Gasoline blending method and device based on large model, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a gasoline blending method and device based on a large model, electronic equipment and a storage medium. According to the method, gasoline blending requirements input by a user are analyzed, blending task parameters are obtained, the blending task parameters comprise the gasoline mark number, the quality condition and the planned yield of gasoline to be blended, the inventory of multiple component oils is obtained, the minimum blending cost is used as a target function through a gasoline blending model corresponding to the gasoline mark number, and the minimum blending cost is used as the target function. The quality condition, the planned yield and the stock are used as constraint conditions, the blending proportion of each component oil is iteratively optimized, the optimal blending proportion of the multiple component oils is output, and then the blending device is controlled to execute blending operation according to the optimal blending proportion. The technical effects that the optimal blending proportion with the minimum blending cost is efficiently and accurately output under the constraint conditions of inventory, cost and quality standards, and therefore the requirements of users for blending efficiency, blending precision and low-cost blending are met are achieved.

Description

Gasoline blending method, apparatus, electronic equipment, and storage medium based on a large model Technical Field

[0001] This application relates to the field of gasoline blending technology, and in particular to a gasoline blending method, apparatus, electronic device and storage medium based on a large model. Background Technology

[0002] In the production of finished gasoline in oil refineries, gasoline blending is a core step. Refineries need to mix various component oils, such as catalytic gasoline, reformed gasoline, MTBE (Methyl-tert-butyl ether), and alkylate oil, in specific proportions to meet the preset standards for key indicators of finished gasoline, such as octane number, vapor pressure, sulfur content, benzene content, and olefin content.

[0003] Traditional blending processes rely on human experience or laboratory test data. Engineers need to adjust the blending ratio based on historical formulas and current inventory. This process involves analysis and calculation of raw material inventory, raw material costs, and gasoline blending standards.

[0004] However, manual calculations struggle to quickly find the optimal blending ratio under constraints such as inventory, cost, and quality standards, making it difficult to meet users' demands for blending efficiency, accuracy, and low-cost blending. Summary of the Invention

[0005] This application provides a gasoline blending method, apparatus, electronic device, and storage medium based on a large model, which can quickly find the optimal gasoline blending ratio under constraints such as inventory, cost, and quality standards, thereby meeting users' needs for blending efficiency, blending accuracy, and low-cost blending.

[0006] In a first aspect, embodiments of this application provide a gasoline blending method based on a large model, comprising:

[0007] The system analyzes the user's input gasoline blending requirements to obtain blending task parameters, which include: the gasoline grade, quality conditions, and planned output of the gasoline to be blended.

[0008] Obtain the inventory levels of multiple component oils;

[0009] Using the gasoline blending model corresponding to the gasoline grade, the minimum blending cost is taken as the objective function, and the quality conditions, planned output and inventory are taken as constraints. The blending ratio of each component oil is iteratively optimized and the optimal blending ratio of multiple component oils is output.

[0010] The control and blending device performs the blending operation according to the optimal blending ratio.

[0011] In one possible implementation, the above-mentioned gasoline blending model includes: an octane number calculation model, a vapor pressure calculation model, and a blending ratio recommendation model. Using the gasoline blending model corresponding to the gasoline grade, the model takes the minimum blending cost as the objective function and quality conditions, planned production volume, and inventory level as constraints to iteratively optimize the blending ratio of each component oil, outputting the optimal blending ratio of multiple component oils, including:

[0012] Generate multiple candidate harmonic ratios;

[0013] The octane number calculation model is based on multiple candidate blending ratios and the octane numbers of multiple component oils, and outputs the first octane number corresponding to each candidate blending ratio.

[0014] The vapor pressure calculation model is used to output the first vapor pressure corresponding to each candidate blending ratio based on the vapor pressure of multiple candidate blending ratios and multiple component oils.

[0015] The linear indices of each component oil are linearly superimposed to obtain the first linear index corresponding to each candidate blending ratio;

[0016] The model recommends a blending ratio with the minimum blending cost as the objective function and quality conditions, planned output, and inventory as constraints. It determines the evaluation parameters of multiple candidate blending ratios and iteratively optimizes the corresponding candidate blending ratios based on the evaluation parameters until the cutoff condition is met, and outputs the optimal blending ratio.

[0017] The evaluation parameters are obtained by evaluating the candidate blending ratio and the indicators to be evaluated based on the constraints. The indicators to be evaluated include: first octane number, first vapor pressure and first linearity index.

[0018] In one possible implementation, the above-mentioned octane number calculation model includes an octane number interpolation table. Based on multiple candidate blending ratios and the octane numbers of multiple component oils, the octane number calculation model outputs a first octane number corresponding to each candidate blending ratio, including:

[0019] The first octane number corresponding to the candidate blending ratio is determined by using an octane number interpolation table.

[0020] In one possible implementation, a vapor pressure calculation model is used to output the first vapor pressure corresponding to each candidate blending ratio based on the vapor pressures of multiple candidate blending ratios and multiple component oils, including:

[0021] Based on the candidate blending ratios, the volume fraction of each component oil is determined;

[0022] Convert the vapor pressure of multiple component oils into a vapor pressure blending index;

[0023] The first vapor pressure is determined based on the volume fraction of multiple component oils and the vapor pressure blending index.

[0024] In one possible implementation, a blending ratio recommendation model is used to determine evaluation parameters for multiple candidate blending ratios, with the minimum blending cost as the objective function and quality conditions, planned output, and inventory levels as constraints. The corresponding candidate blending ratios are then iteratively optimized based on these evaluation parameters until a cutoff condition is met, outputting the optimal blending ratio. This includes:

[0025] The blending cost of each candidate blending ratio is determined by using a blending ratio recommendation model with the minimum blending cost as the objective function and quality conditions, planned output, and inventory as constraints.

[0026] The evaluation parameters are obtained by evaluating the indicators to be evaluated based on quality conditions, and by evaluating the candidate blending ratios based on planned output and inventory levels, the second evaluation parameters are obtained.

[0027] The first fitness is determined based on the reconciliation cost, the first evaluation parameter, and the second evaluation parameter;

[0028] The candidate harmonic ratios are optimized based on the first fitness, and the second fitness of the optimized candidate harmonic ratios is evaluated to determine whether they have converged.

