​​in different scenarios.

[0140] 5.4 Heating sector: The heating sector mainly considers the hydrogen demand for heating in residential areas and pipeline delivery. The model is:

[0141]

[0142] In the formula, Indicates the hydrogen load in the heating sector; Hydrogen demand for residential district heating in year t; is the demand for hydrogen delivered through pipeline in year t; is the hydrogen demand for other applications in the heating sector in year t.

[0143] The specific calculation model for hydrogen load for residential area heating is:

[0144]

[0145] In the formula, are the currently known rural and urban populations in the study area in the kth year; β village , β city are the growth coefficients of rural and urban populations in the study area, respectively; are the per capita hydrogen demand for heating in the rural and urban populations in the study area, respectively.

[0146] The specific calculation model for other hydrogen loads in the heating field is:

[0147]

[0148] In the formula, a heat is the proportionality coefficient, which takes different values ​​in different scenarios.

[0149] Therefore, the required forecast results can be divided through different models to ensure the accuracy of the overall and single forecast results, and at the same time, better development forecasts can be made, including the macro overall and single fields and industries.

[0150] Furthermore, this embodiment can also use the modules of the LEAP model to display the prediction results by field, and visualize them according to the time dimension. Similarly, the hydrogen load development forecast of Qinghai Province is analyzed, and the hydrogen load in the industrial field, power field, transportation field and heating field of Qinghai Province is decomposed. The design of each module is as follows:

[0151] Industrial field module settings; industrial hydrogen loads mainly include hydrogen chemical industry, hydrogen metallurgy and others. Among them, hydrogen chemical industry is divided into five parts: green aviation kerosene, wind and solar hydrogen production and ammonia synthesis, Golmud hydrogenation to methanol, production of dimethyl carbonate and heating; hydrogen metallurgy includes hydrogen reduction of iron. Module settings are as follows Figure 4 shown.

[0152] Module settings in the power sector: Hydrogen loads in the power sector mainly include hydrogen for power generation and others. Hydrogen for power generation includes two parts: Haixi Prefecture hydrogen hybrid engine and Golmud hydrogen fuel cell power generation peak load regulation. Module settings are as follows: Figure 5 shown.

[0153] Module settings in the transportation field: Hydrogen load in the transportation field mainly includes hydrogen transportation and others. Hydrogen transportation includes four parts: buses, public transportation, heavy trucks and cruise ships. The module settings are as follows: Figure 6 shown.

[0154] Module settings in the heating field; module settings are as follows Figure 7 shown.

[0155] S6. Calculate energy benefits based on the classification prediction results. Energy benefits include economic benefits, new energy consumption benefits, and carbon reduction benefits. Analyze the economic and social benefits brought by the development of hydrogen load based on the prediction results. Study the overall benefits of consuming new energy electricity and reducing carbon emissions under the growth of hydrogen load. Make relevant development suggestions based on the development rate of hydrogen load under different scenarios.

[0156] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for predicting the development of hydrogen load based on the LEAP model, characterized in that the steps include: S1. Establishing a LEAP model; wherein the LEAP model is provided with multiple hydrogen-using departments; S2. Collecting energy data of the multiple hydrogen-using departments through a data acquisition system; S3, establishing an influencing factor prediction model, and obtaining an influencing factor prediction result based on the influencing factor prediction model; S4. Substitute the energy data and the prediction results of the influencing factors into the LEAP model, and solve to obtain the hydrogen load prediction results of the multiple hydrogen-using departments.

2. The method for predicting the development of hydrogen load based on the LEAP model according to claim 1, characterized in that: Also includes: S5. Classifying the hydrogen load prediction results based on the multiple hydrogen-using departments and obtaining classified prediction results; S6. Calculating energy efficiency based on the classification prediction results; The energy benefits include economic benefits, new energy consumption benefits and carbon reduction benefits.

3. The method for predicting the development of hydrogen load based on the LEAP model according to claim 1 or 2, characterized in that: In the LEAP model, the multiple hydrogen-using sectors include industrial sector, electric power sector, transportation sector and heating sector.

4. The method for predicting the development of hydrogen load based on the LEAP model according to claim 3, characterized in that: Each hydrogen-using sector is also divided into multiple sub-sectors, and each sub-sector has a corresponding energy type.

5. The method for predicting the development of hydrogen load based on the LEAP model according to claim 4, characterized in that: The energy types include at least hydrogen energy.

6. The method for predicting the development of hydrogen load based on the LEAP model according to claim 5, characterized in that: The multiple hydrogen-using sectors are each provided with corresponding activity levels and energy efficiency parameters; The activity level corresponding to the industrial sector is industrial output value, and the energy efficiency parameter is energy consumption per unit output value; The activity level corresponding to the power sector is power generation, and the energy efficiency parameter is energy consumption per unit of power generation; The activity levels corresponding to the transportation sector are vehicle ownership and vehicle mileage, and the energy efficiency parameter is energy consumption per 100 kilometers; The activity level corresponding to the heating department is the area of ​​the heating region and the permanent population of the region, and the energy efficiency parameters are the energy consumption per unit area of ​​heating and the per capita heating energy consumption.

7. The method for predicting the development of hydrogen load based on the LEAP model according to claim 1, characterized in that: The S2 also includes: performing data preprocessing on the energy data; The data preprocessing includes data cleaning and data noise reduction.

8. The method for predicting the development of hydrogen load based on the LEAP model according to claim 1, characterized in that: The influencing factor prediction model is a two-stage hybrid prediction model, and the two-stage hybrid prediction model establishment process also includes: S31. Establish a first-order model based on multi-head attention mechanism and one-dimensional convolutional neural network; S32. Establish a second-order linear regression model based on influencing factors.

9. The method for predicting the development of hydrogen load based on the LEAP model according to claim 8, characterized in that: The influencing factors include constraint conditions and cost conditions; The constraints include hydrogen energy utilization rate, hydrogen energy industry investment and carbon emissions; The forecast results of the influencing factors include expected hydrogen energy utilization rate, expected hydrogen energy industry investment, expected carbon emissions and expected cost conditions.

10. The method for predicting the development of hydrogen load based on the LEAP model according to claim 1, characterized in that: In S4, the LEAP model includes a baseline scenario and an auxiliary scenario; Substituting the energy data into the baseline scenario; The prediction result of the influencing factors is substituted into the auxiliary scenario.

Images