Tobacco flavor processing parameter determination method and device, and electronic equipment

Abstract

This invention discloses a method, apparatus, and electronic device for determining tobacco flavoring processing parameters. Based on the original flavoring compound information, at least one flavoring to be tested is identified, along with raw material characteristic information and original process parameters. The raw material characteristic information and original process parameters are processed using a first model to determine the extract quality prediction result. The original flavoring compound information and the extract quality prediction result are input into a second model to obtain the predicted quality information of the compounded product. When the predicted quality information of the compounded product does not meet preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model. The objective function is solved using a preset optimization algorithm to obtain the target flavoring compound information and target process parameters while satisfying at least one compounding constraint condition. This achieves the adjustment and optimization of the original flavoring compound information and original process parameters, ensuring the quality of the compounded product.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, and electronic device for determining tobacco flavoring processing parameters. Background Technology

[0002] In tobacco production and processing, flavorings are typically extracted to obtain flavoring extracts. Based on multiple flavoring extracts and the production of compound formulations, a final compound flavoring product is determined for addition to the tobacco. When the quality of the compound flavoring product added to the tobacco differs, the quality of the subsequently processed cigarettes will also vary.

[0003] To ensure the quality stability of finished cigarettes, adjustments are usually made to the flavor extraction process and the compounding process of flavor extracts. However, existing technologies are mainly based on trial and error adjustments using human experience, which not only has high labor costs and low adjustment efficiency, but also cannot guarantee the quality of the adjusted finished product. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for determining tobacco flavoring processing parameters, which realizes the adjustment and optimization of original flavoring compounding information and original process parameters, improves the efficiency and accuracy of adjustment and optimization, and ensures the quality of compounded finished products.

[0005] According to one aspect of the present invention, a method for determining tobacco flavoring processing parameters is provided, the method comprising: Based on predetermined original flavor compound information, at least one flavor to be tested associated with the original flavor compound information is identified, and raw material characteristic information and original process parameters of the flavor to be tested are determined; wherein, the original process parameters are used to extract the extract from the flavor to be tested. For at least one of the flavorings to be detected, the raw material characteristic information and the original process parameters are processed based on the first model to determine the extract quality prediction result corresponding to the flavoring to be detected; The original spice blend information and the quality prediction results of the extract of at least one spice to be tested are input into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information. When it is determined that the predicted quality information of the compound finished product does not meet the preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model; The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and the original process parameters under at least one blending constraint condition, so as to obtain the target spice blending information and the target process parameters.

[0006] According to another aspect of the present invention, a device for determining tobacco flavoring processing parameters is provided, the device comprising: The raw data acquisition module is used to determine at least one spice to be detected associated with the pre-determined raw spice blend information, and to determine the raw material characteristic information and raw process parameters of the spice to be detected; wherein, the raw process parameters are used to extract the extract from the spice to be detected. An extract quality prediction module is used to process the raw material characteristic information and the original process parameters based on a first model for at least one of the flavorings to be tested, and determine the extract quality prediction result corresponding to the flavoring to be tested. The finished product quality prediction module is used to input the original spice blend information and the quality prediction results of the extract of at least one spice to be tested into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information. The objective function determination module is used to determine an objective function based on the raw material characteristic information, the first model, and the second model when it is determined that the predicted quality information of the compound finished product does not meet the preset conditions. The parameter adjustment module is used to solve the objective function according to a preset optimization algorithm, so as to adjust the original spice blend information and the original process parameters under the condition of satisfying at least one blending constraint, so as to obtain the target spice blend information and the target process parameters.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the tobacco flavoring processing parameter determination method according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the method for determining tobacco flavoring processing parameters according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method for determining tobacco flavoring processing parameters as described in any embodiment of the present invention.

[0010] The technical solution of this invention involves determining at least one spice to be tested associated with the pre-determined original spice blending information, and determining the raw material characteristic information and original process parameters corresponding to each spice to be tested. For at least one spice to be tested, the raw material characteristic information and original process parameters are processed based on a first model to determine the predicted quality result of the extract corresponding to the spice to be tested. The original spice blending information and the extract quality of at least one spice to be tested are input into a second model to obtain the predicted quality information of the blended finished product. When it is determined that the predicted quality information of the blended finished product does not meet the preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model. The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and original process parameters while satisfying at least one blending constraint condition, thereby obtaining the target spice blending information and target process parameters. This invention addresses the problems of low efficiency, high labor costs, and inability to guarantee the quality of the final product caused by trial-and-error adjustments based on human experience in existing technologies. By predicting the quality of flavor extracts and the quality of compounded products, it assesses whether adjustments to the original flavor compounding information and original process parameters are necessary. When adjustments are required, the above information is quantitatively adjusted based on the objective function and compounding constraints, avoiding the low efficiency caused by repeated manual trial-and-error adjustments, improving the efficiency and accuracy of adjustment and optimization, and ensuring the quality of the compounded product.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a method for determining tobacco flavoring processing parameters provided in an embodiment of the present invention; Figure 2 This is a flowchart of a method for determining tobacco flavoring processing parameters provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a tobacco flavoring processing parameter determination device provided in an embodiment of the present invention; Figure 4This is a schematic diagram of the structure of an electronic device for implementing the method for determining tobacco flavoring processing parameters according to an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0017] Before introducing the technical solutions provided by the embodiments of the present invention, the application scenarios can be described first. The technical solutions provided by the embodiments of the present invention can be applied to any scenario where the processing parameters of tobacco flavorings need to be adjusted. For example, in the cigarette production process, tobacco is usually treated with flavorings to achieve the purpose of sweetening, enhancing aroma, and modifying smoke. The added flavorings are usually compound products composed of one or more flavoring extracts. Flavoring extracts are obtained by processing and extracting the original flavorings. For example, flavoring extracts can be one or more extracts of licorice, tobacco extract, tree moss, chicory, etc.