[0029] If so, the updated candidate harmonic ratio corresponding to the minimum second fitness will be output as the optimal harmonic ratio;

[0030] If not, then iteratively optimize the updated candidate harmonic ratio based on the second fitness until the fitness of the optimized candidate harmonic ratio reaches the cutoff condition, and output the optimal harmonic ratio.

[0031] In one possible implementation, before the mixing device performs the mixing operation according to the optimal mixing ratio, the method further includes:

[0032] The property data and optimal blending ratio of multiple component oils are input into the digital twin model of the blending device to obtain virtual blending results. The property data includes viscosity, component content and density.

[0033] The virtual blending result is compared with the quality conditions, and the optimal blending ratio is adjusted according to the first comparison result, which is used to indicate that the virtual blending result does not meet the quality conditions.

[0034] The digital twin model is built based on the physical parameters of the blending device, including pipeline flow rate and mixing tank volume. The virtual blending result includes the index data of the gasoline to be blended, including density, vapor pressure, octane number and multiple linear indicators.

[0035] In one possible implementation, after the control and blending device performs the blending operation according to the optimal blending ratio, the method further includes:

[0036] The system receives test data for the optimal blending ratio. If the test data does not meet the quality requirements, the gasoline to be blended is designated as a component oil to be reworked. The volume of the gasoline to be blended is updated to the first inventory data of the component oil to be reworked, and the minimum blending cost corresponding to the optimal blending ratio is updated to the first cost data of the component oil to be reworked. The test data includes experimental data for multiple indicators of the gasoline to be blended.

[0037] Obtain the first index data of the component oil to be reworked from the detection equipment in the blending unit;

[0038] Input the blending task parameters, first inventory data, first cost data, and first indicator data into the gasoline blending model to obtain the updated optimal blending ratio;

[0039] The control and blending device performs the blending operation according to the updated optimal blending ratio.

[0040] Secondly, embodiments of this application provide a gasoline blending apparatus based on a large model, comprising:

[0041] The parsing module is used to parse the user's input gasoline blending requirements and obtain blending task parameters, which include: the gasoline grade, quality conditions, and planned output of the gasoline to be blended.

[0042] The acquisition module is used to acquire the inventory levels of multiple component oils;

[0043] The processing module is used to iteratively optimize the blending ratio of each component oil by taking the minimum blending cost as the objective function and the quality conditions, planned output and inventory as constraints through the gasoline blending model corresponding to the gasoline grade, and output the optimal blending ratio of multiple component oils.

[0044] The control module is used to control the blending device to perform blending operations according to the optimal blending ratio.

[0045] In one possible implementation, the gasoline blending model includes: an octane number calculation model, a vapor pressure calculation model, and a blending ratio recommendation model; the device further includes: a generation module.

[0046] The generation module is used to generate multiple candidate harmonic ratios;

[0047] The processing module is also used to output the first octane number corresponding to each candidate blending ratio based on multiple candidate blending ratios and multiple component oils through the octane number calculation model;

[0048] The processing module is also used to output the first vapor pressure corresponding to each candidate blending ratio based on the vapor pressure of multiple candidate blending ratios and multiple component oils through a vapor pressure calculation model;

[0049] The processing module is also used to linearly superimpose the linear indices of each component oil to obtain the first linear index corresponding to each candidate blending ratio.

[0050] The processing module is also used to determine the evaluation parameters of multiple candidate blending ratios by using the blending ratio recommendation model with the minimum blending cost as the objective function and quality conditions, planned output and inventory as constraints. Based on the evaluation parameters, the corresponding candidate blending ratios are iteratively optimized until the cutoff condition is reached, and the optimal blending ratio is output.

[0051] The evaluation parameters are obtained by evaluating the candidate blending ratio and the indicators to be evaluated based on the constraints. The indicators to be evaluated include: first octane number, first vapor pressure and first linearity index.

[0052] In one possible implementation, the above-mentioned octane number calculation model includes an octane number interpolation table, and the above-mentioned device further includes: a determination module;

[0053] The determination module is used to determine the first octane number corresponding to the candidate blending ratio through an octane number interpolation table.

[0054] In one possible implementation, the above-described apparatus further includes: a conversion module;

[0055] The determination module is also used to determine the volume fraction of each component oil based on the candidate blending ratio;

[0056] The conversion module is used to convert the vapor pressure of multiple component oils into a vapor pressure blending index;

[0057] The determination module is also used to determine the first vapor pressure based on the volume fraction and vapor pressure blending index of multiple component oils.

[0058] In one possible implementation, the above-described apparatus further includes: an evaluation module, an output module, and an optimization module;

[0059] The determination module is also used to determine the blending cost of each candidate blending ratio by means of a blending ratio recommendation model with the minimum blending cost as the objective function and with quality conditions, planned output and inventory as constraints.

[0060] The evaluation module is used to evaluate the indicators to be evaluated based on quality conditions to obtain the first evaluation parameter, and to evaluate the candidate blending ratio based on planned output and inventory to obtain the second evaluation parameter.

[0061] The determination module is also used to determine the first fitness based on the reconciliation cost, the first evaluation parameter, and the second evaluation parameter;

[0062] The evaluation module is also used to optimize the corresponding candidate harmonic ratio based on the first fitness, and to evaluate whether the second fitness of the optimized candidate harmonic ratio has converged.

[0063] The output module is used to output the updated candidate harmonic ratio corresponding to the minimum second fitness as the optimal harmonic ratio if the condition is met.

[0064] The optimization module is used to iteratively optimize the updated candidate harmonic ratio based on the second fitness if the condition is not met, until the fitness of the optimized candidate harmonic ratio reaches the cutoff condition, and then output the optimal harmonic ratio.

[0065] In one possible implementation, the above-described apparatus further includes: an input module;

[0066] The input module is used to input the property data and optimal blending ratio of multiple component oils into the digital twin model of the blending device to obtain virtual blending results. The property data includes viscosity, component content and density.

[0067] The processing module is also used to compare the virtual blending result with the quality conditions and adjust the optimal blending ratio according to the first comparison result, which is used to indicate that the virtual blending result does not meet the quality conditions.

[0068] The digital twin model is built based on the physical parameters of the blending device, including pipeline flow rate and mixing tank volume. The virtual blending result includes the index data of the gasoline to be blended, including density, vapor pressure, octane number and multiple linear indicators.

