A freeze-drying control method and system for fruits and vegetables based on multi-stage quality detection
Through multi-stage quality detection methods, including basic attribute information collection, visual inspection and high-performance liquid chromatography analysis, quality detection models are constructed and production process parameters are optimized, and the problem of inaccurate evaluation of fruit and vegetable lyophilization quality in the existing technology is solved, and the production control level and product quality are improved.
Patent Information
- Application Number
- CN202410660644.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-27
- Publication Date
- 2025-06-03
- Estimated Expiration
- 2044-05-27
AI Technical Summary
The evaluation of the lyophilized quality of fruits and vegetables in the prior art is not accurate enough, resulting in a low level of production control and the inability to adjust the detection standards and processing parameters according to consumers' preferences.
A multi-stage quality detection method is adopted, including collecting basic attribute information of fruits and vegetables freeze-dried, performing visual inspection and high-performance liquid chromatography analysis, building a quality detection model, and optimizing production process parameters according to market demand.
It improves the accuracy of the evaluation of freeze-dried quality of fruits and vegetables, realizes accurate analysis and quantification of nutrients, and improves production control level and product quality.
Smart Images

Figure CN118691134B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of food detection, and in particular to a fruit and vegetable freeze-drying control method and system based on multi-stage quality detection. Background Art
[0002] With the improvement of people's living standards and the enhancement of health awareness, the demand for high-quality, nutritious and convenient fruit and vegetable products is increasing. Fruit and vegetable freeze-drying technology has been widely used in the food industry because it can better preserve the nutrients, flavor and color of fruits and vegetables, and has the advantages of a long shelf life and easy storage and transportation. However, the quality inspection standards for freeze-dried fruits and vegetables are too simple to adjust the inspection standards according to actual needs, and optimize the production process parameters accordingly.
[0003] In the prior art, there are technical problems such as inaccurate quality assessment of freeze-dried fruits and vegetables and low level of control over product production. Summary of the invention
[0004] The embodiment of the present application provides a fruit and vegetable freeze-drying control method and system based on multi-stage quality detection. The inventors found that the quality standards and various parameters in the existing dried fruit production process are relatively fixed. However, due to differences in regions and seasons, consumers have different preferences for various indicators of fruit and vegetable freeze-drying, resulting in different degrees of preference for fruit and vegetable freeze-drying. However, in the prior art, there are technical problems such as inaccurate quality assessment of fruit and vegetable freeze-drying and low level of product production control, which makes it impossible for factories to make targeted adjustments to the inspection standards for the exit of fruit and vegetable freeze-drying and various parameters during processing according to the different preferences of consumers for various indicators of fruit and vegetable freeze-drying.
[0005] In view of the above problems, the present application provides a fruit and vegetable freeze-drying control method and system based on multi-stage quality detection.
[0006] In a first aspect of the present application, a method for controlling freeze-drying of fruits and vegetables based on multi-stage quality detection is provided, the method comprising:
[0007] Collect basic attribute information of freeze-dried fruits and vegetables to be tested, wherein the basic attribute information includes appearance attribute information, production attribute information, and physical property attribute information;
[0008] Perform visual inspection based on the appearance attribute information to obtain a first sample to be inspected;
[0009] Connecting to a high performance liquid chromatography module, and performing a nutrient component analysis on the first sample to be detected based on the high performance liquid chromatography module to obtain a nutrient component chromatogram, wherein the nutrient component chromatogram comprises N chromatographic peaks, where N is an integer greater than 0;
[0010] Quantify the nutrient components of the N chromatographic peaks to obtain a nutrient component index group;
[0011] Set a time window, collect the preference status of freeze-dried fruits and vegetables based on the time window, and generate a first market demand;
[0012] Decompose the production attribute information, obtain the production process parameters of the to-be-detected freeze-dried fruits and vegetables, and construct a quality detection model based on the production process parameters;
[0013] Input the physical property attribute information, the nutrient component index group, and the first market demand into the quality detection model for quality detection, and generate a quality performance result;
[0014] If the passing rate of the quality performance result is lower than the quality threshold, calculate the passing rate deviation, retrieve the optimization strategy library based on the passing rate deviation to obtain an optimization strategy, and optimize and control the production process parameters based on the optimization strategy.