[0018] To ensure the stability of the aroma in the produced cigarettes, it is necessary to ensure the quality stability of the added compound products. That is, the stability of the compound products can be ensured by adjusting the process parameters used to extract the original flavorings and the compound formula information used to compose the compound products.

[0019] Example 1 Figure 1 This is a flowchart of a method for determining tobacco flavoring processing parameters according to Embodiment 1 of the present invention. This embodiment is applicable to predicting the quality of tobacco flavoring extracts and the quality of compounded finished products, so as to adjust the original flavoring compounding information and original process parameters, thereby ensuring the quality of the compounded finished products. This method can be executed by a tobacco flavoring processing parameter determining device, which can be implemented in hardware and / or software, and can be configured in electronic devices such as mobile phones, computers, or servers. Figure 1 As shown, the method includes: S110. Based on the predetermined original flavor compound information, determine at least one flavor to be tested that is associated with the original flavor compound information, and determine the raw material characteristic information and original process parameters of the flavor to be tested.

[0020] The original flavor blend information can be pre-acquired and used to combine flavor extracts to obtain the finished blend. Optionally, the original flavor blend information may include: the flavor to be tested corresponding to at least one flavor extract, the flavor dosage information and extract dosage information for each flavor to be tested.

[0021] The flavoring agent to be tested can be understood as the raw flavoring agent used to obtain the flavoring extract. For example, the flavoring agent to be tested can be licorice raw material. By extracting licorice raw material, glycyrrhizic acid can be obtained. Glycyrrhizic acid is the flavoring extract of the flavoring agent to be tested.

[0022] It should be noted that the quality of the spice extract may vary depending on the origin, climate, harvest time, and storage conditions of the same spice being tested. To accurately assess the quality of the spice extract, the raw material characteristics of the spice to be tested can be determined. These raw material characteristics are used to characterize the chemical composition and content of the spice being tested.

[0023] Correspondingly, the quality and content of the extracted fragrance will vary depending on the process parameters applied during the extraction process. Therefore, to accurately assess the quality of the fragrance extract, the original process parameters of the fragrance to be tested can be determined. These original process parameters are used to extract the extract from the fragrance to be tested. Optionally, the original process parameters may include: extraction method (e.g., solvent extraction, supercritical extraction, subcritical extraction, etc.), extraction temperature, extraction time, type of solvent used, and solid-liquid ratio.

[0024] Specifically, upon receiving a request for processing a blended spice product, the system can obtain the original spice blending information that matches the processing request. Based on the original spice blending information, at least one spice extract that makes up the blended spice product and the corresponding spice to be tested for each spice extract are identified.

[0025] For at least one flavoring agent to be tested, feature information acquisition processing is performed on each flavoring agent to obtain the raw material feature information corresponding to each flavoring agent. Additionally, the original process parameters corresponding to each flavoring agent are retrieved.

[0026] Optionally, the raw material characteristic information and original process parameters of the spice to be tested can be determined in the following way: perform detection processing on the spice to be tested in at least one detection dimension to obtain the raw material characteristic information; retrieve the original process parameters corresponding to the spice to be tested.

[0027] The detection dimension includes at least one of the following: near-infrared spectroscopy (NIR) detection, chromatographic fingerprinting detection, chemical composition detection, and raw material traceability detection. The NIR detection dimension is used to detect the near-infrared spectral information of the spice being tested. The chromatographic fingerprinting detection dimension is used to detect the chromatographic fingerprint of the spice being tested. Optionally, high-performance liquid chromatography (HPLC) is used to obtain a characteristic chromatogram representing the overall chemical composition of the spice being tested, i.e., the chromatographic fingerprint. The chemical composition detection dimension is used to detect conventional chemical components in the spice being tested, such as the moisture content, total sugar content, and total nitrogen content, etc., which are characteristic information of the raw materials. The raw material traceability detection dimension is used to determine the origin, harvest time, and storage conditions of the spice being tested, among other characteristic information.

[0028] The near-infrared spectral information, chromatographic fingerprint, moisture content, total sugar content, total nitrogen content, origin information, harvest time information, and storage conditions of the spices to be tested are used as the raw material characteristic information of the spices to be tested.

[0029] The original process parameters include at least the following: the original extraction time, original extraction temperature, original feed solution information, and original extraction pressure information for the fragrance to be tested. The original extraction time can be understood as the extraction duration during the extraction process of the fragrance to be tested. The original extraction temperature can be the extraction temperature used to extract the fragrance to be tested. The original feed solution information can include information such as the type of solvent, solvent concentration, and solvent-to-feed ratio used in the extraction process of the fragrance to be tested. The original extraction pressure can be understood as the extraction pressure used in the extraction process of the fragrance to be tested. For example, the extraction pressure set when using supercritical fluid extraction to extract the fragrance to be tested. It should be noted that the original process parameters can be obtained from the database corresponding to the production control system.

[0030] Specifically, the fragrance to be tested undergoes detection processing at least one of the following dimensions: near-infrared spectroscopy, chromatographic fingerprinting, chemical composition, and raw material traceability. The detection results for each dimension are then determined. The detection results from at least one dimension are used as the raw material characteristic information of the fragrance to be tested. The original process parameters corresponding to each raw material to be tested are retrieved from the database of the production control system.

[0031] S120. For at least one fragrance to be tested, the raw material characteristic information and original process parameters are processed based on the first model to determine the quality prediction result of the extract corresponding to the fragrance to be tested.

[0032] The first model can be a model used to predict the quality of the spice extract to be tested. Optionally, the first model can be a pre-trained machine learning model. Optionally, the first model can be a multi-output regression model built based on at least one sub-model such as random forest, gradient boosting tree, neural network, and Gaussian process regression.

[0033] The input to the first model is the raw material characteristics and original process parameters of each spice to be tested, and the output of the first model is the quality prediction result of the extract of the spice to be tested.

