A method and device for calculating the water footprint of sustainable aviation fuel, and an electronic device

Abstract

The application provides a water footprint calculation method and device of sustainable aviation fuel and electronic equipment, and relates to the technical field of fine management of agricultural water resources, ecological hydrological simulation and agricultural artificial intelligence. The method comprises the following steps: calculating the theoretical water consumption of various raw material crops according to the historical meteorological data of a target region and the physiological parameters of the raw material crops for producing sustainable aviation fuel and growing in the target region; determining a target algorithm corresponding to the environmental feature complexity index of each crop-soil combination and the machine learning weight of the target algorithm from a plurality of machine learning algorithms; determining a first water footprint according to a water footprint measurement equation constructed based on a formation mechanism; applying the target algorithm to determine a second water footprint; and fusing the first water footprint of each crop-soil combination and the second water footprint of each crop-soil combination according to the machine learning weight to obtain the water footprint of the sustainable aviation fuel.

Description

Technical Field

[0001] This application relates to the interdisciplinary fields of agricultural water resource precision management, eco-hydrological simulation and agricultural artificial intelligence, and more specifically, to a method, device and electronic equipment for calculating the water footprint of sustainable aviation fuel. Background Technology

[0002] With the rapid development of the air transport industry, the demand for aviation fuel continues to grow. However, the use of traditional fossil fuels has led to substantial carbon emissions and environmental pollution. Therefore, developing low-carbon, environmentally friendly, and sustainable aviation fuel (SAF) has become one of the important directions for solving this problem. Water footprint is one of the important indicators for measuring the environmental impact of fuel production.

[0003] Accurate calculation of the water footprint (WF) is of vital guiding significance for assessing agricultural water consumption, developing regional drought and flood control plans, and implementing precision agricultural irrigation. The water footprint is generally divided into green water and blue water. Green water comes from the effective consumption of natural precipitation, while blue water comes from the irrigation consumption of surface water or groundwater.

[0004] Traditional mechanistic models are difficult to characterize the complex nonlinear hydrological interactions in reality, while single data-driven models have poor generalization ability and are prone to producing results that violate physical common sense. Summary of the Invention

[0005] In view of this, this application provides a method, apparatus and electronic device for calculating the water footprint of sustainable aviation fuel.

[0006] One aspect of this application provides a method for calculating the water footprint of sustainable aviation fuel, comprising: calculating the theoretical water consumption of various raw material crops for producing sustainable aviation fuel based on historical meteorological data of a target area and physiological parameters of various raw material crops grown in the target area; determining, from multiple machine learning algorithms, a target algorithm corresponding to the environmental feature complexity index of each of the multiple crop-soil combinations and the machine learning weights of the target algorithm; wherein, the multiple crop-soil combinations include various raw material crops and the growing soil of each raw material crop; the environmental feature complexity index is determined by the product of hydrodynamic physical parameters and the physiological parameters; determining a first water footprint of each crop-soil combination based on a water footprint calculation equation constructed based on a formation mechanism; determining a second water footprint of each crop-soil combination by applying the target algorithm based on the historical meteorological data, the physiological parameters, and the hydrodynamic physical parameters; and fusing the first water footprint and the second water footprint of each crop-soil combination according to the machine learning weights to obtain the water footprint of the sustainable aviation fuel.

[0007] According to embodiments of this application, when multiple machine learning algorithms include support vector regression models, random forest models, and deep neural network models, the target algorithm and its machine learning weights are determined based on the correspondence between the environmental feature complexity indices of each of the multiple crop-soil combinations and each of the aforementioned machine learning algorithms. The process of determining the correspondence between the environmental feature complexity indices of each of the multiple crop-soil combinations and the multiple machine learning algorithms includes: for any environmental feature complexity index of the crop-soil combination, if the environmental feature complexity index is less than a preset first threshold, the machine learning algorithm corresponding to the environmental feature complexity index is the support vector regression model, and the machine learning weight of the support vector regression model is a first coupling weight; if the environmental feature complexity index is greater than or equal to the preset first threshold and less than a preset second threshold, the machine learning algorithm corresponding to the environmental feature complexity index is the random forest model, and the machine learning weight of the random forest model is a second coupling weight; if the environmental feature complexity index is greater than or equal to the preset second threshold, the machine learning algorithm corresponding to the environmental feature complexity index is the deep neural network model, and the machine learning weight of the deep neural network model is a third coupling weight.

[0008] According to an embodiment of this application, determining the target algorithm and the machine learning weight of the target algorithm corresponding to the environmental feature complexity index of each of the crop-soil combinations from multiple machine learning algorithms includes: determining the target algorithm and the machine learning weight of the target algorithm corresponding to the environmental feature complexity index of each crop-soil combination based on a first comparison result between the environmental feature complexity index of each crop-soil combination and the preset first threshold and a second comparison result between the environmental feature complexity index of each crop-soil combination and the preset second threshold.

[0009] According to an embodiment of this application, the theoretical water consumption of various raw material crops for producing sustainable aviation fuel is calculated based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area. This includes: calculating the evapotranspiration of various raw material crops based on historical meteorological data of the target area; and, where the physiological parameters include the average crop coefficient and the number of days in the growing season, calculating the theoretical water consumption of various raw material crops based on the evapotranspiration, the average crop coefficient, and the number of days in the growing season.

[0010] According to an embodiment of this application, when the water footprint of each of the above-mentioned crop-soil combinations is a blue water footprint, the first water footprint of each of the above-mentioned crop-soil combinations is determined according to the water footprint calculation equation constructed based on the formation mechanism, including: determining the water surplus or deficit of each of the above-mentioned crop-soil combinations based on the effective precipitation and the theoretical water consumption of each of the above-mentioned crop-soil combinations; when the water surplus or deficit is greater than 0, the first water footprint is the water surplus or deficit; when the water surplus or deficit is less than or equal to 0, the first water footprint is 0.