[0015] In the second aspect of the present application, a control system for freeze-dried fruits and vegetables based on multi-stage quality detection is provided. The system includes:
[0016] A basic attribute information collection module for freeze-dried fruits and vegetables, which is used to collect the basic attribute information of the to-be-detected freeze-dried fruits and vegetables. The basic attribute information includes appearance attribute information, production attribute information, and physical property attribute information;
[0017] A first to-be-detected sample acquisition module, which is used to perform visual detection based on the appearance attribute information to obtain a first to-be-detected sample;
[0018] A connected chromatographic module, which is used to connect to a high-performance liquid chromatography module, perform nutrient component analysis on the first to-be-detected sample based on the high-performance liquid chromatography module, and obtain a nutrient component chromatogram. The nutrient component chromatogram contains N chromatographic peaks, where N is an integer greater than 0;
[0019] A nutrient component index group acquisition module, which is used to quantify the nutrient components of the N chromatographic peaks to obtain a nutrient component index group;
[0020] A first market demand generation module, which is used to set a time window, collect the preference status of freeze-dried fruits and vegetables based on the time window, and generate a first market demand;
[0021] A quality detection model construction module, which is used to decompose the production attribute information, obtain the production process parameters of the to-be-detected freeze-dried fruits and vegetables, and construct a quality detection model based on the production process parameters;
[0022] A quality performance result generation module, which is used to input the physical characteristic attribute information, the nutrient composition index group, and the first market demand into the quality detection model for quality detection and generate a quality performance result;
[0023] A process parameter optimization and control module, which is used to calculate the compliance rate deviation if the compliance rate of the quality performance result is lower than the quality threshold, retrieve the optimization strategy library based on the compliance rate deviation to obtain an optimization strategy, and optimize and control the production process parameters based on the optimization strategy.
[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
[0025] A freeze-dried fruit and vegetable control method based on multi-stage quality detection provided by an embodiment of the present application relates to the technical field of food detection. It collects the basic attribute information of the freeze-dried fruit and vegetable to be detected, conducts visual detection based on the appearance attribute information to obtain a first sample to be detected, connects to a high-performance liquid chromatography module to obtain a nutrient composition chromatogram, quantifies the nutrient composition for the N chromatographic peaks to obtain a nutrient composition index group, sets a time window to generate a first market demand, decomposes the production attribute information, constructs a quality detection model, inputs the physical characteristic attribute information, the nutrient composition index group, and the first market demand into the quality detection model to generate a quality performance result. If the compliance rate of the quality performance result is lower than the quality threshold, calculate the compliance rate deviation, retrieve the optimization strategy library based on the compliance rate deviation to obtain an optimization strategy, and optimize and control the production process parameters based on the optimization strategy. It solves the technical problems in the prior art that the quality evaluation of freeze-dried fruit and vegetables is not accurate enough and the production control level of products is low. It enables the factory to make targeted adjustments to the detection standards and various parameters during processing according to the different preferences of consumers for various indicators of freeze-dried fruit and vegetables. It achieves the technical effects of improving the accuracy of the quality evaluation of freeze-dried fruit and vegetables, realizing the precise analysis and quantification of nutrient components, and enhancing the production control level and product quality of freeze-dried fruit and vegetables.
[0026] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the following specifically gives the specific implementation manners of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a schematic flowchart of a freeze-dried fruit and vegetable control method based on multi-stage quality detection provided by an embodiment of the present application;
[0028] Figure 2 This is a schematic diagram of the process for analyzing the nutrient components of the first sample to be detected in a fruit and vegetable freeze-drying control method based on multi-stage quality detection provided by an embodiment of the present application;
[0029] Figure 3 This is a schematic diagram of the structure of a fruit and vegetable freeze-drying control system based on multi-stage quality detection provided by an embodiment of the present application.
[0030] Explanation of reference numerals: Fruit and vegetable freeze-drying basic attribute information collection module 10, first sample to be detected acquisition module 20, connection chromatograph module 30, nutrient component index group acquisition module 40, first market demand generation module 50, quality detection model construction module 60, quality performance result generation module 70, process parameter optimization control module 80. Detailed implementation manners
[0031] An embodiment of the present application provides a fruit and vegetable freeze-drying control method based on multi-stage quality detection, which is used to solve the technical problems in the prior art that the evaluation of the quality of fruit and vegetable freeze-drying is not accurate enough and the production control level of products is low.
[0032] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without making creative efforts belong to the scope of protection of the present application.
[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.
[0034] Embodiment 1:
[0035] As Figure 1 shown, a fruit and vegetable freeze-drying control method based on multi-stage quality detection, the method includes:
[0036] Step S100: Collect the basic attribute information of the freeze-dried fruits and vegetables to be detected, where the basic attribute information includes appearance attribute information, production attribute information, and physical property attribute information.
