Sustainability evaluation method and system for sustainable aviation fuel

Abstract

The invention discloses a sustainability evaluation method and system for continuous aviation fuel, and relates to the technical field of high-efficiency energy-saving engineering evaluation service. The method comprises the following steps: acquiring carbon footprint data of a sustainable aviation fuel SAF in raw material acquisition, product production, transportation and use stages, and constructing a full-life-cycle carbon footprint dynamic model and a distributed carbon footprint database in combination with an LCA industry database; updating the carbon emission factor in real time, and calculating the carbon emission result of the SAF in combination with the process parameters of each manufacturing process; determining environmental income deduction of waste regeneration, landfill treatment and energy recovery by adopting a regeneration cycle carbon emission allocation algorithm; and generating a sustainability evaluation report including a fossil source, a biological source, land utilization change, resource consumption intensity and a waste treatment environmental effect according to the environmental income deduction and carbon emission result. According to the method and the device, the integrity and the accuracy of the sustainability evaluation of the SAF are improved.

Description

Technical Field

[0001] The present application relates to the technical field of high-efficiency energy-saving engineering assessment services, and more specifically, to a sustainability assessment method and system for sustainable aviation fuel. Background Art

[0002] Sustainable Aviation Fuel (SAF) refers to a non-petroleum-based alternative aviation fuel that meets both aviation safety and airworthiness standards and sustainability assessment criteria. SAF is compatible with existing aircraft and civil aviation infrastructure, and its lifecycle carbon emissions are reduced by over 10% compared to fossil-based jet fuel.

[0003] Traditional SAF evaluation methods only focus on carbon emissions during the SAF production or combustion stages, lacking a coordinated analysis of the carbon footprint of the entire chain; and only use a fixed ratio to estimate the environmental benefit deductions for recycling links such as waste oil regeneration, resulting in evaluation results that deviate from the actual environmental benefits.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of the present application provide a sustainability evaluation method and system for sustainable aviation fuel to solve the above-mentioned technical problems.

[0006] According to one aspect of an embodiment of the present application, a sustainability evaluation method for sustainable aviation fuel is provided, including: obtaining carbon footprint data of sustainable aviation fuel (SAF) in the raw material acquisition, product production, transportation and use stages, and constructing a full life cycle carbon footprint dynamic model and a distributed carbon footprint database in combination with the LCA industry database; updating the carbon emission factor in real time based on the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, and calculating the carbon emission results of SAF in combination with the process parameters of each manufacturing process; using a regeneration cycle carbon emission allocation algorithm, combined with the waste generation amount, regeneration feasibility coefficient and landfill treatment carbon emission factor of each manufacturing process, determining the environmental benefit deduction of waste regeneration, landfill treatment and energy recovery; and generating a sustainability evaluation report based on the environmental benefit deduction and the carbon emission results, including fossil sources, biomass, land use change, resource consumption intensity and waste treatment environmental effects.

[0007] According to another aspect of an embodiment of the present application, a sustainability evaluation system for sustainable aviation fuel is provided, including: a carbon footprint model and database construction module, which is used to obtain carbon footprint data of sustainable aviation fuel SAF in the raw material acquisition, product production, transportation and use stages, and to construct a full life cycle carbon footprint dynamic model and a distributed carbon footprint database in combination with the LCA industry database; a SAF carbon emission result calculation module, which is used to update the carbon emission factor in real time according to the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, and to calculate the carbon emission results of SAF in combination with the process parameters of each manufacturing process; an environmental benefit deduction determination module, which is used to determine the environmental benefit deduction of waste regeneration, landfill treatment and energy recovery by adopting a recycling cycle carbon emission allocation algorithm, in combination with the waste generation amount, recycling feasibility coefficient and landfill treatment carbon emission factor of each manufacturing process; a sustainability evaluation report generation module, which is used to generate a sustainability evaluation report including fossil sources, biomass, land use change, resource consumption intensity and waste treatment environmental effects based on the environmental benefit deduction and the carbon emission results.

[0008] The sustainability assessment method and system for sustainable aviation fuels provided in this application have the following beneficial effects: By integrating carbon footprint data for SAF throughout its entire life cycle, from raw material acquisition to product use, and constructing a dynamic model based on the LCA industry database, this method overcomes the problem of traditional methods focusing solely on carbon emissions during the production or combustion phase, enabling accurate location of carbon emission sources and cross-link impact assessment. A dynamic coupling mechanism, with real-time updates of carbon emission factors and process parameters, overcomes the carbon emission calculation bias caused by fixed parameters and ensures that carbon emission differences between different manufacturing processes are accurately reflected in the model. Through multi-dimensional quantitative analysis of waste generation, regeneration feasibility coefficients, and landfill factors, a recycling cycle allocation algorithm based on actual material flows is constructed, changing the traditional extensive model of fixed-ratio deductions and dynamically matching environmental benefit accounting with waste treatment scenarios. The generated sustainability assessment report integrates multi-dimensional indicators such as fossil sources, biomass sources, and land use change to establish a comprehensive evaluation system covering resource consumption intensity and waste treatment effects, significantly improving the completeness and accuracy of SAF sustainability assessments. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings described herein are used to provide a further understanding of the embodiments of the present invention and constitute a part of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation on this application. In the drawings: Figure 1 This is a flowchart of an optional sustainability assessment method for sustainable aviation fuel according to an embodiment of the present application; Figure 2 Schematic diagram of an optional dynamic prediction and correction of carbon emissions according to an embodiment of the present application; Figure 3 Schematic diagram of an optional multi-objective optimization method under Markov chain constraints according to an embodiment of the present application; Figure 4 An optional SAF product life cycle and life cycle assessment flow chart according to an embodiment of the present application; Figure 5 4 is a structural diagram of an optional sustainable aviation fuel sustainability assessment system according to an embodiment of the present application.

[0010] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0011] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0012] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] Currently, aviation fuel accounts for over 97% of total emissions from my country's civil aviation industry. Electric and hydrogen-powered aircraft cannot meet the aviation industry's emission reduction needs in the short term. SAF is currently the most efficient and energy-saving solution. SAF represents a new direction and solution that comprehensively utilizes my country's existing resources and technologies. The development of SAF will help promote the upgrading and innovation of chemical and energy technologies, addressing energy scarcity and energy security.

[0014] ICAO has established the Carbon Offsetting and Reduction Scheme for International Aviation. As one of the key emission reduction measures, alternative aviation fuels must not only meet the technical safety indicators specified in ASTM D7566, but also the sustainability standards of the "Sustainability Requirements for Aviation Alternative Fuels." The greenhouse gas emissions of alternative aviation fuels (i.e., the product carbon footprint) must be calculated from a full life cycle perspective, and compared with the greenhouse gas emissions of traditional jet fuel to obtain accurate and standardized emission reduction values. ASTM D7566 is the "Standard Specification for Aviation Turbine Fuels Containing Synthetic Hydrocarbons," developed by the American Society for Testing and Materials (ASTM International).