[0011] According to an embodiment of this application, the water footprint of each crop-soil combination and the water footprint of each crop-soil combination are fused according to the aforementioned machine learning weights to obtain the water footprint of the sustainable aviation fuel. This includes: determining the weight of the first water footprint of each crop-soil combination according to the aforementioned machine learning weights; determining the first equivalent water footprint of each crop-soil combination according to the weight of the first water footprint of each crop-soil combination and the first water footprint of each crop-soil combination; determining the second equivalent water footprint of each crop-soil combination according to the aforementioned machine learning weights and the second water footprint of each crop-soil combination; obtaining the initial water footprint of each crop-soil combination according to the first equivalent water footprint of each crop-soil combination and the second equivalent water footprint of each crop-soil combination; and obtaining the water footprint of the sustainable aviation fuel according to the initial water footprint of each crop-soil combination.

[0012] According to embodiments of this application, all of the aforementioned machine learning algorithms are pre-trained machine learning algorithms. The training process of the aforementioned machine learning algorithms includes: acquiring a multimodal dataset; the multimodal dataset includes historical meteorological data of the sample area, physiological parameters of sample crops grown in the sample area for producing sustainable aviation fuel, hydrodynamic physical parameters of the soil in which the sample crops grow, and the corresponding water footprint of the sample crops; performing standardization preprocessing on the aforementioned multimodal dataset to obtain a preprocessed sample dataset; using the preprocessed sample dataset, employing K-fold cross-validation to train each of the untrained aforementioned machine learning algorithms to obtain multiple initial trained models; and determining the final hyperparameters of each of the aforementioned initial trained models using a grid search method to obtain multiple pre-trained machine learning algorithms.

[0013] According to an embodiment of this application, the above-mentioned water footprint calculation method further includes: visualizing the water footprint calculation results of the above-mentioned sustainable aviation fuel; the water footprint calculation results include the above-mentioned first water footprint, the above-mentioned second water footprint, and the water footprint of the above-mentioned raw material crop.

[0014] Another aspect of this application provides a water footprint calculation device for sustainable aviation fuel, comprising: a theoretical water consumption determination module, used to calculate the theoretical water consumption of various raw material crops for producing sustainable aviation fuel based on historical meteorological data of a target area and physiological parameters of various raw material crops grown in the target area; and a target algorithm determination module, used to determine, from multiple machine learning algorithms, a target algorithm corresponding to the environmental feature complexity index of each of the multiple crop-soil combinations and the machine learning weights of the target algorithms; wherein, the multiple crop-soil combinations include various raw material crops and the growing soil of each raw material crop; and the environmental feature complexity index is... The degree index is determined by the product of hydrodynamic physical parameters and the aforementioned physiological parameters; the first water footprint determination module is used to determine the first water footprint of each of the aforementioned crop-soil combinations according to the water footprint calculation equation constructed based on the formation mechanism; the second water footprint determination module is used to determine the second water footprint of each of the aforementioned crop-soil combinations according to the aforementioned historical meteorological data, the aforementioned physiological parameters, and the aforementioned hydrodynamic physical parameters, applying the aforementioned target algorithm; the water footprint result determination module is used to fuse the first water footprint and the second water footprint of each of the aforementioned crop-soil combinations according to the aforementioned machine learning weights to obtain the water footprint of the aforementioned sustainable aviation fuel.

[0015] Another aspect of this application provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the methods described above.

[0016] According to the embodiments of this application, an environmental feature complexity index is innovatively constructed by cross-calculating the total effective water retention capacity of the soil and the root depth of the crop. Based on this complexity index, the optimal machine learning algorithm and its corresponding coupling weight are dynamically matched from multiple included machine learning algorithms. Finally, the predicted water footprint output by the optimal machine learning algorithm is weighted and coupled with the benchmark water footprint based on hydrophysical equilibrium deduction to obtain the final water footprint calculation result. This effectively overcomes the problems of poor adaptability of traditional mechanism models under extreme climates and weak generalization ability of single data-driven models in heterogeneous crop-soil scenarios. It achieves high-fidelity and accurate cross-scenario calculation of agricultural water footprint and provides a visual traceability map of the underlying algorithm. Attached Figure Description

[0017] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0018] Figure 1An exemplary system architecture for calculating the water footprint of sustainable aviation fuels, based on the method and apparatus of this application, is illustrated schematically.

[0019] Figure 2 A flowchart illustrating a method for calculating the water footprint of sustainable aviation fuel according to an embodiment of this application is shown schematically.

[0020] Figure 3 A flowchart illustrating a method for calculating the water footprint of sustainable aviation fuels corresponding to multiple crop-soil combinations according to embodiments of this application is shown.

[0021] Figure 4 A flowchart illustrating a method for calculating the initial water footprint corresponding to any crop-soil combination according to an embodiment of this application is shown.

[0022] Figure 5 A flowchart illustrating a method for calculating the equivalent water footprint corresponding to any crop-soil combination according to an embodiment of this application is shown.

[0023] Figure 6 A block diagram of a water footprint calculation device for sustainable aviation fuel according to an embodiment of this application is shown schematically.

[0024] Figure 7 A block diagram of an electronic device suitable for implementing a water footprint calculation method for sustainable aviation fuel, according to an embodiment of this application, is shown schematically. Detailed Implementation

[0025] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0026] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0027] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0028] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0029] In the embodiments of this application, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.

[0030] In the embodiments of this application, the user's authorization or consent was obtained before obtaining or collecting the user's personal information.

[0031] The inventive concept of this application mainly includes: In related technologies, the calculation of water footprint mainly involves traditional mechanism model technology path and single data-driven model technology path, but each technology path faces technical bottlenecks.

[0032] The limitations of traditional mechanistic models are mainly reflected in:

[0033] Purely physical empirical models, such as the FAO-56 Penman-Monteith formula and the CROPWAT model, rely excessively on static crop coefficients (Kc) and idealized assumptions about soil moisture transformation. In real-world, large-scale agricultural scenarios, there is a highly nonlinear hydrological interaction between crop root depth and effective soil water retention capacity. For example, there is the capillary evaporation effect of deep-rooted crops in clay loam. Faced with extreme climates or complex heterogeneous combinations of crop and soil, static mechanistic models struggle to accurately characterize the true nonlinear losses under water stress, leading to measurement biases.