[0037] Specifically, collecting the basic information of the freeze-dried fruits and vegetables to be detected can conduct a detailed evaluation of the freeze-dried fruits and vegetables from multiple perspectives. In terms of quality control, this information can provide a comprehensive basis to more accurately grasp the quality status of the product, and at the same time provide rich data support for the improvement and optimization of the product. The appearance attribute information includes color, shape, integrity, etc., which can intuitively reflect the appearance quality of the product; the production attribute information contains various relevant data in the production process, such as processing temperature, time, process parameters, etc., which helps to analyze the impact of the production link on the product quality; the attribute information of the physical properties such as density and hardness can further reveal the internal physical properties of the product and provide a reference for aspects such as its taste and storage.
[0038] Step S200: Conduct visual inspection based on the appearance attribute information to obtain the first sample to be detected.
[0039] Specifically, visual inspection uses an image acquisition device such as a camera to obtain appearance attribute information, and then analyzes the image through image processing algorithms and technologies, such as identifying the contour of an object, extracting color features, etc., to determine whether the product meets specific standards or conditions. Finally, samples that meet specific requirements or exhibit specific characteristics are selected from among many objects to be detected and marked as the first sample to be detected. The appearance attribute information includes various characteristics such as the shape, size, color, surface texture, and glossiness of an object that can be directly or indirectly observed visually, and this information is used as input information and provided to the visual inspection system.
[0040] Step S300: Connect the high-performance liquid chromatography module, and conduct nutritional component analysis on the first sample to be detected based on the high-performance liquid chromatography module to obtain a nutritional component chromatogram, where the nutritional component chromatogram contains N chromatographic peaks, and N is an integer greater than 0.
[0041] Specifically, the high performance liquid chromatography module is an important instrument module for analyzing and separating chemical components in complex mixtures. It mainly consists of an infusion system, an injection system, a chromatographic column, a detection system, etc. By measuring and calculating parameters such as chromatographic peak area or height, the content of nutritional components can be accurately determined, providing accurate data support for the nutritional evaluation of products. The high performance liquid chromatography module analyzes the nutritional components of the first sample to be detected, obtains a chromatogram of nutritional components, which contains N chromatographic peaks. Each independent chromatographic peak corresponds to a specific nutritional component, and the height and area of the peak are closely related to the content of this nutritional component. The higher the peak or the larger the area, the relatively higher the content of this nutritional component in the sample. This analysis method realizes the precise analysis and quantification of the nutritional components of freeze-dried fruits and vegetables.
[0042] Step S400: Quantify the nutritional components of the N chromatographic peaks to obtain a set of nutritional component indicators.
[0043] Specifically, quantifying the nutritional components of the N chromatographic peaks means accurately determining the specific content of the nutritional component represented by each chromatographic peak. By comparing with known standards and corresponding calculation methods, the nutritional component corresponding to each peak is determined, and various quantitative analysis techniques, such as calculating based on peak area, are used to obtain specific values. The set of nutritional component indicators obtained is a collection composed of these quantitative data, which helps to carry out quality control and quality identification, ensuring the stability and reliability of the quality of freeze-dried fruits and vegetables. This indicator set contains detailed information about the content of various nutritional components.
[0044] Step S500: Set a time window, collect the preference status of freeze-dried fruits and vegetables based on the time window, and generate a first market demand.
[0045] Specifically, setting a time window can specifically collect information about the preference status of freeze-dried fruits and vegetables within a specific time period, and can more accurately understand consumers' preference levels for freeze-dried fruits and vegetables during a specific period, such as in different regions, a certain season, or a certain holiday period. The collected preference status data is sorted and analyzed to generate a first market demand, and the product can be better optimized based on the analysis of the first market demand.
[0046] Step S600: Decompose the production attribute information, obtain the production process parameters of the freeze-dried fruits and vegetables to be detected, and construct a quality detection model based on the production process parameters.
[0047] Specifically, production process parameters are used to construct a quality inspection model, which takes production process parameters as key input variables. The quality inspection model is constructed based on statistical methods, machine learning algorithms or other related technologies, and predicts or evaluates the quality characteristics of freeze-dried fruits and vegetables according to the input production process parameters, helping to monitor and adjust the process parameters in real time during the production process and improving the production control level of freeze-dried fruits and vegetables. Decomposing the production attribute information can accurately obtain the production process parameters of the freeze-dried fruits and vegetables to be detected, such as specific temperature setting values, specific values of drying duration, pressure data at different stages, and so on.
[0048] Step S700: Input the physical property attribute information, nutritional component index group, and first market demand into the quality inspection model for quality inspection to generate a quality performance result.