[0015] The applicant plans to establish an aviation fuel life cycle carbon footprint assessment and management method at the Civil Aviation Big Data and Information Service Technology Research Center, which will be used to carry out full life cycle carbon emission accounting of aviation fuel and sustainability assessment and research of aviation alternative fuels. This will help improve the emission reduction capabilities of aviation fuel and its role in the green development of civil aviation, and build an aviation alternative fuel sustainability assessment system that is benchmarked against ICAOCORSIA. ICAOCORSIA is a global market emission reduction mechanism developed by the International Civil Aviation Organization (ICAO) to address the problem of rising total carbon dioxide emissions in the aviation industry.

[0016] This application is mainly used to accurately and standardizedly quantify the emission reduction values ​​of aviation alternative fuels, develop an aviation fuel full life cycle carbon footprint evaluation model, form an aviation fuel full life cycle carbon footprint database, complete the aviation fuel full life cycle carbon footprint accounting and evaluation analysis, and realize the management of the aviation fuel whole process chain, including various institutions, certificates, processes, standards, etc. through the aviation fuel sustainability evaluation system. It implements unified office management, system management and website management, comprehensively analyzes the aviation fuel emission reduction capacity, and supports the low-carbon green development of civil aviation.

[0017] According to one aspect of the embodiments of the present application, a method for evaluating the sustainability of sustainable aviation fuel is provided, such as Figure 1 As shown, the method includes: S101: Obtain carbon footprint data for sustainable aviation fuel (SAF) during the raw material acquisition, product production, transportation, and use stages. Combined with the LCA industry database, this data will be used to construct a dynamic carbon footprint model for the entire life cycle and a distributed carbon footprint database. In some embodiments of the present application, specifically, the following LCA stage division can be performed: (1) source production (such as raw material planting); (2) source regulation (such as raw material harvesting, collection and recycling); (3) raw material processing and extraction (4) transportation of raw materials to processing and fuel production facilities; (5) conversion of raw materials to fuel; (6) transportation and distribution of fuel to mixing points; (7) transportation of fuel from mixing points to aircraft launch points; and (8) combustion of fuel in aircraft engines.

[0018] In some embodiments of the present application, the raw material sources of SAF include agricultural and forestry waste, energy crops, municipal solid waste, industrial waste gas, animal waste, and microbial biomass; the raw materials of SAF are bound to corresponding sustainability certification documents; the LCA (Life Cycle Assessment) industry database includes data sets for raw materials, energy, transportation, and chemicals, and the data sets have built-in IPCC greenhouse gas equivalent factors; there are no less than 400 data sets; the data set format complies with GB / T24040, GB / T24044, and EU PEF standards; S102, based on the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, the carbon emission factor is updated in real time, and the carbon emission results of the SAF are calculated in combination with the process parameters of each manufacturing process; In the process of calculating the carbon emission results of SAF, the ignorable rules (cut-off) and the data unavailable rules are applied; the cut-off rules include ignoring data with negligible environmental impact throughout the entire life cycle, such as if the carbon emission ratio of a certain link is less than a threshold, such as 1%), and the data unavailable rules include prioritizing the use of industry database alternative data, such as using the average energy consumption of the same area in the LCA database if there is no measured transportation energy consumption. The carbon emission results include direct carbon emission results and indirect carbon emission results. In this embodiment, the manufacturing process of SAF includes but is not limited to the hydrogenated fatty acid ester process: in the production stage, animal and vegetable oils are converted into alkanes by hydrodeoxygenation; alcohol to jet fuel process (AtJ): in the production stage, ethanol generated by biomass fermentation is dehydrated and oligomerized to produce aviation fuel; Fischer-Tropsch synthesis process (FT): in the production stage, synthesis gas (CO+H2) is catalytically synthesized into long-chain hydrocarbons; power-to-liquid process (PtL): in the production stage, green electricity is used to electrolyze water to produce hydrogen, and CO2 is combined to synthesize liquid fuel; raw material pretreatment process: in the raw material acquisition stage, including biomass crushing and impurity removal (such as straw crushing and screening); waste material regeneration process: in the waste stage, such as heavy metal recovery of spent catalysts and regeneration and purification of waste oils.

[0019] Process parameters include but are not limited to physical parameters, chemical parameters and equipment parameters; among them, physical parameters include temperature, pressure, flow rate, etc.; chemical parameters include catalyst dosage, reaction time, pH value, etc.; equipment parameters include stirring speed and heating power.

[0020] S103, using a recycling cycle carbon emission allocation algorithm, combined with the waste generation volume, recycling feasibility coefficient, and landfill carbon emission factor of each manufacturing process, to determine the environmental benefit deduction for waste recycling, landfill disposal, and energy recovery; In some embodiments of the present application, when the SAF raw materials are crops and their by-products, the land use conversion carbon emission calculation method is used to combine various parameters such as crop planting land area, yield, soil organic carbon, biomass, etc. to model and calculate the greenhouse gas emissions caused by land use conversion during the planting process of SAF fuel; When the SAF raw material comes from municipal solid waste, the greenhouse gas emissions and deductible environmental benefits of biomass fuels using municipal solid waste as raw material at the end of their life cycle due to the main treatment methods such as recycling, landfilling, incineration, and composting are calculated based on various parameters such as solid waste landfill management conditions, waste type, material recovery data, landfill gas generation, collection, and utilization; S104: Based on the environmental benefit deduction and carbon emission results, a sustainability evaluation report is generated, including fossil sources, biological sources, land use changes, resource consumption intensity and waste treatment environmental effects.

[0021] It should be noted that the carbon emission factor and the landfill treatment carbon emission factor are parallel specific scenario parameters. The former is a universal concept for the entire life cycle, and the latter is an exclusive parameter for the landfill treatment link in the waste stage.

[0022] refer to Figure 1 The sustainability evaluation method for sustainable aviation fuel provided in this application belongs to the technical field of high-efficiency energy-saving engineering assessment services.

[0023] The method provided in this application has the following beneficial effects: by integrating the carbon footprint data of SAF from raw material acquisition to product use throughout its life cycle, and building a dynamic model based on the LCA industry database, the problem of traditional methods focusing only on carbon emissions in the production or combustion stage is solved, and the accurate positioning of carbon emission sources and cross-link correlation impact assessment are achieved. The use of a dynamic coupling mechanism of real-time updated carbon emission factors and process parameters overcomes the carbon emission calculation deviation caused by fixed parameters, ensuring that the carbon emission differences of different manufacturing processes are truly mapped in the model. Through multi-dimensional quantitative analysis of waste generation, recycling feasibility coefficient and landfill factor, a recycling cycle allocation algorithm based on actual material flow is constructed, which changes the extensive model of traditional fixed ratio deduction and enables dynamic matching of environmental benefit accounting with waste treatment scenarios. The generated sustainability evaluation report integrates multi-dimensional indicators such as fossil sources, biomass sources, and land use changes to build a comprehensive evaluation system covering resource consumption intensity and waste treatment effects, significantly improving the integrity and accuracy of SAF's sustainability evaluation.