[0034] The extract quality prediction results may include extract quality prediction sub-results under at least one quality dimension. Optionally, the at least one quality dimension includes one or more of the following: aroma component content dimension, sensory evaluation dimension, fragrance yield dimension, and fragrance color value dimension.

[0035] Among them, the sub-result of extract quality prediction under the aroma component content dimension can be the aroma component content of the extract of the fragrance to be tested. For example, aroma components can be glycyrrhizic acid, thymol, etc.

[0036] Sensory evaluation dimensions may include at least one sensory evaluation sub-dimensional. Optionally, the at least one sensory evaluation sub-dimensional includes one or more of aroma intensity, sweetness, and impurities. The extract quality prediction sub-result under the sensory evaluation dimension may be the sensory evaluation attribute of the flavor extract of the tested flavoring. Optionally, the sensory evaluation attribute may be characterized by a comprehensive sensory evaluation score. For example, based on the evaluation scores of the flavor extract of the tested flavoring in sensory evaluation dimensions such as aroma intensity, sweetness, and impurities, the comprehensive sensory evaluation score of the flavor extract under the sensory evaluation dimension can be determined.

[0037] The extract quality prediction sub-results under the spice yield dimension can be used to characterize the proportion of spice extract mass to the mass of the spice being tested. The extract quality prediction sub-results under the spice color value dimension can be used to characterize information such as the color, hue, and clarity of the spice extract of the spice being tested.

[0038] Specifically, for at least one spice to be tested, the raw material characteristic information and original process parameters of each spice to be tested are processed based on the first model to obtain the extract quality prediction sub-result for each spice to be tested under at least one detection dimension. Based on the extract quality prediction sub-result for each spice to be tested under at least one detection dimension, the extract quality prediction result corresponding to each spice to be tested is determined.

[0039] It should be noted that before processing the raw material characteristic information and original process parameters based on the first model, a trained first model can be obtained based on training samples. Each training sample can include: sample raw material characteristic information, sample process parameter information, and theoretical extract quality results.

[0040] Specifically, the sample raw material characteristics, process parameters, and theoretical extract quality results for different batches of sample spices are obtained from a database corresponding to historical production data to generate multiple training samples. It should be noted that, to improve the generalization ability of the first model, the acquired training samples cover as much as possible the range of raw material variations and process parameter combinations of the sample spices.

[0041] The acquired training samples are divided into training and testing sets according to a preset ratio (e.g., a training set to test set ratio of 4:1). The first model to be trained is then trained using the training samples in the training set to obtain the output extraction quality result corresponding to each training sample. Based on the output extraction quality result and the theoretical extraction quality result, a loss value is determined, and the model parameters of the first model to be trained are adjusted based on the loss value to obtain the trained first model. The trained first model is then tested using the test set to obtain the well-trained first model.

[0042] When testing the first trained model on a test set, the model can be evaluated using at least one of the following model evaluation metrics. The at least one model evaluation metric includes: the coefficient of determination (R²). 2 The evaluation index is one or more of the following: root mean square error (RMSE) evaluation index and mean absolute percentage error (MAPE) evaluation index.

[0043] S130. Input the original spice blend information and the quality prediction results of the extract of at least one spice to be tested into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information.

[0044] The second model can be a pre-trained model used to predict the quality of the compound product. Optionally, the second model can be a mathematical model or a machine learning model. The second model establishes a quality mapping relationship between the flavor extract of the flavoring to be tested and the compound product. The compound product can be understood as a blend of flavorings composed of one or more flavor extracts. The predicted quality information of the compound product is used to characterize its quality.

[0045] Specifically, the original spice blend information and the quality prediction results of the extract of at least one spice to be tested are input into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information.

[0046] S140. When it is determined that the predicted quality information of the compound finished product does not meet the preset conditions, the objective function is determined based on the raw material characteristic information, the first model and the second model.

[0047] The preset conditions can be pre-set conditions that the predicted quality information of the compounded product must meet. The objective function can be a function used to optimize the original spice compounding information and original process parameters. Optionally, a first function can be determined based on a first model and a second model to input the data. The raw material characteristic information is substituted into the first function to obtain the objective function after inputting the data.

[0048] Specifically, if it is determined that the predicted quality information of the compound finished product does not meet the preset conditions, then based on the first model and the second model, a first function is determined, and the raw material characteristic information is substituted into the first function to obtain the target function.

[0049] It should be noted that by determining whether the predicted quality information of the compound finished product meets the preset conditions before determining the objective function and adjusting the corresponding parameters, invalid or unnecessary adjustments can be effectively avoided.

[0050] S150. The objective function is solved according to the preset optimization algorithm to adjust the original spice blending information and original process parameters under the condition of satisfying at least one blending constraint, so as to obtain the target spice blending information and target process parameters.

[0051] The preset optimization algorithm can be a pre-determined optimization algorithm. Optionally, the preset optimization algorithm can be a multi-objective optimization algorithm (such as NSGA-II, MOEA / D) or an algorithm corresponding to a hierarchical optimization strategy. The compounding constraints can be the constraints that need to be satisfied when adjusting the original flavor compounding information and the original process parameters.

[0052] The target spice blend information can be the spice blend information determined after adjusting the original spice blend information. The target process parameters can be the process parameters determined after adjusting the original process parameters.

[0053] Specifically, the objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and original process parameters while satisfying at least one blending constraint condition, so as to obtain the target spice blending information and target process parameters.

[0054] Optionally, the method further includes: sending the target flavoring compound information and target process parameters to the target terminal to obtain the target compound finished product based on the target flavoring compound information and target process parameters.

[0055] The target terminal can be a terminal device used to send target flavoring compound information and target process parameters to the corresponding workshop. The target compound finished product can be a compound flavoring product obtained by processing at least one flavoring to be tested based on the target process parameters and target flavoring compound information.