[0034] The poor generalization ability of a single data-driven model is mainly reflected in:

[0035] Predicting water consumption using a single machine learning model, such as using only a BP neural network or only a random forest, is problematic. However, agricultural data exhibits strong spatial heterogeneity, specifically including:

[0036] (1) In the scenario of shallow root system in sandy loam soil with low water retention, the data noise is large, and the model needs to have the ability to find noise-resistant boundaries.

[0037] (2) In complex collaborative scenarios of high water retention and deep root systems, due to the sharp increase in the coupling dimension between variables, deep networks are needed to fit high-dimensional nonlinear features.

[0038] (3) If a single machine learning algorithm is used in a "one-size-fits-all" manner, it often performs well on some test sets, but fails in generalization applications across regions and crops. In addition, pure data models are very likely to output results that violate physical common sense in cases with extremely few samples, such as predicting negative blue water consumption.

[0039] In summary, there is an urgent need to propose a prediction method that can dynamically perceive the complexity of the regional environment, automatically allocate the optimal computing power algorithm, and can complement and deeply couple the formula based on water footprint computing mechanism with various machine learning-based algorithms.

[0040] Figure 1 An exemplary system architecture for calculating the water footprint of sustainable aviation fuels, based on the method and apparatus of this application, is schematically illustrated. It should be noted that... Figure 1 The examples shown are merely examples of system architectures that can be applied to the embodiments of this application, in order to help those skilled in the art understand the technical content of this application, but do not mean that the embodiments of this application cannot be used in other devices, systems, environments or scenarios.

[0041] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0042] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays, including but not limited to smartphones, tablets, laptops, and desktop computers, etc.

[0043] Server 105 can be a server that provides various services, such as a backend management server that supports the first terminal device 101, the second terminal device 102, and the third terminal device 103 (for example only). The backend management server can analyze and process data such as received user requests and feed the processing results back to the terminal devices.

[0044] It should be noted that the water footprint calculation method for sustainable aviation fuel provided in this application embodiment can generally be executed by server 105. Correspondingly, the water footprint calculation device for sustainable aviation fuel provided in this application embodiment can generally be located in server 105. The water footprint calculation method for sustainable aviation fuel provided in this application embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the water footprint calculation device for sustainable aviation fuel provided in this application embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Alternatively, the water footprint calculation method for sustainable aviation fuel provided in this application embodiment can also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103, or by other terminal devices different from the first terminal device 101, the second terminal device 102, or the third terminal device 103. Accordingly, the water footprint calculation device for sustainable aviation fuel provided in this application embodiment can also be set in the first terminal device 101, the second terminal device 102 or the third terminal device 103, or in other terminal devices different from the first terminal device 101, the second terminal device 102 or the third terminal device 103.

[0045] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0046] Figure 2 A flowchart illustrating a method for calculating the water footprint of sustainable aviation fuel according to an embodiment of this application is shown schematically.

[0047] like Figure 2 As shown, the method includes operations S201 to S205.

[0048] In operation S201, the theoretical water consumption of various raw material crops is calculated based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area for the production of sustainable aviation fuel.

[0049] In embodiments of this application, historical meteorological data can be multi-scale meteorological baseline data. In response to a water footprint assessment directive initiated for the sustainable aviation fuel to be produced, multi-scale meteorological baseline data is acquired to calculate reference crop evapotranspiration and precipitation baselines; multidimensional physiological parameters of the target crop and hydrodynamic physical parameters of the target soil are acquired; thereby constructing multi-source environmental and basic characteristic data.

[0050] The target crops are raw material crops used to produce sustainable aviation fuel, and the target soil is the soil of the target area. There can be multiple target crops and multiple target soils.

[0051] The physiological parameters of the target crop should include at least the following: average crop coefficient (Kc) over the entire growth period, number of days in the growth period, maximum root depth, and target expected yield; the hydrodynamic physical parameters of the target soil should include at least the total effective water retention capacity (TAM); and the multi-scale meteorological baseline data should include at least the maximum temperature, minimum temperature, daily average temperature, and effective precipitation.

[0052] In operation S202, the target algorithm and its machine learning weights are determined from multiple machine learning algorithms, corresponding to the environmental feature complexity index of each crop-soil combination.

[0053] In the embodiments of this application, multiple crop-soil combinations include multiple raw material crops and the growing soil of each raw material crop; the environmental characteristic complexity index is determined by hydrodynamic physical parameters and physiological parameters. The environmental characteristic complexity index of any crop-soil combination is determined by the product of the hydrodynamic parameters of the growing soil of the target crop and the physiological parameters of the target crop.

[0054] Since there can be multiple target crops and multiple target soils, there are multiple combinations of target crops and target soils. The root depth parameters of the target crops and the total effective water retention parameters of the target soils are extracted. Based on the multiple combinations of target crops and target soils, a cross-mapping matrix is ​​established, and the environmental feature complexity index of each crop-soil combination is calculated.

[0055] The mathematical expression for the environmental feature complexity index can be the product of the root depth parameter of the target crop and the total effective water retention parameter of the target soil, which is used to characterize the nonlinear intensity of groundwater hydrological interaction.

[0056] A pool of machine learning algorithms with different underlying logics is pre-built. Based on the complexity index of environmental features, the target algorithm is adaptively triggered in the algorithm pool, and dynamic coupling weights for mechanism fusion are allocated as machine learning weights based on the complexity range.

[0057] In operation S203, the first water footprint of each crop-soil combination is determined based on the water footprint calculation equation constructed according to the formation mechanism.

[0058] In operation S204, based on historical meteorological data, physiological parameters, and hydrodynamic physical parameters, a target algorithm is applied to determine the second water footprint of each crop-soil combination.

[0059] In the embodiments of this application, historical meteorological data, physiological parameters, and hydrodynamic physical parameters are input into the target algorithm, and the target algorithm predicts the second water footprint based on the historical meteorological data, physiological parameters, and hydrodynamic physical parameters.