[0049] Specifically, the quality inspection model comprehensively processes and analyzes these input information, applies built-in algorithms and rules, correlates, compares and evaluates this information, and the finally generated quality performance result is a comprehensive evaluation and presentation of the quality of this freeze-dried fruit and vegetable. Exemplarily, the quality performance result may show that the physical properties of the product are good, the nutritional components are relatively rich, but there is still room for improvement in some key indicators of market demand, or it may show that the product is outstanding in a certain aspect, but other aspects need to be improved to better meet the market demand. Optimize the production process and adjust the product formula according to the quality performance result to further improve the quality of the product. The physical property attribute information includes specific characteristic information such as the appearance, texture, and hardness of the freeze-dried fruits and vegetables, which can intuitively reflect the performance of the product in terms of physical properties; the nutritional component index group details the specific content and index conditions of various nutritional components contained in the product; the first market demand reflects the current market's preferences, trends and expectations for this type of freeze-dried fruit and vegetable product. The type of target dried fruit and the batch information of the target dried fruit can be obtained through the first market demand, and the physical property attribute information and nutritional component index of the target dried fruit can be obtained according to its batch information.
[0050] Step S800: If the pass rate of the quality performance result is lower than the quality threshold, calculate the pass rate deviation, retrieve the optimization strategy library based on the pass rate deviation to obtain an optimization strategy, and optimize and control the production process parameters based on the optimization strategy.
[0051] Specifically, a pass rate of the quality performance results lower than the quality threshold indicates that there is significant room for improvement in certain aspects of the product; the deviation of the pass rate reflects the degree of the gap between the actual pass situation and the ideal pass situation; the optimization strategy library pre-stores response strategies for various possible situations, and an optimization strategy suitable for the current situation is found through the analysis and matching of the deviation; then, targeted optimization control is performed on the production process parameters according to the obtained optimization strategy; through such optimization control of the production process based on the optimization strategy, the production control level and quality standard of freeze-dried fruits and vegetables are improved. Among them, the quality threshold can be gradually optimized according to the first market demand, and the specific method is as follows:
[0052] Obtain the first market demand, obtain the types of target dried fruits and the batch information of the target dried fruits. According to the batch information, the quality parameters of the target dried fruits (such as physical property attribute information, nutritional component indicators) can be obtained, and the data is cleaned and standardized: missing values and outliers are processed, the data format is standardized, and after ensuring the data quality, the market demand data and the quality inspection data are integrated to form a comprehensive data set;
[0053] Calculate descriptive statistics such as the mean, standard deviation, minimum value, and maximum value of each quality parameter (such as hardness, moisture content, nutritional component content, etc.).
[0054] Obtain the basic situation of the market demand data, such as average sales volume, seasonal fluctuations, holiday sales peaks, etc.
[0055] Calculate the Pearson correlation coefficient between each quality parameter and the market demand to evaluate the linear correlation.
[0056] For non-linear relationships, the Spearman rank correlation coefficient can be calculated.
[0057] Construct a multiple regression model with the market demand (such as sales volume, repurchase rate) as the dependent variable and each quality parameter as the independent variable.
[0058] Evaluate the goodness of fit (R² value) of the regression model and the significance (p value) of each independent variable.
[0059] Through the regression coefficient, judge the specific influence degree of each quality parameter on the market demand.
[0060] Regression model verification: Use the cross-validation method to evaluate the stability and prediction ability of the regression model.
[0061] Check the regression residuals to ensure that the model has no obvious systematic bias.
[0062] Cluster analysis: Use a clustering algorithm (such as K-means clustering) to classify dried fruit products and identify product groups with different quality grades.
[0063] Analyze the market demand characteristics in different clusters to identify product groups with high sales volume and high repurchase rate.
[0064] Classification analysis: Use classification algorithms (such as decision trees, random forests) to build a classification model to predict the market demand under different combinations of quality parameters.
[0065] Evaluate the accuracy of the classification model and the influence of key parameters.
[0066] Use feature selection algorithms (such as recursive feature elimination RFE, Lasso regression) to screen out the most important quality parameters.
[0067] Ensure that the selected features have a significant impact on market demand and the model complexity is moderate.
[0068] Model training and validation: Train machine learning models (such as random forests, support vector machines SVM) to predict the relationship between market demand and quality parameters.
[0069] Use the training set and validation set to evaluate the model to ensure the accuracy and generalization ability of the model.
[0070] Key parameter evaluation: Synthesize the results of correlation analysis, regression analysis, clustering analysis and machine learning models to identify key quality parameters that have a significant impact on market demand.
[0071] Focus on key parameters that show significant influence in multiple analysis methods.
[0072] Optimize the parameters in the quality threshold through the key parameters.
[0073] Continuous data collection: Continuously collect market demand and quality inspection data to maintain the currency and integrity of the data.
[0074] Regularly update the analysis model to ensure the effectiveness of the model.