[0056] Specifically, the target spice blend information and target process parameters are sent to the target terminal, which then distributes the target process parameters to the extraction workshop. In the extraction workshop, at least one spice to be tested is extracted based on the target process parameters to obtain the corresponding spice extract. Additionally, the target terminal distributes the target spice blend information to the blending workshop, where at least one spice extract is blended based on the target spice blend information to obtain the target blended finished product.

[0057] Optionally, during the process of obtaining the target compound product based on the target spice compound information and the target process parameters, feedback information corresponding to the target compound product is obtained, and the model parameters of the first model and the second model are adjusted according to the feedback information.

[0058] The feedback information includes at least: actual extract quality results and actual compound product quality information. The actual extract quality results correspond to the aforementioned extract quality prediction results. The actual extract quality results characterize the actual quality of the flavor extracts extracted and processed in the extraction workshop. The actual compound product quality information corresponds to the compound product prediction quality information. The actual compound product quality information characterizes the actual quality of the target compound product produced in the blending workshop.

[0059] Specifically, in the process of obtaining the target compound product based on the target spice blend information and target process parameters, the spice extracts extracted in the extraction workshop are subjected to quality testing to determine the actual extract quality results. Similarly, the target compound product produced in the blending workshop is subjected to quality testing to determine the actual compound product quality information. The actual extract quality results and the actual compound product quality information are used as feedback information. Based on the actual extract quality results in the feedback information, the model parameters of the first model are adjusted; and the model parameters of the second model are adjusted based on the actual extract quality results and the actual compound product quality information in the feedback information. This allows for periodic iterative updates of the first and second models, forming a closed-loop optimization process.

[0060] The technical solution of this embodiment determines at least one spice to be tested associated with the pre-determined original spice blending information, and determines the raw material characteristic information and original process parameters corresponding to each spice to be tested. For at least one spice to be tested, the raw material characteristic information and original process parameters are processed based on a first model to determine the predicted quality result of the extract corresponding to the spice to be tested. The original spice blending information and the extract quality of at least one spice to be tested are input into a second model to obtain the predicted quality information of the blended finished product. When it is determined that the predicted quality information of the blended finished product does not meet the preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model. The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and original process parameters while satisfying at least one blending constraint condition, thereby obtaining the target spice blending information and target process parameters. This invention addresses the problems of low efficiency, high labor costs, and inability to guarantee the quality of the final product caused by trial-and-error adjustments based on human experience in existing technologies. By predicting the quality of flavor extracts and the quality of compounded products, it assesses whether adjustments to the original flavor compounding information and original process parameters are necessary. When adjustments are required, the above information is quantitatively adjusted based on the objective function and compounding constraints, avoiding the low efficiency caused by repeated manual trial-and-error adjustments. This improves the efficiency and accuracy of adjustment and optimization, thereby ensuring the quality of the compounded product.

[0061] Example 2 Figure 2 This is a flowchart of a method for determining tobacco flavoring processing parameters according to Embodiment 2 of the present invention. This embodiment is a preferred embodiment of the above embodiments. For specific implementation details, please refer to the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here. Figure 2 As shown, the method includes: S210. Based on the predetermined original flavor compound information, determine at least one flavor to be tested that is associated with the original flavor compound information, and determine the raw material characteristic information and original process parameters of the flavor to be tested.

[0062] The original process parameters are used to extract the extract from the fragrance to be tested.

[0063] S220. For at least one fragrance to be tested, the raw material characteristic information and original process parameters are processed based on the first model to determine the quality prediction result of the extract corresponding to the fragrance to be tested.

[0064] Optionally, the first model can process the original process parameters and raw material characteristic information in the following ways: perform data preprocessing on the raw material characteristic information and original process parameters to obtain preprocessed data; perform principal component analysis and / or coding on the preprocessed data to obtain the features to be processed corresponding to the preprocessed data; input the features to be processed into the first model to obtain the extract quality prediction results.

[0065] To ensure the accuracy of quality prediction, data preprocessing can be performed on raw material characteristic information and original process parameters. Optionally, data preprocessing may include at least: outlier detection and removal, spectral data smoothing and baseline correction, and data standardization. The spectral data may be the near-infrared spectral information mentioned above. The preprocessed data includes preprocessed raw material characteristic information and preprocessed original process parameters.

[0066] The features to be processed can be those obtained by feature extraction from preprocessed data. Optionally, principal component analysis can be performed on near-infrared spectral information and chromatographic fingerprints in the raw material feature information to determine the corresponding features to be processed. Additionally, other information in the raw material feature information (such as moisture content, total sugar content, total nitrogen content, origin information, harvest time information, and storage conditions) and original process parameters can be encoded using an encoder to determine the corresponding features to be processed.

[0067] Specifically, raw material characteristic information and original process parameters are preprocessed to obtain preprocessed data. Principal component analysis and / or coding are performed on the preprocessed data to obtain the features to be processed corresponding to the preprocessed data. Based on the first model, the features to be processed are processed to obtain the extract quality information corresponding to the fragrance to be detected.

[0068] If the original spice blend information includes at least the spice dosage information and extract dosage information of at least one spice to be tested, the following information can be input into the second model to obtain the predicted quality information of the blended product.

[0069] S230. Input the quality prediction results of the extract of at least one spice to be tested, the spice dosage information and the extract dosage information into the second model to obtain the predicted quality information of the compound finished product.

[0070] The information on the amount of flavoring used can be used to characterize the mass ratio of the flavoring to be tested required for the production of the compound product. The information on the amount of extract used can be used to characterize the mass ratio of the flavoring extract of the flavoring to be tested required for the production of the compound product.

[0071] The predicted quality information of the compound finished product includes quality prediction information under at least one quality dimension; at least one quality dimension includes one or more of the following: aroma component content dimension, sensory evaluation dimension, fragrance yield dimension, and fragrance color value dimension.