[0060] In operation S205, the first water footprint and the second water footprint of each crop-soil combination are fused according to machine learning weights to obtain the water footprint of sustainable aviation fuel.

[0061] In the embodiments of this application, taking the calculation of blue water footprint as an example, the baseline blue water footprint of each crop-soil combination is calculated based on the computer theoretical equation of water footprint, that is, the first water footprint; according to the target algorithm, the predicted value of the blue water footprint of each crop-soil combination is predicted, that is, the second water footprint.

[0062] Taking Blue Water Footprint as an example again, since the machine learning weights corresponding to each machine learning algorithm are different, the machine learning weights will also change dynamically when different machine learning algorithms are selected as the target algorithm. Therefore, machine learning weights are also called dynamic coupling weights.

[0063] The predicted blue water footprint for each crop-soil combination is fused with the baseline blue water footprint to obtain the fused water footprint for that crop-soil combination. Then, based on the fused water footprint for each crop-soil combination, the water footprint of sustainable aviation fuel feedstock, encompassing multiple crop-soil combinations, is obtained. The calculation process for the green water footprint is similar to that for the blue water footprint and will not be elaborated here.

[0064] According to the embodiments of this application, the technical bottlenecks of the prior art, such as the poor adaptability of a single static physical model in extreme environments and the weak generalization ability of a pure data-driven model in heterogeneous agricultural spatial scenarios, which easily leads to counterintuitive calculation results, are solved.

[0065] According to an embodiment of this application, the theoretical water consumption of various raw material crops is calculated based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area for the production of sustainable aviation fuel. This includes: calculating the evapotranspiration of various raw material crops based on historical meteorological data of the target area; and, when the physiological parameters include the average crop coefficient and the number of days in the growing season, calculating the theoretical water consumption of various raw material crops based on the evapotranspiration, the average crop coefficient, and the number of days in the growing season.

[0066] In the embodiments of this application, historical meteorological data of the target area are first input, and the reference crop evapotranspiration is calculated using the simplified Hargreaves-Samani core equation or the FAO-56 PM core equation. In this embodiment, the following empirical physical formulas are used for derivation:

[0067] ;

[0068] In the formula, This refers to solar radiation from the upper atmosphere. , , These are the daily average temperature, maximum temperature, and minimum temperature, respectively.

[0069] Based on this empirical physical formula, the evapotranspiration of each raw material crop can be obtained. .

[0070] Based on this, combined with the average crop coefficient of the raw material crop Based on the number of days in the growing season, calculate the theoretical total water consumption for each type of raw material crop. , .

[0071] In embodiments of this application, a special crop library and a soil physical property library can be pre-established. The special crop library includes various raw material crops for the production of sustainable aviation fuel. The soil physical property library includes various different soil types.

[0072] When calculating the water footprint of sustainable aviation fuel, the average crop coefficient, growth period, and root depth of raw material crops can be selected from a special crop library to calculate the theoretical total water consumption for each type of raw material crop.

[0073] The total effective water retention capacity, i.e., hydrodynamic physical parameters, of different soil physical properties (soil types) can be selected from the soil physical property database. By selecting corresponding parameters from the special crop database and the soil physical property database, different crop-soil combinations are obtained. Based on the product of the hydrodynamic physical parameters and physiological parameters of the crop-soil combination, the environmental characteristic complexity index of the crop-soil combination is calculated.

[0074] For example, a special crop library and a soil physical property library can be as follows:

[0075] Establish a specialty crop database, for example, including reed, castor bean, and flax, and set an average crop coefficient for each specialty crop. ), reproductive period ( ) and root depth ( Establish a soil physical property database, for example, including clay loam and sandy loam, and set the total effective water retention capacity for each soil type. ).

[0076] Taking reed, castor bean, and flax as examples, the information in the constructed special crop database is as follows:

[0077] C1 (Reed): The growing period is 215 days, the maximum root depth is 2.5 m, and the target yield is 30.0 ton / ha.

[0078] C2 (castor bean): The growing period is 140 days, the maximum root depth is 1.2 m, and the target yield is 2.0 ton / ha.

[0079] C3 (Flammulina): The growing period is 90 days, the maximum root depth is 0.9 m, and the target yield is 2.0 ton / ha.

[0080] Taking clay loam and sandy loam as examples, a soil physical property database is constructed:

[0081] S1 (clay loam): Total effective water retention capacity .

[0082] S2 (sandy loam): Total effective water retention capacity .

[0083] Various crop-soil combinations can be obtained from the special crops in the special crop database and the soil types in the soil physical property database, including: reed + clay loam, reed + sandy loam, castor bean + clay loam, castor bean + sandy loam, flax + clay loam, and flax + sandy loam.

[0084] In addition to the examples above, the special crop library may also include other special crops, and the soil physical property library may also include other soil physical properties, depending on actual needs.

[0085] According to embodiments of this application, when multiple machine learning algorithms include support vector regression models, random forest models, and deep neural network models, a target algorithm and its machine learning weights are determined based on the correspondence between the environmental feature complexity indices of multiple crop-soil combinations and each machine learning algorithm. The process of determining the correspondence between the environmental feature complexity indices of multiple crop-soil combinations and the multiple machine learning algorithms includes: for any crop-soil combination's environmental feature complexity index, if the environmental feature complexity index is less than a preset first threshold, the machine learning algorithm corresponding to the environmental feature complexity index is a support vector regression model, and the machine learning weight of the support vector regression model is a first coupling weight; if the environmental feature complexity index is greater than or equal to the preset first threshold and less than a preset second threshold, the machine learning algorithm corresponding to the environmental feature complexity index is a random forest model, and the machine learning weight of the random forest model is a second coupling weight; if the environmental feature complexity index is greater than or equal to the preset second threshold, the machine learning algorithm corresponding to the environmental feature complexity index is a deep neural network model, and the machine learning weight of the deep neural network model is a third coupling weight.

[0086] In the embodiments of this application, the preset machine learning algorithm pool includes at least a support vector regression (SVR) model for handling small samples and noise-resistant boundaries, a random forest (RF) model for handling medium dimensionality and feature importance evaluation, and a deep neural network (DNN) model for handling high-dimensional nonlinear mappings.