[0069] In one possible implementation, the above-described apparatus further includes: a receiving module and an updating module;

[0070] The receiving module is used to receive the test data of the optimal harmonic ratio;

[0071] The update module is used to, when the inspection data does not meet the quality conditions, designate the gasoline to be blended as a component oil to be reworked, update the volume of the gasoline to be blended to the first inventory data of the component oil to be reworked, and update the minimum blending cost corresponding to the optimal blending ratio to the first cost data of the component oil to be reworked; wherein, the inspection data includes: experimental data of multiple indicators of the gasoline to be blended obtained by experiment.

[0072] The acquisition module is also used to acquire the first index data of the component oil to be reworked collected by the detection equipment in the blending unit;

[0073] The input module is also used to input the blending task parameters, the first inventory data, the first cost data, and the first indicator data into the gasoline blending model to obtain the updated optimal blending ratio;

[0074] The control module is also used to control the blending device to perform blending operations according to the updated optimal blending ratio.

[0075] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0076] The memory stores instructions that the computer executes;

[0077] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0078] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0079] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0080] The gasoline blending method, apparatus, electronic device, and storage medium based on a large model provided in this application embodiment parse the gasoline blending requirements input by the user to obtain blending task parameters. These parameters include: the gasoline grade, quality conditions, and planned output of the gasoline to be blended, and the inventory of multiple component oils. Using the gasoline blending model corresponding to the aforementioned gasoline grade, the method takes the minimum blending cost as the objective function and the quality conditions, planned output, and inventory as constraints to iteratively optimize the blending ratio of each component oil, outputting the optimal blending ratio of multiple component oils. The method then controls the blending apparatus to perform blending operations according to this optimal blending ratio, achieving the technical effect of efficiently and accurately outputting the optimal blending ratio with the minimum blending cost under the constraints of inventory, cost, and quality standards, thereby meeting the user's technical requirements for blending efficiency, blending accuracy, and low-cost blending. Attached Figure Description

[0081] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0082] Figure 1 is a flowchart illustrating the gasoline blending method based on a large model provided in this application.

[0083] Figure 2 is a schematic flowchart of the gasoline blending method based on a large model provided in this application.

[0084] Figure 3 is a flowchart illustrating the gasoline blending method based on a large model provided in this application.

[0085] Figure 4 is a schematic diagram of the structure of the gasoline blending device based on a large model provided in this application;

[0086] Figure 5 is a schematic diagram of the structure of the electronic device provided in this application.

[0087] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0088] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0089] First, let me explain the terms used in this application:

[0090] Research Octane Number (RON): This refers to the gasoline's ability to withstand high pressure and resist unnatural knocking when mixed with isooctane in a specific ratio under standardized experimental conditions. The higher the RON, the better the gasoline's anti-knock performance.

[0091] Motor Octane Number (MON) refers to the ability of the fuel mixture in an engine combustion chamber to resist auto-ignition under high-speed and high-load conditions. MON reflects the anti-knock performance of gasoline under actual engine operating conditions.

[0092] Gasoline blending is a crucial step in the production of finished gasoline in refineries. Its goal is to mix component oils such as catalytic gasoline, reformed gasoline, MTBE, and alkylate oil in specific proportions to obtain blended gasoline that meets key requirements for octane number, vapor pressure, sulfur content, benzene content, and olefin content. Simultaneously, production costs should be minimized during the blending process.

[0093] The blending formula is developed by engineers based on historical experience or laboratory test data. For example, under constraints of inventory, cost, and quality standards, the gasoline blending ratio is adjusted until the optimal blending ratio that meets the constraints and minimizes costs is obtained.

[0094] This method has a long harmonic cycle and requires manual calculation of the optimal solution under complex constraints. It cannot quickly and accurately output the optimal harmonic ratio that meets the constraints, thus failing to meet users' needs for harmonic efficiency, harmonic accuracy, and low-cost harmonics.

[0095] The gasoline blending method based on a large model provided in this application calls the gasoline blending model corresponding to the gasoline grade. The gasoline blending model takes the minimum blending cost as the objective function and quality conditions, planned output, and inventory as constraints. Then, iteratively optimizes the blending ratio of each component oil and outputs the optimal blending ratio of multiple component oils. It can quickly and accurately output the optimal solution under complex constraints, that is, the optimal blending ratio that meets the above constraints and has the lowest cost, thereby meeting the user's needs for blending efficiency, blending accuracy, and low-cost blending.

[0096] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0097] Figure 1 is a flowchart illustrating the gasoline blending method based on a large model provided in this application. As shown in Figure 1, this method can be applied, for example, to a gasoline blending agent. The method includes:

[0098] S101. Parse the user's input gasoline blending requirements to obtain the blending task parameters.

[0099] The blending task parameters include: the octane rating, quality conditions, and planned production volume of the gasoline to be blended.

[0100] After receiving the user's gasoline blending request, semantic analysis is used to extract the key terms in the request. These key terms include, but are not limited to, gasoline octane rating, quality standards, and planned production volume. The system then uses the gasoline octane rating to retrieve the corresponding quality conditions for that gasoline octane rating under a predefined quality standard mapping table. These quality conditions include limits for multiple gasoline indicators. The predefined quality standard mapping table is based on industry standards.

[0101] For example, the gasoline blending agent receives a gasoline blending request as "produce 100 tons of 95# gasoline that meets the China V standard." After identifying and analyzing this request text, the blending task parameters include: 95# octane rating, China V quality standard, and planned output of 100 tons. The China V standard is the fifth stage gasoline quality standard, which imposes strict limits on sulfur content, olefin content, and other parameters.

[0102] Understandably, users input their gasoline blending requirements via voice, text, or other means on the user terminal. For example, if the received gasoline blending requirement is voice data, the voice data is converted into text data, and the text data is analyzed and recognized to obtain the blending task parameters.

[0103] S102. Obtain the inventory of multiple component oils.

[0104] The inventory levels of multiple component oils are retrieved from the material management system. Optionally, the gasoline blending agent determines the multiple component oils corresponding to the gasoline grade in the blending task parameters and retrieves the inventory levels of the multiple component oils from the material management system.

[0105] Optionally, the gasoline blending model is trained based on historical blending data. After obtaining the blending task parameters, the gasoline blending model is automatically triggered to retrieve the inventory levels of multiple component oils from the material management system based on embedded blending material knowledge. The blending material knowledge can, for example, be a set containing multiple component oils.