[0072] The quality prediction information under the aroma component content dimension can be used to characterize the aroma component content of the blended product. The sensory evaluation dimension may include at least one sensory evaluation sub-dimensionality. Optionally, at least one sensory evaluation sub-dimensionality includes one or more of aroma intensity, sweetness, and off-odors. The quality prediction information under the sensory evaluation dimension can be used to characterize the overall sensory evaluation score of the blended product. For example, based on the evaluation scores of the blended product under sensory evaluation dimensions such as aroma intensity, sweetness, and off-odors, the overall sensory evaluation score of the blended product under the sensory evaluation dimension can be determined.

[0073] Quality prediction information under the flavor yield dimension can be used to characterize the proportion of the quality of the compound product to the total quality of all flavorings tested. Quality prediction information under the flavor color value dimension can be used to characterize the color, hue, and clarity of the compound product.

[0074] Optionally, the predicted quality information of the compounded finished product can be represented by the following function: ; in, This indicates the predicted quality information of the compounded finished product. This indicates the amount of extract used, where m represents the type and quantity of flavor extracts. This indicates the amount of spices used, where n represents the number of different types of spices to be tested. This indicates the predicted results for the quality of the extract. This corresponds to the processing of the second model.

[0075] Optionally, the second model can be constructed in two ways. In one case, the second model is a machine learning model, such as a neural network model or a random forest model. The second model can be trained as follows: Acquire at least a preset number of historical compound formula data, and based on each historical compound formula data, determine the quality results of at least one sample spice's extract, the amount of sample spice used, the amount of sample extract used, and the quality information of the finished compound product corresponding to at least one sample spice, thereby constructing multiple training samples for training the second model. Based on these multiple training samples, train the second model to be trained to obtain a trained second model.

[0076] In another scenario, the second model is a linear mathematical model. The second model can be determined as follows: Based on at least a predetermined amount of historical compound formulation data, determine the mapping relationship between the quality results of the sample extracts, the amount of sample spices and extracts, and the quality information of the sample compound finished product. The second model is then constructed based on this mapping relationship. Here, the quality information of the sample compound finished product is equal to the weighted sum of the contribution values ​​of each spice extract and each sample spice, with the weighting coefficients being the corresponding amount of spice or extract. The contribution value of the spice extracts can be obtained based on the quality results of the sample extracts through linear or nonlinear mapping.

[0077] Preferably, a hybrid modeling approach can be adopted: using a linear mathematical model as the basic framework, a neural network is used to learn the residual part to construct a second model, so as to achieve both interpretability and prediction accuracy.

[0078] Specifically, based on the second model, the quality prediction results of the extract of at least one spice to be tested, the spice dosage information, and the extract dosage information are processed to obtain the predicted quality information of the compound finished product.

[0079] S240. If the quality prediction information under at least one quality dimension does not meet the threshold range of the corresponding quality dimension, it is determined that the predicted quality information of the compound finished product does not meet the preset conditions. Based on the raw material characteristic information, the first model and the second model, the objective function is determined.

[0080] The quality dimension threshold range can be a pre-set standard range of quality prediction information corresponding to the quality dimension.

[0081] Specifically, for the quality prediction information under at least one quality dimension in the predicted quality information of the compound finished product, if the quality prediction information under at least one quality dimension does not meet the corresponding quality dimension threshold range, then it is determined that the predicted quality information of the compound finished product does not meet the preset conditions. That is, it is determined that the original spice compounding information and the original process parameters need to be adjusted. Then, based on the raw material characteristic information, the first model, and the second model, the objective function is determined.

[0082] S250. Solve the objective function according to the preset optimization algorithm, and adjust the original spice blending information and original process parameters under the condition of satisfying at least one blending constraint, so as to obtain the target spice blending information and target process parameters.

[0083] Optionally, the compounding constraints include one or more of the following: the quality prediction information under at least one quality dimension meets the threshold range of the corresponding quality dimension; the total production cost corresponding to at least one flavoring to be tested does not exceed a preset cost threshold; and the flavoring utilization rate of the flavoring to be tested is not lower than a preset utilization rate threshold.

[0084] The total production cost can be used to characterize the total cost required to produce the compounded product. It should be noted that, under the constraint that the total production cost corresponding to at least one flavoring to be tested does not exceed a preset cost threshold, the flavoring cost information of the flavoring to be tested and the extract cost information of the flavoring extract of the flavoring to be tested can also be obtained. The flavoring cost information and extract cost information are then input into the objective function to calculate the total production cost based on the flavoring cost information and extract cost information.

[0085] Flavor utilization rate can be used to characterize the utilization of a flavoring agent under test. Optionally, the flavor utilization rate can be determined based on the ratio of the amount of flavoring agent actually utilized to the total amount of flavoring agent input. The preset utilization rate threshold can be a pre-set standard value for the utilization rate of the flavoring agent under test.

[0086] Specifically, the objective function is solved according to the preset optimization algorithm. Under the conditions that the quality prediction information under at least one quality dimension meets the corresponding quality dimension threshold range, the total production cost corresponding to at least one spice to be tested does not exceed the preset cost threshold, and the spice utilization rate of the spice to be tested is not lower than the preset utilization rate threshold, the original spice blending information and the original process parameters are adjusted to obtain the target spice blending information and the target process parameters.

[0087] For example, a first function is constructed based on the first model and the second model, using raw material characteristic information. The first function is used as input to obtain the objective function. At least one quality dimension threshold range is considered. (e.g., aroma intensity ≥ 8 points, sweetness ≥ 7 points, impurities ≤ 3 points, etc.), and the cost information of at least one flavoring agent to be tested. Cost information of the extract of the fragrance to be tested. , which serves as the input parameter for the objective function.

[0088] In addition, the original process parameters P (continuous variables: extraction temperature, extraction time, material-liquid ratio, etc.; discrete variables: extraction method, solvent type, etc.) and the original fragrance compound information R (fragrance dosage information and extract dosage information of at least one fragrance to be tested) are used as variables to be optimized.