[0087] The preset first threshold, preset second threshold, first coupling weight, second coupling weight, and third coupling weight can be set based on experience or actual conditions, or determined through multiple experiments. Among them, the first coupling weight, second coupling weight, and third coupling weight are three dynamic coupling weights.

[0088] According to an embodiment of this application, determining the target algorithm and the machine learning weight of the target algorithm corresponding to the environmental feature complexity index of each crop-soil combination from multiple machine learning algorithms includes: determining the target algorithm and the machine learning weight of the target algorithm corresponding to the environmental feature complexity index of each crop-soil combination based on a first comparison result between the environmental feature complexity index of each crop-soil combination and a preset first threshold and a second comparison result between the environmental feature complexity index of each crop-soil combination and a preset second threshold.

[0089] In the embodiments of this application, when the environmental feature complexity index is less than a preset first threshold, it is determined to be a low nonlinear low-dimensional scenario, triggering the SVR model and assigning a first coupling weight; when the environmental feature complexity index is between the preset first threshold and the second threshold, it is determined to be a medium complexity scenario, triggering the RF model and assigning a second coupling weight; when the environmental feature complexity index is greater than or equal to the preset second threshold, it is determined to be a high nonlinear deep hydrological interaction scenario, triggering the DNN model and assigning a third coupling weight; wherein, the first coupling weight > the second coupling weight > the third coupling weight.

[0090] The embodiments of this application define an environmental feature complexity index (Feature Complexity). ): .

[0091] in, For the total effective water retention capacity of the soil, This represents the maximum root depth of the crop. Total effective water retention capacity of the soil is a hydrodynamic physical parameter.

[0092] Furthermore, taking a preset first threshold of 100 and a preset second threshold of 200 as an example, the adaptive algorithm is explained using the combination of raw material crops and soil physical properties corresponding to scenarios A, B, and C. Scenario A is flax C3 + sandy loam S2, scenario B is castor C2 + clay loam S1, and scenario C is reed C1 + clay loam S1.

[0093] Scene A (Flax C3 + Sandy Loam S2): .because The environmental hydrological interaction is weak, and the system determines that there is a large amount of noise in the surface evaporation data. An SVR model is triggered (using the radial basis function (RBF) kernel to construct a high-dimensional feature space noise-resistant boundary) and assigned a high machine learning trust weight. .

[0094] Scene B (castor bean C2 + clay loam S1): The value is between 100 and 200. The algorithm triggers a Random Forest (RF) model, leveraging the advantages of ensemble learning to handle medium-dimensional feature interactions. .

[0095] Scene C (Reed C1 + Clay Loam S1): . In deep-rooted, high-water-retention systems, capillary rise and deep seepage are extremely complex. The algorithm triggers a DNN model (in this example, a 4-layer fully connected deep neural network with ReLU activation is used to extract high-dimensional implicit features). Due to the extreme complexity of this scenario, to prevent overfitting of the data model, the machine learning trust weight is lowered to the basic level. It relies more on the baseline mechanism for protection.

[0096] In scenario A, the environmental feature complexity index is less than 100; therefore, Support Vector Regression (SVR) is chosen, and machine learning weights are set accordingly. .

[0097] In scenario B, the environmental feature complexity index is greater than 100 and less than 200. Therefore, Random Forest (RF) is chosen, and machine learning weights are set accordingly. .

[0098] In scenario C, the environmental feature complexity index is greater than 200; therefore, a deep neural network (DNN) is chosen, and machine learning weights are set accordingly. .

[0099] This application innovatively proposes a method to quantify and construct an environmental feature complexity index by cross-mapping the total effective water retention capacity (TAM) of the target soil with the maximum root depth of the target crop. This index is used to characterize the nonlinear intensity of groundwater interactions. Furthermore, an algorithm pool including Support Vector Regression (SVR), Random Forest (RF), and Deep Neural Network (DNN) is pre-constructed. Instead of using a single algorithm, a machine learning algorithm that dynamically senses the regional environmental state and triggers optimal matching based on the calculated environmental feature complexity index is used, while simultaneously assigning dynamic computational weights.

[0100] According to an embodiment of this application, when the water footprint of each crop-soil combination is a blue water footprint, the first water footprint of each crop-soil combination is determined according to the water footprint calculation equation constructed based on the formation mechanism, including: determining the water surplus / deficit of each crop-soil combination based on the effective precipitation and theoretical water consumption of each crop-soil combination; when the water surplus / deficit is greater than 0, the first water footprint is the water surplus / deficit; when the water surplus / deficit is less than or equal to 0, the first water footprint is 0.

[0101] In the embodiments of this application, taking the calculation of the final blue water footprint (i.e., artificial irrigation consumption) as an example, the physical water balance equation is combined with effective precipitation. Find the baseline blue water : Among them, the baseline blue water As the first water footprint, This represents the water surplus or deficit.

[0102] According to an embodiment of this application, the water footprint of each crop-soil combination is fused with the first water footprint and the second water footprint of each crop-soil combination based on machine learning weights to obtain the water footprint of sustainable aviation fuel. This includes: determining the weights of the first water footprints of each crop-soil combination based on machine learning weights; determining the first equivalent water footprint of each crop-soil combination based on the weights and the first water footprints of each crop-soil combination; determining the second equivalent water footprint of each crop-soil combination based on the machine learning weights and the second water footprints of each crop-soil combination; obtaining the initial water footprint of each crop-soil combination based on the first equivalent water footprint and the second equivalent water footprint of each crop-soil combination; and obtaining the water footprint of sustainable aviation fuel based on the initial water footprint of each crop-soil combination.

[0103] In the embodiments of this application, taking blue water footprint as an example, the machine learning weights are: The weight of the first water footprint is The first equivalent water footprint is The second equivalent water footprint is The general mathematical expression for the water footprint of any crop-soil combination obtained through fusion calculation is:

[0104] .

[0105] result After normalizing the crop yield, the final blue water footprint value per unit of agricultural product can be output.