[0106] Understandably, the inventory levels of multiple component oils are monitored by level sensors installed in the component oil storage tanks, and these inventory levels are uploaded to the material management system.

[0107] S103. Using the gasoline blending model corresponding to the gasoline grade, the minimum blending cost is taken as the objective function, and the quality conditions, planned output and inventory are taken as constraints. The blending ratio of each component oil is iteratively optimized to output the optimal blending ratio of multiple component oils.

[0108] Specifically, the gasoline blending model corresponding to the gasoline grade is invoked. This gasoline blending model includes: an octane number calculation model, a vapor pressure calculation model, a linear index calculation model, and a blending ratio recommendation model.

[0109] The gasoline blending model generates multiple candidate blending ratios based on historically optimal blending proportions. For any candidate blending ratio, an octane number calculation model is used to output the total octane number of the gasoline to be blended, based on the octane numbers of multiple component oils and the candidate blending ratios. Similarly, a vapor pressure calculation model is used to output the total vapor pressure of the gasoline to be blended, based on the vapor pressures of multiple component oils and the candidate blending ratios. Furthermore, a linear index calculation model obtains the linear indices of multiple component oils and performs linear superposition of these indices to obtain the total linear index of the gasoline to be blended. These linear indices include, but are not limited to, the content of aromatics, olefins, benzene, sulfur, and oxygen.

[0110] After obtaining the indicator data corresponding to multiple candidate harmonization ratios, the harmonization ratio recommendation model uses quality conditions, planned output, and inventory as constraints, and minimum harmonization cost as the objective function to iteratively optimize the candidate harmonization ratios until the objective function converges or the number of iterations reaches the maximum number of iterations, and outputs the optimal harmonization ratio.

[0111] This method uses a multi-objective dynamic optimization algorithm to comprehensively balance cost and inventory while meeting quality requirements, thereby minimizing the reconciliation cost.

[0112] S104. Control the blending device to perform blending operations according to the optimal blending ratio.

[0113] The gasoline blending intelligent agent activates the blending device and controls the opening of multiple incoming pipeline valves and other components to pump multiple component oils into the mixing tank according to the optimal blending ratio. Simultaneously, the catalytic gasoline pump is activated to agitate and mix the multiple component oils in the mixing tank. The blending device includes, but is not limited to, a data acquisition component and a mixing tank; the data acquisition component includes, but is not limited to, a flow meter.

[0114] Understandably, the blending system adjusts the valve opening and the operating parameters of the catalytic converter pump according to the calculated optimal blending ratio. Furthermore, during the blending process, the flow rate ratio is dynamically monitored, and the frequency converter or valve opening is adjusted in real time to ensure uniform mixing.

[0115] In this method, the entire reconciliation process is driven by an intelligent agent, achieving fully automated control from requirements analysis to reconciliation execution, ensuring reconciliation accuracy and compliance.

[0116] The gasoline blending method based on a large model provided in this application analyzes the user's input gasoline blending requirements to obtain blending task parameters. These parameters include the gasoline grade, quality conditions, and planned output of the gasoline to be blended. The inventory levels of multiple component oils are also obtained. Using the gasoline blending model corresponding to the aforementioned gasoline grade, the minimum blending cost is taken as the objective function, and the quality conditions, planned output, and inventory levels are taken as constraints. The blending ratio of each component oil is iteratively optimized to output the optimal blending ratio of multiple component oils. The optimal blending ratio is then controlled to be executed according to this optimal ratio. This method can quickly and accurately output the optimal solution under complex constraints, namely, the optimal blending ratio that meets the aforementioned constraints and has the lowest cost, thereby satisfying the user's needs for blending efficiency, blending accuracy, and low-cost blending.

[0117] Figure 2 is a schematic flowchart of the gasoline blending method based on a large model provided in this application. The gasoline blending model includes: an octane number calculation model, a vapor pressure calculation model, and a blending ratio recommendation model. As shown in Figure 2, this embodiment, based on the embodiment in Figure 1, provides a detailed description of a method for determining the optimal blending ratio using a gasoline blending model. This method includes:

[0118] S201, Generate multiple candidate harmonic ratios.

[0119] Multiple candidate harmonic ratios are generated based on the most recent historical harmonic ratio. For example, the most recent historical harmonic ratio is... Then the candidate harmonic ratio = ,in, , It is a value randomly generated within a preset range. The preset range can be, for example, [-0.02, 0.02].

[0120] This method generates multiple candidate harmonic ratios by using historical harmonic ratios, which improves the convergence speed of the model and thus improves the computational efficiency of the optimal harmonic ratio.

[0121] S202. Based on multiple candidate blending ratios and the octane numbers of multiple component oils, the octane number calculation model outputs the first octane number corresponding to each candidate blending ratio.

[0122] The octane rating calculation model is trained based on a historical gasoline blending dataset, which includes multiple historical blending ratios, the measured octane rating of each component oil in each historical blending ratio, and the measured octane rating corresponding to the historical blending ratio.

[0123] Specifically, the candidate blending ratio and the octane number of each component oil in that ratio are input into the octane number calculation model to obtain the octane number corresponding to the candidate blending ratio. The octane number of each component oil is obtained by detection equipment installed in the incoming material inspection pipeline. For example, online infrared spectrometers, gas chromatographs, and other detection equipment are used to detect the octane number, vapor pressure, and content of aromatics, benzene, olefins, sulfur, mercaptans, oxygen, etc., of each component oil.

[0124] For example, the octane number calculation model can be a deep neural network, which models the nonlinear superposition characteristics of octane numbers. Here, the deep neural network refers to a multilayer perceptron (MLP) model, used to learn the nonlinear superposition rules of component oil octane numbers.

[0125] Optionally, the first octane number corresponding to the candidate blending ratio can be determined by using an octane number interpolation table.

[0126] This method optimizes the generated octane number interpolation table using historical blending data to locally linearize nonlinear superposition characteristics. For example, an octane number interpolation table can be generated based on octane number blending data of different batches of catalytic gasoline and MTBE.

[0127] S203. Based on the vapor pressure of multiple candidate blending ratios and multiple component oils, the vapor pressure calculation model is used to output the first vapor pressure corresponding to each candidate blending ratio.