[0024] As an optional solution, carbon footprint data includes product name, lower calorific value, raw material name, storage and delivery time, quantity, location, transportation method, sustainability information, and inspection information. Sustainability information includes raw material carbon intensity, land use compliance, and non-grain raw material certification. Inspection information includes but is not limited to raw material heavy metal content test results, production wastewater COD / BOD5 indicators, and fuel calorific value test reports. Obtain carbon footprint data for sustainable aviation fuel (SAF) during the raw material acquisition, product production, transportation, and use stages. Combined with the LCA industry database, a dynamic carbon footprint model for the entire life cycle and a distributed carbon footprint database are constructed, including: Determine the system boundary of the carbon footprint throughout the SAF life cycle; the system boundary covers the raw material acquisition, product production, transportation and use stages, and the system boundary is defined in three dimensions: stage, space and time; In this embodiment, the stage dimension covers the core links of raw material acquisition, product production, transportation, and use. The spatial dimension clarifies the implementation locations of each stage, such as the latitude and longitude coordinates of the raw material production area, the geographical address of the processing plant, and the location of the transportation route nodes. The time dimension records key time nodes, such as the specific time of raw material entry and exit, the operation cycle of each process unit, the start and end time of transportation, and the implementation period of waste treatment. Divide the system into multiple core stages according to its boundaries, identify the direct and indirect carbon emission sources in each stage, and bind them to the corresponding manufacturing process nodes; Based on the direct and indirect carbon emission sources at each stage, the carbon input and output at each stage are calculated using the material balance method. Combined with the LCA industry database, a distributed carbon footprint database and a dynamic carbon footprint model for the entire life cycle are constructed. Among them, the dynamic model of carbon footprint for the whole life cycle couples the carbon intensity of raw materials, resource consumption and environmental impact factors of the LCA industry database; the dynamic model of carbon footprint for the whole life cycle is dynamically updated based on real-time data of the distributed carbon footprint database.

[0025] In the embodiment of the present application, real-time data includes energy consumption and raw material data after changes in process parameters; the environmental impact factors of the LCA industry database include but are not limited to global warming potential (GWP), ozone depletion potential (ODP), acidification potential (AP), eutrophication potential (EP), photochemical oxidation potential (POCP), abiotic resource depletion potential (ADP), human toxicity potential (HTP) and IPCC carbon emission coefficient, etc.

[0026] In the embodiments of this application, direct carbon emission sources include, but are not limited to, the following: Raw material acquisition: fuel emissions from biomass planting / collection machinery, such as CO2 produced by diesel fuel used in agricultural machinery; fugitive emissions from fossil fuel extraction, such as methane leaks from natural gas extraction. Production: other direct process emissions: gasification-Fischer-Tropsch (FT) synthesis, which produces CO2 from biomass pyrolysis and gasification; alcohol-to-liquids (ATJ), which produces fuel combustion emissions due to the high-temperature energy consumption of alcohol dehydration reactions; and fuel combustion in production equipment, such as natural gas consumption in heating furnaces and steam boilers. Transportation: fuel consumption by transportation vehicles, such as diesel / gasoline combustion during transportation by tanker trucks and ships. Use: aircraft engine combustion emissions, such as CO2 produced by the combustion of SAF in aircraft engines.

[0027] Indirect carbon emission sources include, but are not limited to: Raw material acquisition phase: land use change in biomass cultivation, such as carbon sink loss caused by deforestation; fertilizer production emissions, such as N2O emissions from nitrogen fertilizer production in agricultural planting. Production phase: Implicit emissions from electricity consumption, such as grid electricity used in the production process, such as coal-fired power generation; Catalyst regeneration energy consumption, such as the energy consumption for catalyst regeneration in the HEFA process. Transportation phase: Emissions from transportation infrastructure construction, such as the implicit emissions from pipeline and tanker truck manufacturing. Equipment manufacturing phase: Emissions from production equipment manufacturing, such as carbon emissions from metal processing processes such as reactors and hydrogenation units; Transportation vehicle manufacturing emissions, such as the implicit emissions from tanker truck and ship manufacturing.

[0028] Based on the embodiments provided in this application, by defining the system boundaries in three dimensions: stage, space, and time throughout the SAF life cycle, full coverage of multiple core stages from raw material acquisition to product use is achieved, solving the problem of carbon footprint omissions caused by fuzzy boundaries in traditional evaluations, such as ignoring the implicit emissions of cross-regional transportation or the time effect of long-term storage. The material balance method is used to accurately calculate the carbon input and output of each stage, and the direct / indirect carbon emission sources are bound to specific manufacturing process nodes (such as the energy consumption of the hydrogenation reactor in the HEFA process and the carbon emissions of biomass fermentation in the AtJ process), so that the distributed database can store real-time data at the process granularity (such as catalyst replacement time and raw material in and out positioning). By coupling the carbon intensity of raw materials, resource consumption and LCA environmental impact factors, the dynamic model can automatically respond to data updates, such as real-time correction of fossil resource depletion potential calculation parameters when the transportation mode changes. Compared with the traditional static model, the spatiotemporal resolution and process adaptability of carbon emission accounting for different processes and stages are significantly improved, providing an accurate data base for subsequent contribution rate analysis.

[0029] As an optional option, carbon emission results include no less than 10 environmental impact indicators, including: climate change indicators, primary energy consumption, water resource consumption, abiotic resource depletion potential, acidification potential, eutrophication potential, ozone layer depletion potential, terrestrial ecotoxicity potential, marine ecotoxicity potential, and fossil resource depletion potential; Based on the full life cycle carbon footprint dynamic model and distributed carbon footprint database, the carbon emission factor is updated in real time. Combined with the process parameters of each manufacturing process, the carbon emission results of SAF are calculated, including: Based on the LCA allocation rules, the carbon emission contribution rate of each manufacturing process is calculated according to the resource consumption ratio and process parameters; In this example, LCA allocation rules include mass balance or economic allocation. Resource consumption ratios include raw material conversion rate and energy consumption percentage. For example, if a process's energy consumption accounts for 30% of the total lifecycle energy consumption, its contribution is 30% multiplied by the carbon emission factor for that stage. Manufacturing processes include hydrogenated fatty acid esters, alcohol-to-jet fuel, Fischer-Tropsch synthesis, electro-to-liquid conversion, raw material pretreatment (including biomass pulverization and impurity removal), and waste recycling (such as heavy metal recovery from spent catalysts and regeneration and purification of waste oils).

[0030] Process parameters include physical parameters, chemical parameters, and equipment parameters, among which physical parameters include temperature, pressure, flow rate, etc.; chemical parameters include catalyst dosage, reaction time, pH value, etc.; and equipment parameters include stirring speed, heating power, etc.