[0075] Feedback and optimization: Continuously optimize the quality parameters and production process of the dried fruits according to the latest data analysis results.
[0076] Enable a closed-loop feedback mechanism to quickly transmit market feedback to the production link for timely adjustment, thereby guiding production optimization, improving product quality, and meeting market demand.
[0077] In a possible implementation manner, such as Figure 2 , step S300 further includes:
[0078] Step S310: Obtain the standard fruit and vegetable freeze-dried product of the to-be-detected fruit and vegetable freeze-dried product, and retrieve the nutritional composition information of the standard fruit and vegetable freeze-dried product.
[0079] Step S320: Based on the nutritional composition information, perform polarity judgment, and based on the polarity judgment result, set the mobile phase composition information and the mobile phase flow rate.
[0080] Step S330: Based on the nutritional composition information, set the parameter information of the detector.
[0081] Step S340: Configure the high performance liquid chromatography module based on the mobile phase composition information, the mobile phase flow rate, and the parameter information.
[0082] Specifically, call out the detailed nutritional composition information contained in the standard freeze-dried fruits and vegetables through a specific database or storage system, including the content data of specific components such as various vitamins, minerals, sugars, proteins, etc. Different nutritional components have different polarities, and it is crucial to select a suitable mobile phase. Based on the result of the polarity judgment, reasonably set the composition of the mobile phase and determine the flow rate of the mobile phase to ensure the effective separation and detection of different nutritional components. The setting of the mobile phase flow rate determines the running speed and separation effect of the sample in the chromatographic column. A suitable flow rate can enable good separation of different components, and combined with the previously set detector parameter information, the detector can work in coordination with the mobile phase and the chromatographic column to sensitively and accurately detect the separated components at a specific wavelength. When setting the parameter information of the detector based on the nutritional composition information, multiple aspects need to be considered. For example, set the detection wavelength of the detector according to the characteristic absorption wavelength of different nutritional components to ensure that the signal of specific components can be detected to the greatest extent. If there are multiple components to be detected and their absorption wavelengths are different, it may be necessary to select a suitable range or adopt a multi-wavelength detection mode. Other parameters, such as the response time and linear range of the detector, also need to be optimized specifically according to the nutritional composition information to meet the detection requirements of different components and improve the reliability and accuracy of detection. Accurately input the mobile phase composition information into the system to ensure that the mobile phase can be mixed and supplied according to the set ratio during actual operation.
[0083] In a possible implementation manner, step S300 further includes:
[0084] Step S340: Input the first sample to be detected into the high performance liquid chromatography module. The high performance liquid chromatography module separates the first sample to be detected based on the mobile phase, in combination with the stationary phase and the partition coefficient. The detector receives the separation result, records the chromatogram, and obtains N chromatographic peaks.
[0085] Specifically, the first sample to be detected is input into the high-performance liquid chromatography module. The mobile phase runs in the system according to the set composition and flow rate, interacts with the stationary phase. Different components have different distribution coefficients between the stationary phase and the mobile phase, resulting in different moving speeds in the chromatographic column, thus achieving separation. The detector will receive these separated components in real time, convert them into electrical signals for recording, and finally form N chromatographic peaks on the chromatogram. Each chromatographic peak represents one or a class of components.
[0086] In a possible implementation manner, step S400 further includes:
[0087] Step S410: Connect to the fruit and vegetable nutrient database, where the fruit and vegetable nutrient database contains multiple chromatographic peak areas and multiple fruit and vegetable nutrients, and the multiple chromatographic peak areas correspond one-to-one to the multiple fruit and vegetable nutrients.
[0088] Step S420: Perform symmetry analysis on the N chromatographic peaks to obtain I symmetric chromatographic peaks and J asymmetric chromatographic peaks, where I + J = N.
[0089] Step S430: Calculate the areas of the I symmetric chromatographic peaks based on the symmetric area formula to obtain I symmetric areas, and calculate the areas of the J asymmetric chromatographic peaks based on the asymmetric area formula to obtain J asymmetric areas.
[0090] Specifically, connecting to the fruit and vegetable nutrient database is to compare and analyze the chromatographic peak areas detected by the high-performance liquid chromatography module with the multiple known fruit and vegetable nutrients in the database. This data contains information on multiple fruit and vegetable nutrients and their corresponding chromatographic peak areas. By matching the actually detected chromatographic peak areas with the data in the database, the nutrients in the sample can be accurately identified and quantified.
[0091] Performing symmetry analysis on the N chromatographic peaks, a symmetric chromatographic peak means that the shapes on its left and right sides are relatively symmetric, and the rising and falling trends are relatively similar and gentle, representing a single and pure component with less interference during the separation process. An asymmetric chromatographic peak indicates that there are differences in the shapes on its left and right sides. This asymmetry is caused by various factors, such as the influence of impurities or the action of certain specific conditions during the separation process that cause the peak shape to change.