[0128] The vapor pressure calculation model is trained based on multiple historical candidate harmonic ratios and the measured vapor pressure values ​​of each historical candidate harmonic ratio.

[0129] Optionally, the volume fraction of each component oil is determined based on the candidate blending ratio using a vapor pressure calculation model, and the vapor pressures of multiple component oils are converted into a vapor pressure blending index. Then, based on the volume fractions of multiple component oils and the vapor pressure blending index, a first vapor pressure is determined.

[0130] Optionally, the vapor pressure calculation model calculates the vapor pressure blending index of the gasoline to be blended based on the following formula:

[0131]

[0132] in, The total number of component oils, For the first Volume fraction of each component oil For the first Vapor pressure of each component oil.

[0133] After obtaining the vapor pressure blending index, the first vapor pressure is determined based on the mapping relationship between the vapor pressure of the gasoline to be blended and the vapor pressure blending index. This mapping relationship is shown in the following formula:

[0134]

[0135] in, The vapor pressure blending index of the gasoline to be blended. This is the first vapor pressure of the gasoline to be blended.

[0136] S204. Linearly superimpose the linear indices of each component oil to obtain the first linear index corresponding to each candidate blending ratio.

[0137] For linearly superimposed component oil properties (such as sulfur content), a direct weighted calculation is performed. For example, if the sulfur content of catalytic gasoline is 15 ppm (parts per million) and the sulfur content of MTBE is 2 ppm, and the blending ratio is 60%:40%, then the sulfur content after blending is 15 × 0.6 + 2 × 0.4 = 10.6 ppm.

[0138] Steps S202 to S204 above calculate the nonlinear indicators of each component oil using a pre-trained model, and then superimpose the linear indicators to significantly improve the accuracy of blending ratio prediction. For example, using nonlinear interpolation to calculate the first octane number reduces the prediction error of octane number caused by fluctuations in the properties of the component oils, while the linear property calculation model ensures strict compliance with environmental indicators (such as sulfur content). This technique reduces the risk of blending results deviating from expectations and reduces the rework rate caused by prediction errors.

[0139] S205 uses a blending ratio recommendation model with minimum blending cost as the objective function and quality conditions, planned output, and inventory as constraints to determine evaluation parameters for multiple candidate blending ratios. Based on the evaluation parameters, the corresponding candidate blending ratios are iteratively optimized until the cutoff condition is met, and the optimal blending ratio is output.

[0140] The evaluation parameters are obtained by evaluating the candidate blending ratio and the indicators to be evaluated based on the constraints. The indicators to be evaluated include: first octane number, first vapor pressure and first linearity index.

[0141] Specifically, with minimizing the procurement cost of component oils as the core optimization objective, an objective function is constructed. For example, if MTBE is more expensive than catalytic gasoline, the model prioritizes using catalytic gasoline to reduce blending costs. Based on the objective function, a multi-objective optimization algorithm is used to solve for the optimal blending ratio under constraints of quality conditions, planned production volume, and inventory levels.

[0142] This method employs a multi-objective optimization algorithm, setting the minimum blending cost as the objective function, and combining planned production volume, quality conditions (e.g., octane number ≥ 95), and inventory levels (e.g., MTBE inventory ≤ 200 tons) as constraints. A nonlinear programming algorithm is then used to solve for the optimal blending ratio. Through mathematical modeling and iterative calculations, this process ensures that the blending ratio minimizes cost while meeting quality and inventory requirements.

[0143] Optionally, a method for outputting the optimal harmonic ratio based on a multi-objective optimization algorithm is provided here. This method includes: determining the harmonic cost of each candidate harmonic ratio using a harmonic ratio recommendation model with minimum harmonic cost as the objective function and quality conditions, planned output, and inventory as constraints; evaluating the evaluation index based on the quality conditions to obtain a first evaluation parameter; and evaluating the candidate harmonic ratios based on planned output and inventory to obtain a second evaluation parameter; determining a first fitness based on the harmonic cost, the first evaluation parameter, and the second evaluation parameter; optimizing the corresponding candidate harmonic ratio based on the first fitness, and evaluating whether the second fitness of the optimized candidate harmonic ratio has converged; if so, outputting the updated candidate harmonic ratio corresponding to the minimum second fitness as the optimal harmonic ratio; if not, iteratively optimizing the updated candidate harmonic ratio based on the second fitness until the fitness of the optimized candidate harmonic ratio reaches the cutoff condition, and outputting the optimal harmonic ratio.

[0144] The first evaluation parameter is a quality deviation index. If the first octane number, first vapor pressure, and all linearity indicators corresponding to the candidate blending ratio meet the quality requirements, this first evaluation parameter is 0. Otherwise, the first evaluation parameter is determined based on the degree of deviation. The greater the degree of deviation, the larger the first evaluation parameter.

[0145] After obtaining the first evaluation parameter, the second evaluation parameter, and the harmonization cost corresponding to the candidate harmonization ratio, the first evaluation parameter, the second evaluation parameter, and the harmonization cost are weighted and calculated to obtain the first fitness.

[0146] Understandably, the above cutoff conditions include the second fitness convergence, or the number of iterations reaching the preset maximum number of iterations.

[0147] The gasoline blending method based on a large model provided in this application generates multiple candidate blending ratios. Then, using an octane number calculation model, it outputs the first octane number corresponding to each candidate blending ratio based on the octane numbers of the multiple candidate blending ratios and the component oils. Similarly, using a vapor pressure calculation model, it outputs the first vapor pressure corresponding to each candidate blending ratio based on the vapor pressures of the multiple candidate blending ratios and the component oils. Furthermore, it linearly superimposes the linear indices of each component oil to obtain the first linear index corresponding to each candidate blending ratio. Finally, using a blending ratio recommendation model with minimum blending cost as the objective function and quality conditions, planned production volume, and inventory level as constraints, it determines the evaluation parameters for multiple candidate blending ratios. Based on these evaluation parameters, it iteratively optimizes the corresponding candidate blending ratios until a cutoff condition is met, outputting the optimal blending ratio.

[0148] In this method, the octane number calculation model can identify the nonlinear superposition law of the octane numbers of each component oil, thereby improving the prediction accuracy of the first octane number corresponding to the candidate blending ratio and reducing the prediction deviation caused by fluctuations in the properties of the component oils (such as differences in the octane numbers of different batches of catalytic gasoline).