[0031] Group process parameters by function and define the strategy space for each group of parameters; In this embodiment, the process parameters are grouped according to their functions, namely, a raw material processing group, an equipment operation group, and a waste material regeneration group.

[0032] With the goal of minimizing the contribution to total carbon emissions, a group strategy gradient optimization algorithm is used to perform gradient search on each group of parameters to generate multiple groups of parameter combinations that cover process fluctuations and external uncertainties. In this embodiment, process fluctuations are covered, such as decreased equipment efficiency; external uncertainties, such as policy carbon tax adjustments; The extreme learning machine model is used to dynamically predict the carbon emission results under various parameter combinations.

[0033] Among them, the extreme learning machine model (Extreme Learning Machine, referred to as ELM); the group policy gradient optimization algorithm, namely Group Policy Gradient Optimization Algorithm.

[0034] Based on the examples provided in this application, a multi-dimensional quantitative assessment of SAF sustainability is achieved by defining 10 environmental impact indicators (covering dimensions such as climate change, resource consumption, and ecotoxicity). The carbon emission contribution rate of each process is calculated based on the LCA allocation rule, and process parameters are grouped by function (for example, the "hydrogen consumption-reaction temperature" of the HEFA process is set as a group of parameters). The strategy space for each group of parameters is defined, enabling the model to analyze the impact of different parameters on each indicator at the process granularity level (for example, identifying the contribution weight of alcohol conversion rate to eutrophication potential in the AtJ process). Parameter combinations are searched using a group strategy gradient optimization algorithm to generate multiple sets of parameter solutions that cover process fluctuations (such as a 10% decrease in catalyst efficiency) and external uncertainties (such as the addition of an ozone layer depletion indicator to the policy). This allows the evaluation results to reflect parameter change scenarios in actual production (for example, the deduction of emission results when the carbon intensity of the raw material fluctuates by ±15%), providing more practical guidance for process optimization.

[0035] The dynamic adjustment rules of the policy space include: When it is detected that the carbon intensity fluctuation of the raw material exceeds the threshold, the strategy space of the raw material processing group is automatically shrunk to the high conversion efficiency range; If the waste recycling feasibility coefficient is lower than the preset value, the strategy space switches to the interval dominated by landfill treatment.

[0036] For example, if the historical mean carbon intensity of raw materials is 50gCO2 / MJ with a standard deviation of 5gCO2 / MJ, a threshold of twice the standard deviation (±10gCO2 / MJ) is used. This means that when the carbon intensity is >60gCO2 / MJ or <40gCO2 / MJ, the policy space is contracted. The default value is 30%. When the recyclability factor of a process waste falls below 30% (e.g., 25%), the policy space automatically switches to landfill-dominated treatment (weight >60%).

[0037] As an optional solution, the dynamic prediction and correction of the extreme learning machine model includes: The number of hidden layer nodes is dynamically expanded according to the input parameter dimension, and the activation function adopts an adaptive piecewise function; Through reverse error propagation, direct emission factors, indirect emission factors and implicit emission factors are dynamically corrected.

[0038] In this embodiment, direct emission factors include combustion exhaust gas concentration, indirect emission factors include carbon intensity of electricity production, and implicit emission factors include carbon sequestration effect of land tillage.

[0039] In the embodiments of this application, Figure 2 Figure 2 shows a schematic diagram of dynamic carbon emissions prediction and correction. This diagram includes: Dynamic input parameter detection: The input parameter set includes process parameters for each manufacturing process (such as conversion rate, energy intensity, and equipment maintenance cycle). The input parameter dimension detection module monitors parameter changes in real time (such as new raw material types or transportation methods), triggering dynamic adjustments to hidden layer nodes. For example, if "enzyme catalysis process parameters" are added, the input parameter dimension detection module recognizes the increase in dimension and triggers the expansion of hidden layer nodes.

[0040] Dynamic adjustment of hidden layer nodes: Hidden layer node expansion: When the parameter dimension increases, new hidden layer nodes are added to cover the new features, and the activation function uses an adaptive piecewise function (such as a linear function for sparse input areas and a sigmoid function for dense areas). Hidden layer node merging: When the parameter dimension decreases (such as when a process is eliminated), redundant nodes are merged to reduce model complexity.

[0041] ELM prediction and error correction: The ELM model outputs carbon emission forecasts based on the adjusted network structure. The error calculation module compares the forecasts with actual carbon emission data to generate an error signal. Backpropagation correction: The error signal dynamically corrects three types of emission factors through backpropagation: Direct emission factors correct for carbon emissions monitored in real time, such as production energy consumption and transportation fuel combustion; Indirect emission factors correct for indirect effects such as land use changes caused by raw material acquisition and the embodied carbon in equipment manufacturing; and Implicit emission factors correct for carbon emissions not directly monitored in the supply chain (such as packaging material production and unaccounted logistics).

[0042] Based on the examples provided in this application, a dynamic network structure breaks through the traditional fixed ELM architecture and addresses the variable parameter dimensions caused by the diversity of SAF feedstocks (such as waste oils and agricultural and forestry wastes). A targeted correction mechanism: This mechanism categorizes and corrects direct, indirect, and implicit emission factors to accurately adapt to the complexity of SAF's full lifecycle carbon emissions (for example, biogenic carbon offsets require strict separation of implicit emissions).

[0043] Based on the embodiments provided herein, the dynamic expansion of the number of hidden layer nodes in the extreme learning machine model (automatically adding computational nodes based on the input parameter dimensions, such as adding mercury ion toxicity computational nodes when adding "marine ecotoxicity potential") addresses the lack of compatibility of traditional fixed-structure models with newly added environmental indicators, enabling the model to flexibly adapt to assessment dimensions that may be included in the future (such as microplastic pollution and photochemical ozone creation potential). Backward error propagation is used to dynamically correct three types of emission factors. For example, when the carbon intensity test value of a batch of biomass feedstock deviates from the database estimate by more than 5%, the model automatically corrects the land use change impact factor corresponding to that feedstock. This avoids the lag and subjectivity caused by manual correction factors in traditional methods, ensuring that emission factors are always synchronized with the latest test data.

[0044] As an optional solution, a recycling carbon emission allocation algorithm is used to combine the waste generation volume, recycling feasibility coefficient and landfill carbon emission factor of each manufacturing process to determine the environmental benefit deduction of waste recycling, landfill disposal and energy recovery, including: Based on the amount of waste generated, the regeneration feasibility coefficient, and the carbon emission contribution rate, a dynamic weighting model for waste regeneration, landfill treatment, and energy recovery is constructed. The impact of historical regeneration success rates on the carbon benefits of the current cycle is determined through a Markov chain, generating a state-dependent apportionment ratio. In this embodiment, the historical regeneration success rate, such as the enzyme-catalyzed decomposition efficiency; A multi-objective particle swarm optimization algorithm is used to optimize the weight distribution under the constraints of Markov chains, with the goal of maximizing environmental benefits and balancing treatment methods. Combine landfill permeability and methane emission factors, convert to CO2 equivalent according to GWP100 value, and quantify geological stability risk; The environmental benefit deductions of waste recycling, landfill disposal and energy recovery are normalized to generate a total environmental benefit value.