[0106] By normalizing the water footprint of multiple crop-soil combinations, the water footprint of each crop-soil combination can be directly compared horizontally. When the unit water footprint is used as the water footprint of sustainable aviation fuel, the result of normalizing the water footprint of multiple crop-soil combinations can be used as the water footprint of sustainable aviation fuel.

[0107] According to embodiments of this application, the traditional farmland water balance physical baseline (Base Blue / GreenWF) is used as a physical lower limit guarantee and dynamically weighted and coupled with the triggered algorithm prediction value. This ensures that the prediction results do not deviate from physical common sense (eliminating negative or abnormally high values), while also significantly improving the local measurement accuracy under extreme and complex conditions using AI technology.

[0108] According to embodiments of this application, all machine learning algorithms are pre-trained machine learning algorithms. The training process of the multiple machine learning algorithms includes: acquiring a multimodal dataset; the multimodal dataset includes historical meteorological data of the sample area, physiological parameters of sample crops grown in the sample area for producing sustainable aviation fuel, hydrodynamic physical parameters of the soil in which the sample crops grow, and the water footprint of the corresponding sample crops; performing standardization preprocessing on the multimodal dataset to obtain a preprocessed sample dataset; using K-fold cross-validation based on the preprocessed sample dataset to train each untrained machine learning algorithm to obtain multiple initial trained models; and determining the final hyperparameters of each initial trained model using a grid search method to obtain multiple pre-trained machine learning algorithms.

[0109] In the embodiments of this application, the standardization preprocessing adopts Z-score standardization preprocessing to eliminate the influence of dimensions; K-fold cross validation is used to train SVR, RF and DNN in the algorithm pool respectively, and the optimal hyperparameters of each model are determined by grid search.

[0110] According to an embodiment of this application, the method for calculating the water footprint of sustainable aviation fuel further includes: visualizing the calculation results of the water footprint of sustainable aviation fuel.

[0111] For example, the water footprint of sustainable aviation fuel can be generated into a stacked bar chart, with the horizontal axis representing different combinations of raw material crops and soil physical properties; the vertical axis represents the unit water footprint measurement value for each combination, in units of... Different colors can be used to represent different identifiers. For example, green can be used to represent green water and blue to represent blue water. Different algorithms can be pre-defined to use different colors when calculating the second water footprint. For example, DNN can be represented by dark red and SVR by orange. Above each combined column, the algorithm code label used to calculate the second water footprint can be printed to achieve a direct mapping between the prediction results and the underlying computing power of the algorithm code.

[0112] The following is for reference. Figures 3-5 In conjunction with specific embodiments, Figure 2 The method shown will be further explained.

[0113] Figure 3 A flowchart illustrating a method for calculating the water footprint of sustainable aviation fuels corresponding to multiple crop-soil combinations according to embodiments of this application is shown.

[0114] like Figure 3As shown, taking sustainable aviation fuel raw materials including a first raw material crop and a second raw material crop, and the target area soil types including a first soil type and a second soil type as an example, the first raw material crop and the second raw material crop, as well as the first soil type and the second soil type, can be combined into four crop-soil combinations. These four crop-soil combinations are respectively the first crop-soil combination composed of the first raw material crop and the first soil type, the second crop-soil combination composed of the first raw material crop and the second soil type, the third crop-soil combination composed of the second raw material crop and the first soil type, and the fourth crop-soil combination composed of the second raw material crop and the second soil type.

[0115] Historical meteorological data, physiological parameters, and hydrodynamic physical parameters of each crop-soil combination are obtained. Based on these parameters, the initial water footprint of each crop-soil combination is determined. Finally, based on the initial water footprints of the four crop-soil combinations, the water footprint of sustainable aviation fuel is obtained.

[0116] Figure 4 A flowchart illustrating a method for calculating the initial water footprint corresponding to any crop-soil combination according to an embodiment of this application is shown.

[0117] like Figure 4 As shown, with Figure 3 Taking the first crop-soil combination as an example, the process of determining the initial water footprint of the crop-soil combination is explained.

[0118] Based on the historical meteorological data, physiological parameters, and hydrodynamic physical parameters of the first crop-soil combination, the environmental feature complexity index FC of the first crop-soil combination is determined. An adaptive algorithm is then applied, that is, based on the comparison results of the environmental feature complexity index FC of the crop-soil combination with the two thresholds of 100 and 200, the corresponding target algorithm and the weight of the target algorithm are determined.

[0119] After obtaining the target algorithm and its weights, the machine learning weights of the target algorithm and its corresponding FC are applied to obtain the first equivalent water footprint and the second equivalent water footprint. The first equivalent water footprint and the second equivalent water footprint are then added together to determine the initial water footprint of the first crop-soil combination.

[0120] Figure 5 A flowchart illustrating a method for calculating the equivalent water footprint corresponding to any crop-soil combination according to an embodiment of this application is shown.

[0121] like Figure 5 As shown, with Figure 4 The calculation method of the equivalent water footprint of the first crop-soil combination is used as an example to illustrate the calculation method, which includes operations S501-S505.

[0122] In operation S501, the first water footprint of the first crop-soil combination is determined based on the water footprint calculation equation constructed based on the formation mechanism.

[0123] In operation S502, based on historical data, physiological parameters, and hydrodynamic physical parameters, a target algorithm is applied to determine the second water footprint of the first crop-soil combination.

[0124] In operation S503, the weight of the first water footprint of the first crop-soil combination is determined based on the machine learning weights corresponding to the target algorithm.

[0125] In operation S504, the first equivalent water footprint of the first crop-soil combination is determined based on the weight of the first water footprint of the first crop-soil combination and the product of the first water footprint of the first crop-soil combination.

[0126] In operation S505, the second equivalent water footprint of the first crop-soil combination is determined by multiplying the machine learning weights corresponding to the target algorithm with the second water footprint of the first crop-soil combination.