[0149] Furthermore, this method uses a multi-objective optimization algorithm to dynamically solve for the blending ratio corresponding to the lowest blending cost under the constraints of quality, planned output, and inventory. This avoids the limitations of fixed formulas and ensures the stability and compliance of the blending results while reducing blending costs.

[0150] Figure 3 is a flowchart illustrating the gasoline blending method based on a large model provided in this application. As shown in Figure 3, based on the embodiments in Figures 1-2, this embodiment, after obtaining the optimal blending ratio, provides a detailed description of a gasoline blending method based on a large model. This method includes:

[0151] S301. Input the property data of multiple component oils and the optimal blending ratio into the digital twin model of the blending device to obtain the virtual blending result.

[0152] The attribute data includes viscosity, component content, and density. The digital twin model is built based on the physical parameters of the blending device, including pipeline flow rate and mixing tank volume. The virtual blending result includes the index data of the gasoline to be blended, including density, vapor pressure, octane number, and multiple linear indicators.

[0153] By using a digital twin model to simulate the fluid dynamics (such as stratification) during the mixing process according to the optimal blending ratio, a virtual blending result is obtained. This method outputs the virtual blending result through the digital twin model before actual blending, which helps to predict blending risks in advance and reduce trial-and-error costs during actual blending.

[0154] Understandably, the aforementioned digital twin model is calibrated based on multiple historical blending ratios and multiple blended gasoline indices corresponding to each historical blending ratio. These blended gasoline indices can include, for example, the measured octane number, measured vapor pressure, and the content of components such as aromatics, benzene, olefins, sulfur, mercaptans (RSH), and oxygen in the blended gasoline.

[0155] S302. Compare the virtual blending result with the quality conditions, and adjust the optimal blending ratio based on the first comparison result.

[0156] The first comparison result is used to indicate that the virtual reconciliation result does not meet the quality requirements.

[0157] The system determines whether the virtual blending result meets the quality conditions. If yes, it controls the blending device to perform the blending operation according to the optimal blending ratio. If not, it optimizes the parameters of the gasoline blending model based on the first comparison result, and inputs the quality conditions, planned output, and inventory into the optimized gasoline blending model to obtain the optimized optimal blending ratio. Further, it determines whether the optimized optimal blending ratio meets the quality conditions. If not, it further optimizes the gasoline blending model until the optimized optimal blending ratio meets the quality conditions, or until the preset maximum number of optimizations is reached. Finally, it outputs the optimized optimal blending ratio and controls the blending device to perform the blending operation according to this optimized optimal blending ratio.

[0158] S303, The control and blending device performs the blending operation according to the optimal blending ratio.

[0159] This step is similar to the explanation of step S103 above, and will not be repeated here.

[0160] S304. Receive the inspection data of the optimal blending ratio, and if the inspection data does not meet the quality conditions, treat the gasoline to be blended as a component oil to be reworked, update the volume of the gasoline to be blended to the first inventory data of the component oil to be reworked, and update the minimum blending cost corresponding to the optimal blending ratio to the first cost data of the component oil to be reworked.

[0161] The test data includes experimental data on multiple indicators of the gasoline to be blended. These multiple indicators include, but are not limited to, the octane number, vapor pressure, and the content of aromatics, benzene, olefins, sulfur, mercaptans, and oxygen in the gasoline to be blended.

[0162] Determine whether the experimental data for the above-mentioned indicators meet the standard ranges specified in the quality conditions. If so, generate a gasoline blending report to complete the blending task. The gasoline blending report includes, but is not limited to, the finished product batch number, the values ​​of each indicator, and a detailed cost breakdown.

[0163] If the experimental data for any indicator does not meet the corresponding standard range in the quality conditions, the gasoline to be blended will be treated as a component oil to be reworked, and the optimal blending ratio will be recalculated.

[0164] Specifically, a rework material code is established for each component oil to be reworked, to distinguish it from the original component oil. The volume and / or mass of the component oil to be reworked is detected by the testing equipment in the blending unit, and this volume and / or mass is used as the initial inventory data of the component oil to be reworked and entered into the raw material inventory management system. In addition, the minimum blending cost corresponding to the optimal blending ratio is used as the initial cost data of the component oil to be reworked and entered into the raw material inventory management system.

[0165] S305. Obtain the first index data of the component oil to be reworked from the detection equipment in the blending unit.

[0166] The first indicator data includes the measured octane number and measured vapor pressure of the component oil to be reworked, as well as the content of components such as aromatics, benzene, olefins, sulfur, mercaptans, and oxygen in the component oil to be reworked.

[0167] S306. Input the blending task parameters, first inventory data, first cost data, and first indicator data into the gasoline blending model to obtain the updated optimal blending ratio.

[0168] In this step, the method for generating the updated optimal blending ratio using the gasoline blending model is similar to the method for determining the optimal blending ratio using the gasoline blending model shown in Figure 2 above, and will not be repeated here.

[0169] S307. The control and blending device performs the blending operation according to the updated optimal blending ratio.

[0170] This step is similar to the explanation of step S103 above, and will not be repeated here.

[0171] The gasoline blending method based on a large model provided in this application introduces a virtual blending result by inputting the attribute data of multiple component oils and the optimal blending ratio into the digital twin model of the blending device. The virtual blending result is then compared with the quality conditions, and the optimal blending ratio is adjusted according to the first comparison result. This method anticipates blending risks in advance and reduces resource waste caused by trial and error.

[0172] After obtaining the adjusted optimal blending ratio, the blending device is controlled to perform the blending operation according to the optimal blending ratio, and the inspection data of the optimal blending ratio is received. If the inspection data does not meet the quality conditions, the gasoline to be blended is designated as a component oil to be reworked, the volume of the gasoline to be blended is updated to the first inventory data of the component oil to be reworked, and the minimum blending cost corresponding to the optimal blending ratio is updated to the first cost data of the component oil to be reworked. Subsequently, the first index data of the component oil to be reworked is obtained from the detection equipment in the blending device, and the blending device is controlled to perform the blending operation according to the updated optimal blending ratio.

[0173] This method automatically triggers a rework process when the inspection data of the gasoline to be blended does not meet the quality requirements. The gasoline to be blended in the blending unit is treated as a component oil to be reworked, and the optimal blending ratio is recalculated. This realizes a closed-loop mechanism of blending-detection-feedback, ensuring that the finished gasoline meets the quality requirements while minimizing blending costs.