[0045] In this embodiment, after the total environmental benefit value is generated, the total environmental benefit value and the deduction details (including the allocation ratio, calorific value matching logic, and penetration data) are linked to the blockchain for evidence storage.

[0046] Among them, the multi-objective particle swarm optimization algorithm (Multi-ObjectiveParticleSwarmOptimization, MOPSO).

[0047] Optionally, in this embodiment, each deduction result can be linked to the process node data on the blockchain to record the allocation ratio, risk quantification parameters and PSO optimization path; automatically verify whether the environmental benefits comply with regional policies, such as the EU ReFuelEU's minimum requirements for recycling weights, and trigger an early warning in the event of an abnormality.

[0048] Based on the embodiments provided herein, a dynamic weighting model (combining waste generation volume, regeneration feasibility coefficient, and carbon emission contribution rate) is constructed to achieve differentiated weight allocations for waste from different processes. For example, due to the low regeneration feasibility of high-ash waste generated by the Fischer-Tropsch synthesis process, the model automatically reduces its regeneration weight to below 40% and increases its energy recovery weight. A Markov chain is introduced to determine the impact of historical regeneration success rates on current benefits, enabling carbon benefit allocation to reflect the dynamic maturity of the regeneration technology. For example, processes with continuously successful operations receive a higher carryover rate. This avoids the static allocation flaw of traditional methods that "ignore" historical performance, such as applying the same deduction ratio to newly commissioned regeneration equipment and mature equipment. Using a multi-objective particle swarm optimization algorithm (with the objectives of maximizing environmental benefits and balancing treatment methods), weight allocation is optimized within the constraints of the Markov chain. For example, when regeneration equipment capacity is insufficient, the regeneration weight is automatically capped, ensuring that the evaluation results meet actual production constraints. This addresses the disconnect between theoretical benefits and engineering feasibility in traditional methods.

[0049] As an optional solution, the construction of Markov chain includes: Define the regeneration system state set: S1: regeneration success rate ≥ 80%; S2: 50% ≤ success rate < 80%; S3: success rate < 50%; State transition rule: Based on the historical regeneration success rate, the state transition probability matrix is ​​trained. The state transition probability matrix represents the transition probability between each state; Carbon benefit allocation ratio binding: Each state corresponds to a specific carbon benefit allocation ratio, among which S1: allows 90% of historical recycled carbon benefits to be carried forward to the current life cycle; S2: allows 60% to be carried forward; S3: allows 30% to be carried forward.

[0050] Based on the embodiments provided in this application, by defining a three-level regeneration state set (S1 / S2 / S3) and binding differentiated allocation ratios (90% / 60% / 30%), the carbon benefit carryover rules can quantitatively reflect the historical performance of the regeneration technology. For example, when the initial regeneration success rate of a process is 45% (S3 state), only 30% of the historical benefits are allowed to be carried forward. As the technology improves to 85% (S1 state), the carryover ratio increases to 90%. This incentivizes companies to continuously optimize their regeneration processes and solves the fairness issue of "indifferent allocation between good and bad processes" in traditional methods. Based on the state transition probability matrix trained based on historical regeneration success rates, the model can predict transition trends between different states (for example, the S2 state has a 50% probability of maintaining, a 30% probability of upgrading, and a 20% probability of downgrading). This provides data-driven decision support for companies to develop regeneration technology improvement paths (for example, prioritizing resources to reduce the transition probability from S2 to S3), which is more scientific than traditional empirical judgment.

[0051] As an optional solution, the optimization rules of the multi-objective particle swarm optimization algorithm include: Objective function: Maximize total environmental benefits and minimize weight differences among treatment methods; where total environmental benefits = waste recycling benefits + energy recovery benefits - landfill carbon emissions; Dynamic constraint: The upper limit of the regeneration weight is dynamically adjusted by the allocation ratio of the current Markov state; for example, when the state is S1, the upper limit of the regeneration weight = 0.8 × the allocation ratio corresponding to S1 (i.e. 0.8 × 90% = 72%); when the state is S3, the upper limit of the regeneration weight = 0.8 × 30% = 24%; Particle update rule optimization: The particle velocity update formula introduces the Markov state offset. The Markov state offset is calculated by the transition probability difference between the current state and the historical state, and is used to guide the weight distribution trend.

[0052] In this embodiment, if Figure 3 As shown in , it is a schematic diagram of a multi-objective optimization method under Markov chain constraints. The figure includes a left subgraph and a right subgraph; Among them, in the right sub-graph, the Markov chain state transition: State definition: S1: regeneration success rate ≥ 80%, indicating a high-efficiency regeneration state (such as a mature enzyme-catalyzed waste oil and grease process); S2: 50% ≤ success rate < 80%, indicating a medium regeneration state (such as a waste plastic chemical decomposition process); S3: success rate < 50%, indicating an inefficient regeneration state (such as uncertified waste material treatment).

[0053] State transition rules: Arrows represent the probability of transition between states (e.g., S1 has a 70% probability of remaining in its own state (P=0.7), a 20% probability of transitioning to S2, and a 10% probability of transitioning to S3); Among them, in the left sub-figure, multi-objective PSO optimization: Optimization engine (PSO): Input: Goal 1 (maximize total environmental benefits): total benefits = waste recycling benefits + energy recovery benefits − landfill carbon emissions; Goal 2 (minimize treatment method differences): weight difference = maximum weight − minimum weight (ensure balanced distribution of recycling, landfill, and energy recovery).

[0054] Dynamic constraints: The upper limit of the regeneration weight is dynamically limited by the allocation ratio α of the current Markov state (for example, W1≤0.8×90%=72% at S1); the landfill weight W2 is constrained by the geological risk factor, and the energy recovery weight W3 is dynamically associated with the regional power grid factor.

[0055] Output: W1 (recycling weight): the proportion of environmental benefits allocated to waste recycling; W2 (landfill weight): the proportion of carbon emission deductions allocated to landfill; W3 (energy recovery weight): the proportion of emission reduction contributions allocated to energy recovery.

[0056] Between the left and right subgraphs, the carbon benefit allocation ratio α (90% / 60% / 30%) for S1 / S2 / S3 serves as the upper bound constraint for the PSO weights. For example, in state S1, the regeneration weight W1 can be allocated up to 72% (0.8 × 90%) to avoid over-reliance on inefficient processes. A Markov state offset Δ is introduced into the particle velocity formula, calculated from the difference in state transition probabilities (for example, as the probability of S2 → S1 increases, Δ guides the particle search toward a higher W1).