[0089] Quality prediction information based on at least one quality dimension All meet the corresponding quality dimension threshold range The primary optimization objective is to minimize the total production cost (e.g., the total production cost may include extraction energy consumption, material loss and the cost of the spices to be tested) and maximize the spice utilization rate (e.g., the preset spice to be tested has the highest utilization rate). The secondary optimization objectives are that the total production cost does not exceed the preset cost threshold and the spice utilization rate of the spices to be tested is not lower than the preset utilization rate threshold.

[0090] The objective function is solved using multi-objective optimization algorithms or hierarchical optimization strategies to obtain the target spice blend information and target process parameters.

[0091] For example, let's take licorice as the spice to be tested, glycyrrhizic acid as the spice extract, and a compound spice blend composed of glycyrrhizic acid as an example. The glycyrrhizic acid content varies depending on the origin of the licorice, typically fluctuating between 1.5% and 4.2%. However, fluctuations in glycyrrhizic acid content can lead to unstable sweetness in the final compound spice blend. Therefore, after obtaining the licorice to be tested, we can acquire its raw material characteristic information, such as: the near-infrared spectrum (900-2500 nm) of the licorice, its origin information, and moisture content information. We can also determine the original process parameters corresponding to the licorice to be tested, such as: extraction temperature (60-90℃), time (2-6 h), material-to-liquid ratio (1:8-1:15), and ethanol concentration (50%-80%).

[0092] Principal component analysis (PCA) was performed on the near-infrared spectra of the raw materials to obtain spectral features after 30-dimensional PCA dimensionality reduction. Origin information was also encoded to obtain origin codes. Features corresponding to the original process parameters were determined. The 30-dimensional PCA-reduced spectral features, origin codes, and features corresponding to the original process parameters were input into a first model based on the XGBoost algorithm to obtain extract quality prediction results. For example, extract quality prediction results may include: glycyrrhizic acid content, glycyrrhizic acid yield, glycyrrhizic acid color value, and glycyrrhizic acid sensory sweetness score. It should be noted that the first model was trained based on training samples corresponding to 120 batches of licorice samples from the past three years. The 5-fold cross-validation results showed a coefficient of determination R0. 2 =0.87, RMSE=0.32.

[0093] The original flavor blend information and the predicted quality of the licorice extract to be tested are input into the second model to obtain the predicted quality information of the blended finished product. When the predicted quality information of the blended finished product corresponding to the licorice to be tested does not meet the preset conditions, parameter adjustments are made based on the first and second models. The original flavor blend information and original process parameters are adjusted with the sweetness score of the target blended finished product ≥8.0 as the blending constraint.

[0094] For example, there are three possible adjusted process parameters: 1) Extraction temperature of 75℃, extraction time of 4 hours, ethanol concentration of 65%, and glycyrrhizic acid content of 3.8%; 2) Extraction temperature of 82℃, extraction time of 5 hours, ethanol concentration of 70%, and glycyrrhizic acid content of 4.5%; 3) Extraction temperature of 68℃, extraction time of 3 hours, ethanol concentration of 60%, and glycyrrhizic acid content of 3.2%. The third set of process parameters, which offers the best cost, can be selected as the target process parameters. Actual production verification shows that the target compound product achieves a sweetness score of 8.0, meeting the target requirements. Compared to the traditional fixed process parameters, this method reduces costs by 23%.

[0095] For example, let's take licorice, tree moss, and chicory as examples. The corresponding fragrance extracts are glycyrrhizic acid, causalin, and caramel aroma, respectively. The compound product is a compound fragrance composed of glycyrrhizic acid, causalin, and caramel aroma.

[0096] If the glycyrrhizic acid content of the licorice to be tested is low (1.8%), the styraciferol content of the tree moss to be tested is normal (0.32%), and the caramel aroma component of the chicory to be tested is high (+15%), then the process parameters can be adjusted based on the above method.

[0097] Specifically, the extraction parameters for the licorice to be tested were adjusted to increase the glycyrrhizic acid content (up to a maximum of 3.5%). The extraction parameters for the tree moss to be tested were also adjusted to increase the causalin content (up to a maximum of +8%). Finally, the extraction parameters for the chicory to be tested were adjusted to reduce caramel aroma components by lowering the extraction temperature. Specifically, the licorice to be tested underwent enhanced extraction (temperature +5℃, time +1h), the chicory underwent mild extraction (temperature -8℃), and the tree moss underwent standard extraction. In the compounding process, the amount of licorice to be tested was reduced by 15%, the amount of chicory underwent increased by 10%, and the amount of tree moss underwent remained unchanged. Compared to the traditional method, this method reduced costs by 12%.

[0098] The technical solution of this embodiment determines at least one spice to be tested associated with the pre-determined original spice blending information, and determines the raw material characteristic information and original process parameters corresponding to each spice to be tested. For at least one spice to be tested, the raw material characteristic information and original process parameters are processed based on a first model to determine the extract quality prediction result corresponding to the spice to be tested. The extract quality prediction result of at least one spice to be tested, the spice dosage information, and the extract dosage information are input into a second model to obtain the predicted quality information of the blended finished product. If the quality prediction information under at least one quality dimension does not meet the corresponding quality dimension threshold range, it is determined that the predicted quality information of the blended finished product does not meet the preset conditions. Based on the raw material characteristic information, the first model, and the second model, an objective function is determined. The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and original process parameters while satisfying at least one blending constraint condition, thereby obtaining the target spice blending information and target process parameters. This invention achieves synergistic prediction of flavor compounding information and process parameters by predicting the quality of flavor extracts and compounded products. This allows for accurate determination of which data point in the flavor compounding information and process parameters needs quantitative adjustment when adjustments are required. Furthermore, it can determine the quality of the raw material under corresponding process parameters and flavor compounding information before practical application, providing forward-looking guidance. Setting at least one compounding constraint can meet production needs across multiple dimensions, ensuring the quality stability of the compounded product while reducing production costs and improving raw material utilization, thus enhancing the efficiency and accuracy of adjustment and optimization.