[0174] Figure 4 is a schematic diagram of the gasoline blending device based on a large model provided in this application. As shown in Figure 4, the gasoline blending device 40 based on a large model provided in this embodiment includes:

[0175] The parsing module 401 is used to parse the gasoline blending requirements input by the user and obtain the blending task parameters, which include: the gasoline grade, quality conditions and planned output of the gasoline to be blended.

[0176] The acquisition module 402 is used to acquire the inventory of multiple component oils;

[0177] The processing module 403 is used to iteratively optimize the blending ratio of each component oil by taking the minimum blending cost as the objective function and the quality conditions, planned output and inventory as constraints through the gasoline blending model corresponding to the gasoline grade, and output the optimal blending ratio of multiple component oils.

[0178] The control module 404 is used to control the blending device to perform blending operations according to the optimal blending ratio.

[0179] In one possible implementation, the gasoline blending model includes: an octane number calculation model, a vapor pressure calculation model, and a blending ratio recommendation model; the device further includes: a generation module 405.

[0180] Generation module 405 is used to generate multiple candidate harmonic ratios;

[0181] The processing module 403 is also used to output the first octane number corresponding to each candidate blending ratio based on multiple candidate blending ratios and multiple component oils through the octane number calculation model;

[0182] The processing module 403 is also used to output the first vapor pressure corresponding to each candidate blending ratio based on the vapor pressure of multiple candidate blending ratios and multiple component oils through a vapor pressure calculation model;

[0183] The processing module 403 is also used to linearly superimpose the linear indices of each component oil to obtain the first linear index corresponding to each candidate blending ratio.

[0184] The processing module 403 is also used to determine the evaluation parameters of multiple candidate blending ratios by using the blending ratio recommendation model with the minimum blending cost as the objective function and quality conditions, planned output and inventory as constraints, and to iteratively optimize the corresponding candidate blending ratios based on the evaluation parameters until the cutoff condition is reached, and output the optimal blending ratio.

[0185] The evaluation parameters are obtained by evaluating the candidate blending ratio and the indicators to be evaluated based on the constraints. The indicators to be evaluated include: first octane number, first vapor pressure and first linearity index.

[0186] In one possible implementation, the above-mentioned octane number calculation model includes an octane number interpolation table, and the above-mentioned device further includes: a determination module 406;

[0187] The determination module 406 is used to determine the first octane number corresponding to the candidate blending ratio through an octane number interpolation table.

[0188] In one possible implementation, the above-described device further includes: a conversion module 407;

[0189] The determination module 406 is also used to determine the volume fraction of each component oil based on the candidate blending ratio;

[0190] The conversion module 407 is used to convert the vapor pressure of multiple component oils into a vapor pressure blending index;

[0191] The determination module 406 is also used to determine a first vapor pressure based on the volume fraction and vapor pressure blending index of multiple component oils.

[0192] In one possible implementation, the above-described apparatus further includes: an evaluation module 408, an output module 409, and an optimization module 410;

[0193] The determination module 406 is also used to determine the blending cost of each candidate blending ratio by means of the minimum blending cost as the objective function and the quality conditions, planned output and inventory as constraints via the blending ratio recommendation model.

[0194] Evaluation module 408 is used to evaluate the indicator to be evaluated based on quality conditions to obtain the first evaluation parameter, and to evaluate the candidate blending ratio based on planned output and inventory to obtain the second evaluation parameter.

[0195] The determination module 406 is also used to determine the first fitness based on the reconciliation cost, the first evaluation parameter, and the second evaluation parameter;

[0196] The evaluation module 408 is also used to optimize the corresponding candidate harmonic ratio based on the first fitness, and to evaluate whether the second fitness of the optimized candidate harmonic ratio has converged.

[0197] Output module 409 is used to output the updated candidate harmonic ratio corresponding to the minimum second fitness as the optimal harmonic ratio if the condition is met.

[0198] The optimization module 410 is used to iteratively optimize the updated candidate harmonic ratio based on the second fitness if not, until the fitness of the optimized candidate harmonic ratio reaches the cutoff condition, and output the optimal harmonic ratio.

[0199] In one possible implementation, the above-described device further includes: an input module 411;

[0200] Input module 411 is used to input the property data and optimal blending ratio of multiple component oils into the digital twin model of the blending device to obtain virtual blending results. The property data includes viscosity, component content and density.

[0201] The processing module 403 is further configured to compare the virtual blending result with the quality conditions, and adjust the optimal blending ratio according to the first comparison result, wherein the first comparison result is used to indicate that the virtual blending result does not meet the quality conditions.

[0202] The digital twin model is built based on the physical parameters of the blending device, including pipeline flow rate and mixing tank volume. The virtual blending result includes the index data of the gasoline to be blended, including density, vapor pressure, octane number and multiple linear indicators.

[0203] In one possible implementation, the above-described apparatus further includes: a receiving module 412 and an updating module 413;

[0204] The receiving module 412 is used to receive the test data of the optimal harmonic ratio;

[0205] The update module 413 is used to update the gasoline to be blended as a component oil to be reworked when the inspection data does not meet the quality conditions, update the volume of the gasoline to be blended to the first inventory data of the component oil to be reworked, and update the minimum blending cost corresponding to the optimal blending ratio to the first cost data of the component oil to be reworked; wherein, the inspection data includes: experimental data of multiple indicators of the gasoline to be blended obtained by experiment.

[0206] The acquisition module 402 is also used to acquire the first index data of the component oil to be reworked collected by the detection equipment in the blending device;

[0207] The input module 411 is also used to input the blending task parameters, the first inventory data, the first cost data and the first index data into the gasoline blending model to obtain the updated optimal blending ratio;

[0208] The control module 404 is also used to control the blending device to perform blending operations according to the updated optimal blending ratio.