[0057] Based on the examples provided in this application, dynamic state-weight binding is achieved by directly mapping the maturity of recycling technologies (S1 / S2 / S3) to weight constraints, resolving the conflict between technology fluctuations and multi-objective optimization in SAF waste treatment. The landfill weight W2 is constrained by geological risk factors (such as clay layer thickness), and the energy recovery weight W3 is linked to the grid carbon emission factor, ensuring that the calculation results are accurate for real-world scenarios.

[0058] Based on the embodiments provided herein, a dual-objective function design—maximizing total environmental benefits and minimizing differences in treatment method weights—avoids the concentration of systemic risks (e.g., equipment overload caused by a single treatment method weight exceeding 70%) caused by a singular pursuit of recycling benefits. For example, when the recycling feasibility coefficient suddenly drops, the algorithm automatically increases the landfill / recycling weights, ensuring that the assessment results meet the multi-objective balance requirements of actual production. The introduction of Markov state dynamic constraints (the upper limit of the recycling weight is determined by the current state allocation ratio, such as an upper limit of 80% for the S1 state and 40% for the S3 state) allows weight allocation to be correlated with technology maturity. For example, mature processes (S1 state) receive higher recycling weights, aligning with the practical principle that "more mature technology, more significant recycling benefits" and addressing the flaw of traditional methods in which weight allocation is disconnected from technology status. A Markov state offset (calculated from the difference in transition probabilities between the current state and historical states) is introduced through the particle update rule to guide weight allocation trends (e.g., adjusting the weight allocation towards increasing the recycling weight in the S1 state), enabling the algorithm to converge more quickly to the optimal solution that aligns with the direction of technological evolution, improving optimization efficiency and rationality in complex scenarios.

[0059] As an optional solution, the method further includes: The data generated during the sustainability assessment process is encrypted using a hash algorithm and then privately chained using a blockchain traceability strategy. This blockchain traceability strategy uses a consortium chain architecture, whereby distributed ledgers are jointly maintained by all supply chain nodes, including raw material collectors, processors, manufacturers, logistics nodes, and certification bodies. The blockchain traceability strategy utilizes a Book & Claim mechanism, which uses smart contracts to automatically verify the qualifications of each supply chain node. For example, it retrieves ISCCEU certification numbers and corporate carbon emission quota data, supporting tamper-proof traceability of the entire supply chain, from raw material collection to fueling. The Book & Claim mechanism separates the physical and environmental properties of aviation fuel, generating a traceable green certificate that is linked to a specific production batch.

[0060] Based on the embodiments provided herein, a blockchain traceability strategy enables the immutable storage of sustainability assessment data. For example, any modification to raw material carbon intensity data after it is uploaded to the blockchain is recorded, resolving the trust pain point of traditional methods, where data tampering can distort assessment results, such as the artificial adjustment of land use compliance data to pass certification. The Book & Claim mechanism and smart contracts automatically verify node qualifications, enabling full-chain traceability from raw material collection to fuel refueling. For example, scanning a SAF fuel batch number can be traced back to the raw material planting records and recycling equipment numbers for a specific plot. This meets the requirements of international standards (such as CORSIA) for verifiable emission reductions and traceable accountability, providing the technical infrastructure for SAF to participate in global carbon market transactions. The consortium chain distributed ledger enables real-time synchronization and sharing of data across the entire supply chain. For example, when a logistics node updates transportation energy consumption data, the manufacturer and certification body simultaneously obtain and trigger a model recalculation. This eliminates the assessment delays and disagreements caused by traditional "data silos," significantly improving assessment efficiency and data consistency in multi-agent collaborative scenarios.

[0061] Blockchain evidence further includes: Each environmental impact indicator is linked to a green certificate on the blockchain, such as the Book & Claim certificate, which records the LCA calculation version and participant information; Smart contracts are used to verify whether indicator data complies with regional policies, such as the EU ReFuelEU blending ratio, and trigger early warnings when abnormalities occur.

[0062] As an optional option, based on environmental benefit offsets and carbon emission results, a sustainability assessment report is generated that includes fossil sources, biomass sources, land use change, resource consumption intensity, and waste treatment environmental effects, including: Based on a dynamic model of carbon footprint throughout the entire life cycle, carbon emissions from fossil sources and biomass are separated and bound to the raw material source data stored on the blockchain; Verify the sustainability certification information of bio-sourced raw materials through blockchain smart contracts to ensure their carbon emission offsets are valid; In this embodiment, the sustainability certification information may include, but is not limited to, non-grain certification and RSB certification; The carbon sink gains and losses of biogenic raw materials are calculated according to IPCC standards using historical land use data in the distributed carbon footprint database, and the carbon sink loss value is dynamically revised based on real-time land use changes. In this embodiment, real-time land use change is such as conversion of forestland to farmland; Extract resource consumption indicators from carbon emission results and classify and calculate the resource consumption intensity ratio of each process node; In this embodiment, resource consumption indicators include primary energy consumption, water resource consumption, and non-biological resource consumption potential; process parameters include conversion rate and energy intensity; Unify the landfill risk quantification results and recycled carbon benefits into CO2 equivalents based on the GWP100 value and link them to blockchain for evidence storage; In this embodiment, the risk quantification results include the confidence interval of methane emission; the regenerative carbon benefits include the Markov chain-based allocation ratio; The classification results are linked to the supply chain map stored in the blockchain, and the changing trends of various indicators are dynamically displayed through the timeline; In this embodiment, the classification results include fossil sources, biogenic sources, land use change, resource consumption intensity, and waste effects; Generate a sustainability evaluation report that includes a core indicator table, source contribution chart, treatment effect details, and a blockchain QR code. Scan the code to verify the data source and calculation logic.

[0063] In this embodiment, the core indicator table lists 10 indicators such as climate change and resource consumption and their confidence intervals; the source contribution chart shows the proportion of net carbon emissions from fossil sources and biogenic sources, and the carbon sink profit and loss value of land use; the treatment effect details include the deduction values ​​and risk parameters of regeneration, landfill and energy recovery; and the blockchain verification code allows users to scan the code to trace the data of the entire life cycle.

[0064] Optionally, in this embodiment, if Figure 4 The figure shows a SAF product life cycle and life cycle assessment flow chart. The left side of the figure illustrates the raw material sources and production steps for various SAF production processes (such as HEFA, ATJ, G-FT, and PtL). For example, the HEFA process uses waste oils and energy plants as raw materials and undergoes pretreatment, hydrodeoxygenation, hydrocracking, and isomerization. The ATJ process utilizes biomass waste and exhaust gas and undergoes pretreatment, fermentation / hydrolysis, alcoholization, dehydration, oligomerization, and hydrogenation. The text on the right explains that the SAF product life cycle assessment covers stages such as raw material production / acquisition, product production, transportation / storage, and use, and quantifies and analyzes environmental impact indicators (carbon footprint) throughout the entire life cycle. The life cycle assessment subsystem (also known as the sustainable aviation fuel sustainability assessment system) complies with the SAF product life cycle assessment requirements of international ISO standards, GB / T standards, the EU PEF, and CORSIA (Carbon Offsetting and Reduction Scheme for International Aviation) standards. The following process shows the complete process from determining boundary conditions, conducting life cycle carbon footprint modeling and calculation, to data quality assessment, and finally generating a carbon footprint report.