[0092] Divide N chromatographic peaks into I symmetric chromatographic peaks and J asymmetric chromatographic peaks, where I represents the number of symmetric chromatographic peaks and J represents the number of asymmetric chromatographic peaks. The symmetric area can more accurately evaluate the content of relatively pure and single components; the asymmetric area can determine the relative content of complex components. Based on the symmetric area formula: A = 1.064h×0.5w, and the asymmetric area formula: A = h×(0.15w + 0.85w) / 2; where A is the estimated chromatographic peak area, h is the peak height of the chromatographic peak, and w is the width of the chromatographic peak. Calculate the areas of I symmetric chromatographic peaks and J asymmetric chromatographic peaks to more accurately understand the distribution and content of various nutritional components in the sample.
[0093] In one possible implementation, step S420 further includes:
[0094] Based on the N chromatographic peaks, obtain the first chromatographic peak.
[0095] Set the tailing peak height, and calculate the tailing factor of the first chromatographic peak based on the tailing peak height.
[0096] If the tailing factor is 1, the first chromatographic peak is a symmetric chromatographic peak; if the tailing factor is not 1, the first chromatographic peak is an asymmetric chromatographic peak.
[0097] Specifically, based on the N chromatographic peaks, select a representative chromatographic peak as the first chromatographic peak according to the experimental purpose and sample characteristics; setting the tailing peak height to calculate the tailing factor of the first chromatographic peak is a quantitative method to judge the asymmetry degree of the chromatographic peak, and accurately judge whether the first chromatographic peak is a symmetric chromatographic peak or an asymmetric chromatographic peak through the tailing factor. Set the width of the first half of the peak at 10% of the peak height of the chromatographic peak as a, and the width of the second half of the peak at the same height as b. The tailing factor = b / a. If the tailing factor is 1, the first chromatographic peak is a symmetric chromatographic peak; if the tailing factor is not 1, the first chromatographic peak is an asymmetric chromatographic peak.
[0098] In one possible implementation, step S600 further includes:
[0099] Step S610: Based on the production process parameters, retrieve and obtain the historical physical property attribute information, historical nutritional component index group, and historical market demand under the production process parameters.
[0100] Step S620: Based on the neural network, combine the historical physical property attribute information, the historical nutritional component index group, and the historical market demand to construct the quality detection model.
[0101] Specifically, the production process parameters are the key indices or basis for finding historical physical property information, historical nutritional component index groups, and historical market demands through specific databases, data storage systems, or query mechanisms. The neural network has powerful information recognition and fitting capabilities and can process complex data relationships. Inputting the above-mentioned different types of data into the neural network allows the neural network to learn the associations and potential patterns among them, thereby constructing a detection model that can accurately predict product quality based on newly input production process parameters, making the model's evaluation of product quality more comprehensive and accurate. Through this model, the quality status of the product can be understood in advance so as to take targeted measures in a timely manner to improve and optimize the product.
[0102] In a possible implementation manner, step S800 further includes:
[0103] Step S810: Construct a quality traceability database.
[0104] Step S820: Trace raw material information based on the quality performance result.
[0105] Step S830: Concatenate the raw material information, the production attribute information, the nutritional component index, and the quality performance result to generate a quality traceability record and store it in the quality traceability database.
[0106] Specifically, constructing a quality traceability database establishes a dedicated database for storing and managing product quality-related data. This database will become the basic resource for subsequent traceability and analysis, ensuring the centralized storage and orderly management of data for subsequent query, analysis, and traceability, and enabling it to trace and analyze the target dried fruit-related information obtained from the first market demand through the quality traceability database.
[0107] According to the obtained product quality performance result, using the association information in the database, trace back in reverse to the specific raw material information related to this quality, and query the detailed information such as the source, batch, and composition of the raw material to determine which batch of raw materials has an impact on the quality of the final product, quickly locate the source problem, and take measures in a timely manner.
[0108] By splicing raw material information, production attribute information, nutritional component indicators, and quality performance results together, a comprehensive and detailed quality traceability record is formed, providing a product file that can be consulted at any time, facilitating quality analysis, problem troubleshooting, and subsequent improvement work. The raw material information includes the specific conditions of various raw materials that make up the product, such as sources, batches, etc., which can help determine the basic composition of the product and the possible sources of problems; the production attribute information covers various parameters and characteristics during the production process; the production attribute information covers various parameters and characteristics during the production process, reflecting the nutritional value and characteristics of the product. The quality performance result is an evaluation and description of the final quality of the product.