[0127] According to embodiments of this application, the algorithm mechanism to be used to calculate the second water footprint is determined by the feature complexity index. In scenarios with low water retention and shallow root systems (high noise), SVR is used to find noise-resistant boundaries, while in scenarios with deep root systems and high water retention (high-dimensional nonlinear coupling), DNN is used to extract deep features, significantly improving the model's generalization ability in heterogeneous agricultural scenarios. Furthermore, adaptive color anchoring and code-level annotation can be performed on the water footprint stacked feature chart, thus providing intuitive traceability.

[0128] Figure 6 A block diagram of a water footprint calculation device for sustainable aviation fuel according to an embodiment of this application is shown schematically.

[0129] like Figure 6 As shown, the water footprint calculation device 600 for sustainable aviation fuel includes a theoretical water consumption determination module 610, a target algorithm determination module 620, a first water footprint determination module 630, a second water footprint determination module 640, and a water footprint result determination module 650.

[0130] The theoretical water consumption determination module 610 is used to calculate the theoretical water consumption of various raw material crops based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area for the production of sustainable aviation fuel.

[0131] The target algorithm determination module 620 is used to determine the target algorithm and the machine learning weight of the target algorithm corresponding to the environmental feature complexity index of each of the multiple crop-soil combinations from multiple machine learning algorithms; wherein, the multiple crop-soil combinations include multiple raw material crops and the growing soil of each raw material crop; the environmental feature complexity index is determined by the product of hydrodynamic physical parameters and physiological parameters.

[0132] The first water footprint determination module 630 is used to determine the first water footprint of each crop-soil combination based on the water footprint calculation equation constructed based on the formation mechanism.

[0133] The second water footprint determination module 640 is used to determine the second water footprint of each crop-soil combination by applying a target algorithm based on historical meteorological data, physiological parameters, and hydrodynamic physical parameters.

[0134] The water footprint determination module 650 is used to fuse the first water footprint and the second water footprint of each crop-soil combination according to machine learning weights to obtain the water footprint of sustainable aviation fuel.

[0135] Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be at least partially implemented as hardware circuits, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or implemented by hardware or firmware in any other reasonable manner by integrating or packaging circuits, or implemented in any one of software, hardware, and firmware, or in a suitable combination of any of these. Alternatively, one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.

[0136] For example, any multiple of the following modules can be combined into one module / unit / subunit: the theoretical water consumption determination module 610, the target algorithm determination module 620, the first water footprint determination module 630, the second water footprint determination module 640, and the water footprint result determination module 650. Alternatively, any one of these modules / units / subunits can be split into multiple modules / units / subunits. Or, at least some of the functionality of one or more of these modules / units / subunits can be combined with at least some of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this application, at least one of the theoretical water consumption determination module 610, target algorithm determination module 620, first water footprint determination module 630, second water footprint determination module 640, and water footprint result determination module 650 can be at least partially implemented as a hardware circuit, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable method of integrating or packaging the circuit, or implemented in software, hardware, or firmware, or in any appropriate combination of any of these three methods. Alternatively, at least one of the theoretical water consumption determination module 610, target algorithm determination module 620, first water footprint determination module 630, second water footprint determination module 640, and water footprint result determination module 650 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0137] It should be noted that the water footprint calculation device part of sustainable aviation fuel in the embodiments of this application corresponds to the water footprint calculation method part of sustainable aviation fuel in the embodiments of this application. The description of the water footprint calculation device part of sustainable aviation fuel is specifically referred to in the water footprint calculation method part of sustainable aviation fuel, and will not be repeated here.

[0138] Figure 7 A block diagram of an electronic device suitable for implementing a water footprint calculation method for sustainable aviation fuel, according to an embodiment of this application, is shown schematically. Figure 7 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0139] like Figure 7As shown, an electronic device 700 according to an embodiment of this application includes a processor 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage portion 708 into a random access memory (RAM) 703. The processor 701 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 701 may also include onboard memory for caching purposes. The processor 701 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0140] RAM 703 stores various programs and data required for the operation of electronic device 700. Processor 701, ROM 702, and RAM 703 are interconnected via bus 704. Processor 701 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 702 and / or RAM 703. It should be noted that programs may also be stored in one or more memories other than ROM 702 and RAM 703. Processor 701 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in one or more memories.

[0141] According to embodiments of this application, the electronic device 700 may further include an input / output (I / O) interface 705, which is also connected to a bus 704. The electronic device 700 may also include one or more of the following components connected to the input / output (I / O) interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output (I / O) interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.

[0142] According to embodiments of this application, the method flow according to embodiments of this application can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by processor 701, it performs the functions defined in the system of embodiments of this application. According to embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0143] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0144] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0145] For example, according to embodiments of this application, a computer-readable storage medium may include the ROM 702 and / or RAM 703 described above and / or one or more memories other than ROM 702 and RAM 703.

[0146] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this application. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the water footprint calculation method for sustainable aviation fuel provided in the embodiments of this application.

[0147] When the computer program is executed by the processor 701, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0148] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 709, and / or installed from a removable medium 711. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0149] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0150] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations are not explicitly described in this application. In particular, without departing from the spirit and teachings of this application, the features described in the various embodiments of this application can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of this application.

[0151] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Without departing from the scope of this application, those skilled in the art can make various substitutions and modifications, all of which should fall within the scope of this application.

Claims

1. A method for calculating the water footprint of sustainable aviation fuel, characterized in that, The water footprint calculation method includes: Based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area for the production of sustainable aviation fuel, the theoretical water consumption of each of the raw material crops is calculated. From multiple machine learning algorithms, a target algorithm corresponding to the environmental feature complexity index of each of multiple crop-soil combinations and the machine learning weight of the target algorithm are determined; wherein, the multiple crop-soil combinations include multiple raw material crops and the growing soil of each raw material crop; the environmental feature complexity index is determined by the product of hydrodynamic physical parameters and physiological parameters; Based on the water footprint calculation equation constructed according to the formation mechanism, the first water footprint of each crop-soil combination is determined. Based on the historical meteorological data, the physiological parameters, and the hydrodynamic physical parameters, the target algorithm is applied to determine the second water footprint of each crop-soil combination. Based on the machine learning weights, the first water footprint and the second water footprint of each crop-soil combination are fused to obtain the water footprint of the sustainable aviation fuel.