[0065] It should be noted that fossil sources generally refer to energy materials formed from the remains of ancient organisms through long-term geological processes, mainly including coal, oil, and natural gas. Biogenic sources refer to substances or energy that are directly or indirectly derived from organisms, including: waste oils and fats, energy plants, biomass waste gas, biomass solid waste, green electricity, and water.

[0066] Based on the embodiments provided in this application, by separating fossil and biogenic carbon emissions and binding them to blockchain raw material data, accurate traceability of SAF carbon sources is achieved, such as distinguishing the carbon emission contributions of petroleum-based raw materials and waste oil raw materials. This solves the sustainability misjudgment problem caused by the confusion of fossil and biogenic emissions in traditional methods, such as mistakenly counting fossil fuel emissions as biogenic deduction items. Blockchain smart contracts are used to automatically verify the sustainability certification of biogenic raw materials, ensuring that only carbon emissions from biogenic raw materials that meet the standards can be deducted. This avoids the illegal deduction loophole caused by "forged or expired certification documents" in traditional manual review, and provides machine consensus-level trust guarantee for carbon benefit accounting. Based on historical land use data and real-time use changes, carbon sink gains and losses are dynamically corrected. For example, when a biomass raw material production area is converted from forest to arable land, the model automatically increases the corresponding land use change carbon loss value, solving the defect of "static carbon sink parameters cannot reflect real-time land changes" in traditional methods, making the sustainability assessment of biogenic raw materials more in line with actual ecological impacts. By categorizing and calculating the resource consumption intensity ratios of each process node, such as the proportion of hydrogen consumption in the HEFA process to total energy consumption during the production phase and the proportion of water consumption in the AtJ process, high-resource consumption links, such as coal consumption in the Fischer-Tropsch synthesis process, can be accurately identified. This provides a quantitative basis for companies to develop "process node-level" resource optimization plans, changing the extensive model of traditional methods that lack detailed overall resource consumption statistics. By unifying the quantified results of landfill disposal risks and recycled carbon benefits into CO2 equivalents (converted according to the GWP100 value and incorporating a geological stability risk adjustment factor), and linking them to blockchain evidence, the environmental impact of the waste disposal stage can be directly compared with indicators at other stages. This addresses the problem of traditional methods that "multi-dimensional environmental impacts cannot be uniformly measured," and forms a standardized assessment system for environmental impacts throughout the entire life cycle. The blockchain timeline dynamically displays the changing trends of various indicators (such as the downward curve of the fossil resource depletion potential at a certain process node as the catalyst is upgraded), and generates a visual report containing a blockchain QR code (scanning the code can verify the source of raw materials, calculation logic and stored data). This makes the sustainability evaluation results traceable, verifiable and interpretable, meets the requirements of international certification bodies such as CORSIA for transparency in the assessment process, and provides visual technical endorsement for SAF's participation in global carbon market transactions.

[0067] According to another aspect of the embodiments of the present application, a sustainability evaluation system for sustainable aviation fuel is provided. Figure 5As shown in , the system includes: The carbon footprint model and database construction module 501 is used to obtain carbon footprint data of sustainable aviation fuel (SAF) during the raw material acquisition, product production, transportation, and use stages, and to build a full life cycle carbon footprint dynamic model and a distributed carbon footprint database in combination with the LCA industry database; The SAF carbon emission result calculation module 502 is used to update the carbon emission factor in real time based on the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, and calculate the carbon emission results of the SAF in combination with the process parameters of each manufacturing process; Environmental benefit deduction determination module 503 is used to determine environmental benefit deductions for waste recycling, landfill disposal, and energy recovery using a recycling cycle carbon emission allocation algorithm, taking into account the waste generation volume, recycling feasibility coefficient, and landfill carbon emission factor of each manufacturing process; The sustainability evaluation report generation module 504 is used to generate a sustainability evaluation report including fossil sources, biomass sources, land use changes, resource consumption intensity and waste treatment environmental effects based on environmental benefit deductions and carbon emission results.

[0068] refer to Figure 5 The waste cooking oil traceability management system for sustainable aviation fuel provided in this application belongs to the technical field of high-efficiency energy-saving engineering assessment services.

[0069] Optionally, in this embodiment, the embodiments to be implemented by the above-mentioned various unit modules can refer to the above-mentioned various method embodiments, which will not be repeated here.

[0070] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0071] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A sustainability assessment method for sustainable aviation fuel, characterized in that: include: Obtain carbon footprint data for sustainable aviation fuel (SAF) during the raw material acquisition, product production, transportation, and use stages. Combined with the LCA industry database, a dynamic carbon footprint model for the entire life cycle and a distributed carbon footprint database will be constructed. Based on the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, the carbon emission factor is updated in real time, and the carbon emission results of the SAF are calculated in combination with the process parameters of each manufacturing process; A recycling carbon emissions allocation algorithm is used to determine the environmental benefit deductions for waste recycling, landfill disposal, and energy recovery, taking into account the waste generation volume, recycling feasibility coefficient, and landfill carbon emission factor of each manufacturing process; Based on the environmental benefit deduction and the carbon emission results, a sustainability evaluation report including fossil sources, biological sources, land use changes, resource consumption intensity and waste treatment environmental effects is generated.

2. The sustainability assessment method for sustainable aviation fuel according to claim 1, characterized in that: The carbon footprint data includes product name, lower calorific value, raw material name, storage and delivery time, quantity, location, transportation method, sustainability information and inspection information; the sustainability information includes raw material carbon intensity, land use compliance and non-grain raw material certification; The carbon footprint data of sustainable aviation fuel (SAF) during the raw material acquisition, product production, transportation and use stages are obtained and combined with the LCA industry database to build a full life cycle carbon footprint dynamic model and a distributed carbon footprint database, including: Determine the system boundary of the carbon footprint throughout the SAF life cycle; the system boundary covers the raw material acquisition, product production, transportation and use stages, and the system boundary is defined in three dimensions: stage, space and time; Divide the system into multiple core stages according to the system boundaries, identify the direct and indirect carbon emission sources in each stage, and bind them to the corresponding manufacturing process nodes; Based on the direct and indirect carbon emission sources at each stage, the carbon input and output at each stage are calculated using the material balance method, and combined with the LCA industry database, the distributed carbon footprint database and the full life cycle carbon footprint dynamic model are constructed; Among them, the full life cycle carbon footprint dynamic model couples the raw material carbon intensity, resource consumption and LCA industry database environmental impact factors; the full life cycle carbon footprint dynamic model is dynamically updated based on real-time data of the distributed carbon footprint database.