[0099] Example 3 Figure 3 This is a schematic diagram of a tobacco flavoring processing parameter determination device provided in Embodiment 3 of the present invention. Figure 3As shown, the device includes: a raw data acquisition module 310, an extract quality prediction module 320, a finished product quality prediction module 330, an objective function determination module 340, and a parameter adjustment module 350.

[0100] The raw data acquisition module 310 is used to determine at least one spice to be tested associated with the pre-determined raw spice blending information, and to determine the raw material characteristic information and original process parameters of the spice to be tested; wherein the original process parameters are used to extract extracts from the spice to be tested; the extract quality prediction module 320 is used to process the raw material characteristic information and the original process parameters based on a first model for at least one spice to be tested, and to determine the extract quality prediction result corresponding to the spice to be tested; the finished product quality prediction module 330 is used to combine the raw spice blending information and the original process parameters. The quality prediction result of the extract of one less spice to be tested is input into the second model to obtain the predicted quality information of the compound product corresponding to the original spice compound information; the objective function determination module 340 is used to determine the objective function based on the raw material characteristic information, the first model and the second model when it is determined that the predicted quality information of the compound product does not meet the preset conditions; the parameter adjustment module 350 is used to solve the objective function according to the preset optimization algorithm, so as to adjust the original spice compound information and the original process parameters under the condition of satisfying at least one compounding constraint, to obtain the target spice compound information and the target process parameters.

[0101] The technical solution of this embodiment determines at least one spice to be tested associated with the pre-determined original spice blending information, and determines the raw material characteristic information and original process parameters corresponding to each spice to be tested. For at least one spice to be tested, the raw material characteristic information and original process parameters are processed based on a first model to determine the predicted quality result of the extract corresponding to the spice to be tested. The original spice blending information and the extract quality of at least one spice to be tested are input into a second model to obtain the predicted quality information of the blended finished product. When it is determined that the predicted quality information of the blended finished product does not meet the preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model. The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and original process parameters while satisfying at least one blending constraint condition, thereby obtaining the target spice blending information and target process parameters. This invention addresses the problems of low efficiency, high labor costs, and inability to guarantee the quality of the final product caused by trial-and-error adjustments based on human experience in existing technologies. By predicting the quality of flavor extracts and the quality of compounded products, it assesses whether adjustments to the original flavor compounding information and original process parameters are necessary. When adjustments are required, the above information is quantitatively adjusted based on the objective function and compounding constraints, avoiding the low efficiency caused by repeated manual trial-and-error adjustments. This improves the efficiency and accuracy of adjustment and optimization, thereby ensuring the quality of the compounded product.

[0102] Based on the above embodiments, optionally, the device further includes: a parameter application module, used to send the target spice blend information and the target process parameters to a target terminal, so as to obtain a target blend product based on the target spice blend information and the target process parameters.

[0103] Optionally, the device further includes: a feedback adjustment module, used to acquire feedback information corresponding to the target compound product, and to adjust the model parameters of the first model and the second model based on the feedback information; wherein the feedback information includes at least: actual extract quality results and actual compound product quality information.

[0104] Optionally, the raw data acquisition module includes: a parameter determination unit, used to perform detection processing on the spice to be detected under at least one detection dimension to obtain raw material characteristic information; wherein, at least one of the detection dimensions includes one or more of the following: near-infrared spectroscopy detection dimension, chromatographic fingerprint detection dimension, chemical composition detection dimension, and raw material traceability detection dimension; and to retrieve the original process parameters corresponding to the spice to be detected; wherein, the original process parameters include at least: the original extraction time, original extraction temperature, original material liquid information, and original extraction pressure information of the spice to be detected.

[0105] Optionally, the extract quality prediction module is used to preprocess the raw material characteristic information and the original process parameters to obtain preprocessed data; perform principal component analysis and / or encoding on the preprocessed data to obtain the features to be processed corresponding to the preprocessed data; and input the features to be processed into the first model to obtain the extract quality prediction result.

[0106] Optionally, the original flavoring compound information includes at least: flavoring dosage information and extract dosage information of at least one flavoring to be tested; and a finished product quality prediction module, used to input the extract quality prediction result, flavoring dosage information and extract dosage information of at least one flavoring to be tested into a second model to obtain the compound finished product prediction quality information; wherein, the compound finished product prediction quality information includes quality prediction information under at least one quality dimension; the at least one quality dimension includes one or more of the following: aroma component content dimension, sensory evaluation dimension, flavoring yield dimension and flavoring color value dimension.

[0107] Optionally, the objective function determination module includes: a quality information judgment unit, used to determine that the predicted quality information of the compound finished product does not meet the preset conditions if the quality prediction information under at least one of the quality dimensions does not meet the threshold range of the corresponding quality dimension.

[0108] Optionally, the compounding constraints include one or more of the following: the quality prediction information under at least one quality dimension meets the corresponding quality dimension threshold range; the total production cost corresponding to at least one of the tested spices does not exceed a preset cost threshold; and the spice utilization rate of the tested spices is not lower than a preset utilization rate threshold.

[0109] The tobacco flavoring processing parameter determination device provided in this embodiment of the invention can execute the tobacco flavoring processing parameter determination method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0110] Example 4 Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0111] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0112] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0113] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for determining tobacco flavoring processing parameters.

[0114] In some embodiments, the method for determining tobacco flavoring processing parameters may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for determining tobacco flavoring processing parameters described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for determining tobacco flavoring processing parameters by any other suitable means (e.g., by means of firmware).