[0109] Embodiment 2: Based on the same inventive concept as the method for controlling freeze-drying of fruits and vegetables based on multi-stage quality detection in the foregoing embodiment, as Figure 3 shown, the present application provides a control system for freeze-drying fruits and vegetables based on multi-stage quality detection, and the system includes:
[0110] A basic attribute information collection module 10 for freeze-drying fruits and vegetables, which is used to collect the basic attribute information of the fruits and vegetables to be detected for freeze-drying. The basic attribute information includes appearance attribute information, production attribute information, and physical property attribute information.
[0111] A first sample to be detected acquisition module 20, which is used to perform visual detection based on the appearance attribute information to obtain a first sample to be detected.
[0112] A connected chromatographic module 30, which is used to connect to a high-performance liquid chromatography module, perform nutritional component analysis on the first sample to be detected based on the high-performance liquid chromatography module, and obtain a nutritional component chromatogram. The nutritional component chromatogram contains N chromatographic peaks, and N is an integer greater than 0.
[0113] A nutritional component index group acquisition module 40, which is used to perform nutritional component quantification on the N chromatographic peaks to obtain a nutritional component index group.
[0114] A first market demand generation module 50, which is used to set a time window, collect the preference status of freeze-dried fruits and vegetables based on the time window, and generate a first market demand.
[0115] A quality detection model construction module 60, which is used to decompose the production attribute information, obtain the production process parameters of the fruits and vegetables to be detected for freeze-drying, and construct a quality detection model based on the production process parameters.
[0116] A quality performance result generation module 70, which is configured to input the physical property attribute information, the nutritional component index group, and the first market demand into the quality detection model for quality detection and generate a quality performance result.
[0117] A process parameter optimization and control module 80, which is configured to calculate a compliance rate deviation if the compliance rate of the quality performance result is lower than a quality threshold, retrieve an optimization strategy from an optimization strategy library based on the compliance rate deviation, and optimize and control the production process parameters based on the optimization strategy.
[0118] Furthermore, the system further includes:
[0119] A nutritional composition information retrieval module, which is configured to obtain a standard freeze-dried fruit and vegetable of the to-be-detected freeze-dried fruit and vegetable and retrieve the nutritional composition information of the standard freeze-dried fruit and vegetable.
[0120] A polarity judgment module, which is configured to perform polarity judgment based on the nutritional composition information and set mobile phase composition information and mobile phase flow rate based on the polarity judgment result.
[0121] A detector setting module, which is configured to set parameter information of the detector based on the nutritional composition information.
[0122] A high performance liquid chromatography configuration module, which is configured to configure a high performance liquid chromatography module based on the mobile phase composition information, the mobile phase flow rate, and the parameter information.
[0123] Furthermore, the system further includes:
[0124] A chromatographic peak acquisition module, which is configured to input the first to-be-detected sample into the high performance liquid chromatography module. The high performance liquid chromatography module separates the first to-be-detected sample based on the mobile phase in combination with a stationary phase and a partition coefficient. The detector receives the separation result, records a chromatogram, and obtains N chromatographic peaks.
[0125] Furthermore, the system further includes:
[0126] A database connection module, which is configured to connect to a fruit and vegetable nutritional component database. The fruit and vegetable nutritional component database contains multiple chromatographic peak areas and multiple fruit and vegetable nutritional components, and the multiple chromatographic peak areas correspond to the multiple fruit and vegetable nutritional components one by one.
[0127] A symmetry analysis module, which is configured to perform symmetry analysis on the N chromatographic peaks to obtain I symmetric chromatographic peaks and J asymmetric chromatographic peaks, where I + J = N.
[0128] A chromatographic peak area calculation module, configured to calculate the areas of the I symmetric chromatographic peaks based on the symmetric area formula to obtain I symmetric areas, and calculate the areas of the J asymmetric chromatographic peaks based on the asymmetric area formula to obtain J asymmetric areas.
[0129] Furthermore, the system further includes:
[0130] A first chromatographic peak acquisition module, configured to obtain a first chromatographic peak based on the N chromatographic peaks.
[0131] A tailing factor calculation module, configured to set the height of the tailing peak and calculate the tailing factor of the first chromatographic peak based on the height of the tailing peak.
[0132] If the tailing factor is 1, the first chromatographic peak is a symmetric chromatographic peak; if the tailing factor is not 1, the first chromatographic peak is an asymmetric chromatographic peak.
[0133] Furthermore, the system further includes:
[0134] An attribute information acquisition module, configured to retrieve and obtain historical physical property attribute information, a historical nutritional component index group, and historical market demand under the production process parameters based on the production process parameters.