[0209] The gasoline blending device based on a large model provided in this embodiment can execute the method provided in the above-described method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0210] Figure 5 is a schematic diagram of the structure of the electronic device provided in this application. As shown in Figure 5, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0211] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0212] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0213] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0214] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0215] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0216] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0217] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0218] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0219] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0220] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0221] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0222] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0223] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0224] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0225] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A gasoline blending method based on a large model, characterized in that, include: The system parses the user's input gasoline blending requirements to obtain blending task parameters, including: the gasoline grade, quality conditions, and planned output of the gasoline to be blended; obtains the inventory of multiple component oils; uses the gasoline blending model corresponding to the gasoline grade as the objective function, with the minimum blending cost as the objective function and the quality conditions, planned output, and inventory as constraints, to iteratively optimize the blending ratio of each component oil, outputting the optimal blending ratio of the multiple component oils; and controls the blending device to perform the blending operation according to the optimal blending ratio.

2. The method according to claim 1, characterized in that, The gasoline blending model includes: an octane number calculation model, a vapor pressure calculation model, and a blending ratio recommendation model. The gasoline blending model corresponding to the gasoline grade uses the minimum blending cost as the objective function and the quality conditions, planned production volume, and inventory level as constraints to iteratively optimize the blending ratio of each component oil, outputting the optimal blending ratio of the multiple component oils. This includes: generating multiple candidate blending ratios; using the octane number calculation model based on the multiple candidate blending ratios and the octane numbers of the multiple component oils, outputting the first octane number corresponding to each candidate blending ratio; and using the vapor pressure calculation model based on the multiple candidate blending ratios and the vapor pressure of the multiple component oils... Gas pressure, outputting the first vapor pressure corresponding to each candidate blending ratio; linearly superimposing the linear indices of each component oil to obtain the first linear index corresponding to each candidate blending ratio; using the blending ratio recommendation model with minimum blending cost as the objective function and quality conditions, planned production, and inventory as constraints, determining the evaluation parameters of the multiple candidate blending ratios, and iteratively optimizing the corresponding candidate blending ratios based on the evaluation parameters until the cutoff condition is reached, outputting the optimal blending ratio; wherein, the evaluation parameters are obtained by evaluating the candidate blending ratios and the indicators to be evaluated based on the constraints, and the indicators to be evaluated include: first octane number, first vapor pressure, and first linear index.

3. The method according to claim 2, characterized in that, The octane number calculation model includes an octane number interpolation table. The step of outputting the first octane number corresponding to each candidate blending ratio based on the multiple candidate blending ratios and the octane numbers of the multiple component oils through the octane number calculation model includes: determining the first octane number corresponding to the candidate blending ratio through the octane number interpolation table.

4. The method according to claim 2, characterized in that, The step of outputting a first vapor pressure corresponding to each candidate blending ratio based on the vapor pressure calculation model and the vapor pressures of the multiple candidate blending ratios and the multiple component oils includes: determining the volume fraction of each component oil based on the candidate blending ratios; converting the vapor pressures of the multiple component oils into a vapor pressure blending index; and determining the first vapor pressure corresponding to the candidate blending ratio based on the volume fractions of the multiple component oils and the vapor pressure blending index.

5. The method according to claim 2, characterized in that, The process of determining evaluation parameters for multiple candidate blending ratios using the blending ratio recommendation model with minimum blending cost as the objective function and quality conditions, planned output, and inventory as constraints, and iteratively optimizing the corresponding candidate blending ratios based on the evaluation parameters until a cutoff condition is met, and outputting the optimal blending ratio, includes: determining the blending cost for each candidate blending ratio using the blending ratio recommendation model with minimum blending cost as the objective function and quality conditions, planned output, and inventory as constraints; evaluating the indicators to be evaluated based on the quality conditions to obtain a first evaluation parameter; and, based on the planned output... The quantity and the inventory quantity are used to evaluate the candidate harmonization ratio to obtain a second evaluation parameter; based on the harmonization cost, the first evaluation parameter, and the second evaluation parameter, a first fitness is determined; the corresponding candidate harmonization ratio is optimized based on the first fitness, and the second fitness of the optimized candidate harmonization ratio is evaluated to see if it converges; if so, the updated candidate harmonization ratio corresponding to the minimum second fitness is output as the optimal harmonization ratio; if not, the updated candidate harmonization ratio is iteratively optimized based on the second fitness until the fitness of the optimized candidate harmonization ratio reaches the cutoff condition, and the optimal harmonization ratio is output.

6. The method according to any one of claims 1-5, characterized in that, Before the blending device performs the blending operation according to the optimal blending ratio, the method further includes: inputting the attribute data of the multiple component oils and the optimal blending ratio into the digital twin model of the blending device to obtain a virtual blending result, wherein the attribute data includes viscosity, component content, and density; comparing the virtual blending result with the quality conditions, and adjusting the optimal blending ratio according to a first comparison result, wherein the first comparison result is used to indicate that the virtual blending result does not meet the quality conditions; wherein the digital twin model is constructed based on the physical parameters of the blending device, wherein the physical parameters include: pipeline flow rate and mixing tank volume, and the virtual blending result includes the index data of the gasoline to be blended, wherein the index data includes: density, vapor pressure, octane number, and multiple linear indices.

7. The method according to claim 1 or 6, characterized in that, After the blending device performs a blending operation according to the optimal blending ratio, the method further includes: receiving inspection data of the optimal blending ratio, and if the inspection data does not meet the quality conditions, designating the gasoline to be blended as a component oil to be reworked, updating the volume of the gasoline to be blended to the first inventory data of the component oil to be reworked, and updating the minimum blending cost corresponding to the optimal blending ratio to the first cost data of the component oil to be reworked; wherein the inspection data includes: experimental data of multiple indicators of the gasoline to be blended; acquiring the first indicator data of the component oil to be reworked collected by the detection equipment in the blending device; inputting the blending task parameters, the first inventory data, the first cost data, and the first indicator data into the gasoline blending model to obtain the updated optimal blending ratio; and controlling the blending device to perform a blending operation according to the updated optimal blending ratio.

8. A gasoline blending device based on a large model, characterized in that, include: The system includes a parsing module for parsing user-inputted gasoline blending requirements and obtaining blending task parameters, including the gasoline grade, quality conditions, and planned output of the gasoline to be blended; an acquisition module for acquiring the inventory levels of multiple component oils; a processing module for iteratively optimizing the blending ratio of each component oil using the gasoline blending model corresponding to the gasoline grade, with the minimum blending cost as the objective function and the quality conditions, planned output, and inventory levels as constraints, and outputting the optimal blending ratio of the multiple component oils; and a control module for controlling the blending device to perform the blending operation according to the optimal blending ratio.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images