[0115] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0116] The computer program for implementing the method for determining tobacco flavoring processing parameters of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are performed. The computer program can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0117] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0118] Example 5 Embodiment 5 of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute a method for determining tobacco flavoring processing parameters, the method comprising: Based on predetermined original flavor compound information, at least one flavor to be tested associated with the original flavor compound information is identified, and raw material characteristic information and original process parameters of the flavor to be tested are determined; wherein, the original process parameters are used to extract the extract from the flavor to be tested. For at least one of the flavorings to be detected, the raw material characteristic information and the original process parameters are processed based on the first model to determine the extract quality prediction result corresponding to the flavoring to be detected; The original spice blend information and the quality prediction results of the extract of at least one spice to be tested are input into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information. When it is determined that the predicted quality information of the compound finished product does not meet the preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model; The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and the original process parameters under at least one blending constraint condition, so as to obtain the target spice blending information and the target process parameters.

[0119] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0120] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0121] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0122] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0123] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the method for determining tobacco flavoring processing parameters as provided in any embodiment of this application.

[0124] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider). This program product belongs to the same inventive concept as the tobacco flavoring processing parameter determination method disclosed in the embodiments of this application, and therefore will not be described further here.

[0125] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0126] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining tobacco flavoring processing parameters, characterized in that, include: Based on predetermined original flavor compound information, at least one flavor to be tested associated with the original flavor compound information is identified, and raw material characteristic information and original process parameters of the flavor to be tested are determined; wherein, the original process parameters are used to extract the extract from the flavor to be tested. For at least one of the flavorings to be detected, the raw material characteristic information and the original process parameters are processed based on the first model to determine the extract quality prediction result corresponding to the flavoring to be detected; The original spice blend information and the quality prediction results of the extract of at least one spice to be tested are input into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information. When it is determined that the predicted quality information of the compound finished product does not meet the preset conditions, an objective function is determined based on the raw material characteristic information, the first model, and the second model; The objective function is solved according to a preset optimization algorithm to adjust the original spice blending information and the original process parameters under at least one blending constraint condition, so as to obtain the target spice blending information and the target process parameters.

2. The method according to claim 1, characterized in that, The method further includes: The target spice blend information and the target process parameters are sent to the target terminal to obtain the target blended finished product based on the target spice blend information and the target process parameters.

3. The method according to claim 2, characterized in that, The method further includes: Obtain feedback information corresponding to the target compound product, and adjust the model parameters of the first model and the second model based on the feedback information; The feedback information includes at least: the actual quality results of the extract and the actual quality information of the compounded product.

4. The method according to claim 1, characterized in that, The determination of the raw material characteristic information and original process parameters of the spice to be detected includes: The spice to be tested is subjected to detection processing under at least one detection dimension to obtain raw material characteristic information; wherein, at least one of the detection dimensions includes one or more of the following: near-infrared spectroscopy detection dimension, chromatographic fingerprint detection dimension, chemical composition detection dimension, and raw material traceability detection dimension; Retrieve the original process parameters corresponding to the spice to be tested; wherein the original process parameters include at least: the original extraction time, the original extraction temperature, the original liquid information, and the original extraction pressure information of the spice to be tested.

5. The method according to claim 1, characterized in that, The step of processing the raw material characteristic information and the original process parameters based on the first model to determine the extract quality prediction result corresponding to the flavoring to be detected includes: The raw material characteristic information and the original process parameters are preprocessed to obtain preprocessed data; Principal component analysis and / or encoding processing are performed on the preprocessed data to obtain the features to be processed corresponding to the preprocessed data; The features to be processed are input into the first model to obtain the quality prediction results of the extract.

6. The method according to claim 1, characterized in that, The original spice blend information includes at least: spice dosage information and extract dosage information for at least one spice to be tested. The step of inputting the original spice blend information and the extract quality prediction results of at least one spice to be tested into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information includes: The quality prediction results of the extract of at least one spice to be tested, the spice dosage information, and the extract dosage information are input into the second model to obtain the predicted quality information of the compound finished product. The predicted quality information of the compound finished product includes quality prediction information under at least one quality dimension; the at least one quality dimension includes one or more of the following: aroma component content dimension, sensory evaluation dimension, fragrance yield dimension, and fragrance color value dimension.

7. The method according to claim 6, characterized in that, Determining that the predicted quality information of the compounded finished product does not meet the preset conditions includes: If the quality prediction information under at least one of the quality dimensions does not meet the corresponding quality dimension threshold range, then it is determined that the predicted quality information of the compound finished product does not meet the preset conditions.

8. The method according to claim 1, characterized in that, The compounding constraints include one or more of the following: The quality prediction information for at least one quality dimension meets the corresponding quality dimension threshold range. The total production cost corresponding to at least one of the detected spices does not exceed a preset cost threshold; The utilization rate of the spice to be tested is not lower than a preset utilization rate threshold.

9. A device for determining tobacco flavoring processing parameters, characterized in that, include: The raw data acquisition module is used to determine at least one spice to be detected associated with the pre-determined raw spice blend information, and to determine the raw material characteristic information and raw process parameters of the spice to be detected; wherein, the raw process parameters are used to extract the extract from the spice to be detected. An extract quality prediction module is used to process the raw material characteristic information and the original process parameters based on a first model for at least one of the flavorings to be tested, and determine the extract quality prediction result corresponding to the flavoring to be tested. The finished product quality prediction module is used to input the original spice blend information and the quality prediction results of the extract of at least one spice to be tested into the second model to obtain the predicted quality information of the blended finished product corresponding to the original spice blend information. The objective function determination module is used to determine an objective function based on the raw material characteristic information, the first model, and the second model when it is determined that the predicted quality information of the compound finished product does not meet the preset conditions. The parameter adjustment module is used to solve the objective function according to a preset optimization algorithm, so as to adjust the original spice blend information and the original process parameters under the condition of satisfying at least one blending constraint, so as to obtain the target spice blend information and the target process parameters.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for determining tobacco flavoring processing parameters according to any one of claims 1-8.

Images