3. The sustainability assessment method for sustainable aviation fuel according to claim 1, characterized in that: The carbon emission results include no less than 10 environmental impact indicators, including: climate change indicators, primary energy consumption, water resource consumption, abiotic resource depletion potential, acidification potential, eutrophication potential, ozone layer depletion potential, terrestrial ecotoxicity potential, marine ecotoxicity potential, and fossil resource depletion potential; The carbon emission factor is updated in real time based on the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, and the carbon emission results of the SAF are calculated in combination with the process parameters of each manufacturing process, including: Based on the LCA allocation rules, the carbon emission contribution rate of each manufacturing process is calculated according to the resource consumption ratio and process parameters; Group process parameters by function and define the strategy space for each group of parameters; With the goal of minimizing the contribution to total carbon emissions, a group strategy gradient optimization algorithm is used to perform gradient search on each group of parameters to generate multiple groups of parameter combinations that cover process fluctuations and external uncertainties. The extreme learning machine model is used to dynamically predict the carbon emission results under various parameter combinations.

4. The sustainability assessment method for sustainable aviation fuel according to claim 3, characterized in that: The dynamic prediction and correction of the extreme learning machine model includes: The number of hidden layer nodes is dynamically expanded according to the input parameter dimension, and the activation function adopts an adaptive piecewise function; Through reverse error propagation, direct emission factors, indirect emission factors and implicit emission factors are dynamically corrected.

5. The sustainability assessment method for sustainable aviation fuel according to claim 1, characterized in that: The recycling carbon emission allocation algorithm is used to determine the environmental benefit deductions for waste recycling, landfill treatment, and energy recovery, taking into account the waste generation volume, recycling feasibility coefficient, and landfill carbon emission factor of each manufacturing process. The deductions include: Based on the amount of waste generated, the regeneration feasibility coefficient and the carbon emission contribution rate, a dynamic weighting model for waste regeneration, landfill treatment and energy recovery is constructed; The impact of historical regeneration success rates on the current cycle’s carbon revenue is determined through a Markov chain, generating a state-dependent apportionment ratio; A multi-objective particle swarm optimization algorithm is used to optimize the weight distribution under the constraints of Markov chains, with the goal of maximizing environmental benefits and balancing treatment methods. Combine landfill permeability and methane emission factors, convert to CO2 equivalent according to GWP100 value, and quantify geological stability risk; The environmental benefit deductions of waste recycling, landfill disposal and energy recovery are normalized to generate a total environmental benefit value.

6. The sustainability assessment method for sustainable aviation fuel according to claim 5, characterized in that: The construction of the Markov chain includes: Define the regeneration system state set: S1: regeneration success rate ≥ 80%; S2: 50% ≤ success rate < 80%; S3: success rate < 50%; State transition rule: training a state transition probability matrix based on the historical regeneration success rate, wherein the state transition probability matrix represents the transition probability between states; Carbon benefit allocation ratio binding: Each state corresponds to a specific carbon benefit allocation ratio, among which S1: allows 90% of historical recycled carbon benefits to be carried forward to the current life cycle; S2: allows 60% to be carried forward; S3: allows 30% to be carried forward.

7. The sustainability assessment method for sustainable aviation fuel according to claim 5, characterized in that: The optimization rules of the multi-objective particle swarm optimization algorithm include: Objective function: Maximize total environmental benefits and minimize weight differences in treatment options; where total environmental benefits = waste recycling benefits + energy recovery benefits - landfill carbon emissions; Dynamic constraint: The upper limit of the regeneration weight is dynamically adjusted by the allocation ratio of the current Markov state; Particle update rule optimization: The particle velocity update formula introduces a Markov state offset. The Markov state offset is calculated by the transition probability difference between the current state and the historical state, and is used to guide the weight distribution trend.

8. The sustainability assessment method for sustainable aviation fuel according to any one of claims 1 to 7, characterized in that: The method further comprises: The data generated during the sustainability assessment process is encrypted using a hash algorithm and then privately chained using a blockchain traceability strategy. This blockchain traceability strategy uses a consortium chain architecture, whereby distributed ledgers are jointly maintained by various supply chain nodes, including raw material collectors, processors, manufacturers, logistics nodes, and certification bodies. Among them, the blockchain traceability strategy adopts the Book & Claim mechanism, which automatically verifies the qualifications of each node in the supply chain through smart contracts, and supports the tamper-proof traceability of the entire chain of data from raw material collection to fuel filling.

9. The sustainability assessment method for sustainable aviation fuel according to claim 8, characterized in that: Based on the environmental benefit deduction and the carbon emission results, a sustainability assessment report is generated including fossil sources, biomass sources, land use changes, resource consumption intensity and waste treatment environmental effects, including: Based on the dynamic model of the full life cycle carbon footprint, carbon emissions from fossil sources and biomass are separated and bound to the raw material source data stored on the blockchain; Verify the sustainability certification information of bio-sourced raw materials through blockchain smart contracts to ensure their carbon emission offsets are valid; Calculate the carbon sink gains and losses of biogenic raw materials according to IPCC standards using historical land use data in the distributed carbon footprint database, and dynamically revise the carbon sink loss value based on real-time land use changes; Extract resource consumption indicators from the carbon emission results, and classify and calculate the resource consumption intensity ratio of each process node; Unify the landfill risk quantification results and recycled carbon benefits into CO2 equivalents based on the GWP100 value and link them to blockchain for evidence storage; The classification results are linked to the supply chain map stored in the blockchain, and the changing trends of various indicators are dynamically displayed through the timeline; Generate a sustainability evaluation report that includes a core indicator table, source contribution chart, treatment effect details, and a blockchain QR code. Scan the code to verify the data source and calculation logic.

10. A sustainability assessment system for sustainable aviation fuel, characterized in that: include: The carbon footprint model and database construction module is used to obtain carbon footprint data of sustainable aviation fuel (SAF) during the raw material acquisition, product production, transportation and use stages. Combined with the LCA industry database, it builds a dynamic carbon footprint model for the entire life cycle and a distributed carbon footprint database. A SAF carbon emission result calculation module is used to update the carbon emission factor in real time based on the full life cycle carbon footprint dynamic model and the distributed carbon footprint database, and calculate the carbon emission results of the SAF in combination with the process parameters of each manufacturing process; An environmental benefit deduction determination module is used to determine environmental benefit deductions for waste recycling, landfill disposal, and energy recovery using a recycling cycle carbon emission allocation algorithm, taking into account the waste generation volume, recycling feasibility coefficient, and landfill carbon emission factor of each manufacturing process; The sustainability evaluation report generation module is used to generate a sustainability evaluation report including fossil sources, biological sources, land use changes, resource consumption intensity and waste treatment environmental effects based on the environmental benefit deduction and the carbon emission results.

Images