2. The method for calculating the water footprint of sustainable aviation fuel according to claim 1, characterized in that, In the case of multiple machine learning algorithms including support vector regression model, random forest model and deep neural network model, the target algorithm and the machine learning weight of the target algorithm are determined according to the correspondence between the environmental feature complexity index of each of the multiple crop-soil combinations and each of the machine learning algorithms. The process of determining the correspondence between the environmental feature complexity indices of the multiple crop-soil combinations and the multiple machine learning algorithms includes: For any of the crop-soil combinations, the environmental feature complexity index is... When the environmental feature complexity index is less than a preset first threshold, the machine learning algorithm corresponding to the environmental feature complexity index is the support vector regression model, and the machine learning weight of the support vector regression model is the first coupling weight. When the environmental feature complexity index is greater than or equal to the preset first threshold and less than the preset second threshold, the machine learning algorithm corresponding to the environmental feature complexity index is the random forest model, and the machine learning weight of the random forest model is the second coupling weight. When the environmental feature complexity index is greater than or equal to a preset second threshold, the machine learning algorithm corresponding to the environmental feature complexity index is the deep neural network model, and the machine learning weight of the deep neural network model is the third coupling weight.

3. The method for calculating the water footprint of sustainable aviation fuel according to claim 2, characterized in that, Determining the target algorithm and its machine learning weights from multiple machine learning algorithms, corresponding to the environmental feature complexity index of each crop-soil combination, including: Based on the first comparison result between the environmental feature complexity index of each crop-soil combination and the preset first threshold, and the second comparison result between the environmental feature complexity index of each crop-soil combination and the preset second threshold, the target algorithm corresponding to the environmental feature complexity index of each crop-soil combination and the machine learning weight of the target algorithm are determined.

4. The method for calculating the water footprint of sustainable aviation fuel according to claim 1, characterized in that, Based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area for the production of sustainable aviation fuel, the theoretical water consumption of each of these raw material crops is calculated, including: Based on historical meteorological data of the target area, calculate the evapotranspiration of various raw material crops; When the physiological parameters include the average crop coefficient and the number of days in the growing season, the theoretical water consumption of each of the raw material crops is calculated based on the evapotranspiration, the average crop coefficient, and the number of days in the growing season.

5. The method for calculating the water footprint of sustainable aviation fuel according to claim 1, characterized in that, When the water footprint of each crop-soil combination is a blue water footprint, the first water footprint of each crop-soil combination is determined according to the water footprint calculation equation constructed based on the formation mechanism, including: The water surplus or deficit of each crop-soil combination is determined based on the effective precipitation and theoretical water consumption of each crop-soil combination. When the water deficit is greater than 0, the first water footprint is the water deficit. When the water deficit is less than or equal to 0, the first water footprint is 0.

6. The method for calculating the water footprint of sustainable aviation fuel according to claim 1, characterized in that, Based on the machine learning weights, the first water footprint and the second water footprint of each crop-soil combination are fused to obtain the water footprint of the sustainable aviation fuel, including: Based on the machine learning weights, determine the weight of the first water footprint for each crop-soil combination; Based on the weight of the first water footprint of each crop-soil combination and the first water footprint of each crop-soil combination, the first equivalent water footprint of each crop-soil combination is determined; The second equivalent water footprint of each crop-soil combination is determined based on the machine learning weights and the second water footprint of each crop-soil combination. The initial water footprint of each crop-soil combination is obtained based on the first equivalent water footprint and the second equivalent water footprint of each crop-soil combination. The water footprint of the sustainable aviation fuel is obtained based on the initial water footprint of each crop-soil combination.

7. The method for calculating the water footprint of sustainable aviation fuel according to claim 1, characterized in that, All of the machine learning algorithms mentioned are pre-trained machine learning algorithms; The training process of the aforementioned machine learning algorithms includes: Obtain a multimodal dataset; the multimodal dataset includes historical meteorological data of the sample area, physiological parameters of sample crops grown in the sample area for the production of sustainable aviation fuel, hydrodynamic physical parameters of the soil in which the sample crops grow, and the corresponding water footprint of the sample crops. The multimodal dataset is standardized and preprocessed to obtain a preprocessed sample dataset; Based on the preprocessed sample dataset, K-fold cross-validation is used to train each of the untrained machine learning algorithms to obtain multiple initial trained models. The final hyperparameters of each initially trained model are determined by grid search, resulting in multiple pre-trained machine learning algorithms.

8. The method for calculating the water footprint of sustainable aviation fuel according to claim 1, characterized in that, The water footprint calculation method also includes: The water footprint calculation results of the sustainable aviation fuel are visualized; the water footprint calculation results include the first water footprint, the second water footprint, and the water footprint of the raw material crop.

9. A water footprint calculation device for sustainable aviation fuel, characterized in that, The water footprint calculation device includes: The theoretical water consumption determination module is used to calculate the theoretical water consumption of various raw material crops based on historical meteorological data of the target area and physiological parameters of various raw material crops grown in the target area for the production of sustainable aviation fuel. A target algorithm determination module is used to determine, from multiple machine learning algorithms, the target algorithm corresponding to the environmental feature complexity index of each of multiple crop-soil combinations, and the machine learning weight of the target algorithm; wherein, the multiple crop-soil combinations include multiple raw material crops and the growing soil of each raw material crop; the environmental feature complexity index is determined by the product of hydrodynamic physical parameters and physiological parameters; The first water footprint determination module is used to determine the first water footprint of each crop-soil combination based on the water footprint calculation equation constructed based on the formation mechanism. The second water footprint determination module is used to determine the second water footprint of each crop-soil combination based on the historical meteorological data, the physiological parameters, and the hydrodynamic physical parameters, and by applying the target algorithm. The water footprint determination module is used to fuse the first water footprint of each crop-soil combination and the second water footprint of each crop-soil combination according to the machine learning weights to obtain the water footprint of the sustainable aviation fuel.

10. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 8.

Images