[0135] A quality detection model construction module, configured to construct the quality detection model based on a neural network in combination with the historical physical property attribute information, the historical nutritional component index group, and the historical market demand.
[0136] Furthermore, the system further includes:
[0137] A quality traceability database construction module, configured to construct a quality traceability database.
[0138] A raw material information traceability module, configured to trace raw material information based on the quality performance result.
[0139] A quality traceability data storage module, configured to splice the raw material information, the production attribute information, the nutritional component index, and the quality performance result to generate a quality traceability record and store it in the quality traceability database.
[0140] It should be noted that the above-mentioned order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above description of specific embodiments of this specification has been made. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0141] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.
[0142] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.
Claims
1. A method for controlling freeze-drying of fruits and vegetables based on multi-stage quality detection, characterized in that: The method comprises: Collect basic attribute information of freeze-dried fruits and vegetables to be tested, wherein the basic attribute information includes appearance attribute information, production attribute information, and physical property attribute information; Perform visual inspection based on the appearance attribute information to obtain a first sample to be inspected; Connecting to a high performance liquid chromatography module, and performing a nutrient component analysis on the first sample to be detected based on the high performance liquid chromatography module to obtain a nutrient component chromatogram, wherein the nutrient component chromatogram comprises N chromatographic peaks, where N is an integer greater than 0; Quantifying the nutrients of the N chromatographic peaks to obtain a nutrient index group; Setting a time window, collecting freeze-dried fruit and vegetable preferences based on the time window, and generating a first market demand; Decomposing the production attribute information, obtaining the production process parameters of the freeze-dried fruits and vegetables to be tested, and constructing a quality detection model based on the production process parameters; Inputting the physical property attribute information, the nutritional component index group, and the first market demand into the quality detection model, performing quality detection, and generating quality performance results; If the compliance rate of the quality performance result is lower than the quality threshold, the compliance rate deviation is calculated, an optimization strategy library is searched based on the compliance rate deviation to obtain an optimization strategy, and the production process parameters are optimized and controlled based on the optimization strategy; The production attribute information is decomposed to obtain the production process parameters of the freeze-dried fruits and vegetables to be tested, and a quality detection model is constructed based on the production process parameters. The method includes: Based on the production process parameters, retrieve and obtain historical physical property attribute information, historical nutritional component index groups, and historical market demand under the production process parameters; The quality detection model is constructed based on a neural network and combined with the historical physical property attribute information, the historical nutritional component index group, and the historical market demand.
2. The method according to claim 1, characterized in that A high performance liquid chromatography module is connected, and a nutrient component analysis is performed on the first sample to be detected based on the high performance liquid chromatography module to obtain a nutrient component chromatogram, wherein the nutrient component chromatogram comprises N chromatographic peaks. The method comprises: Obtain standard freeze-dried fruits and vegetables for the freeze-dried fruits and vegetables to be tested, and retrieve nutritional composition information of the standard freeze-dried fruits and vegetables; Based on the nutritional composition information, polarity judgment is performed, and based on the polarity judgment result, mobile phase composition information and mobile phase flow rate are set; Based on the nutritional composition information, setting parameter information of the detector; A high performance liquid chromatography module is configured based on the mobile phase composition information, the mobile phase flow rate, and the parameter information.
3. The method according to claim 2, characterized in that The first sample to be detected is input into the HPLC module, and the HPLC module separates the first sample to be detected based on the mobile phase, in combination with the stationary phase and the distribution coefficient. The detector receives the separation result, records the chromatogram, and obtains N chromatographic peaks.
4. The method according to claim 1, characterized in that The N chromatographic peaks are quantified for nutrient components to obtain a nutrient component index group, and the method further comprises: Connecting to a fruit and vegetable nutritional component database, wherein the fruit and vegetable nutritional component database includes a plurality of chromatographic peak areas and a plurality of fruit and vegetable nutritional components, and the plurality of chromatographic peak areas correspond to the plurality of fruit and vegetable nutritional components one by one; Performing symmetrical analysis on the N chromatographic peaks to obtain I symmetrical chromatographic peak and J asymmetrical chromatographic peaks, wherein I+J=N; The areas of the I symmetrical chromatographic peaks are calculated based on the symmetrical area formula to obtain I symmetrical areas, and the areas of the J asymmetrical chromatographic peaks are calculated based on the asymmetrical area formula to obtain J asymmetrical areas.
5. The method according to claim 4, characterized in that The N chromatographic peaks are subjected to symmetrical analysis, the method comprising: Based on the N chromatographic peaks, obtaining a first chromatographic peak; Setting a tailing peak height, and calculating a tailing factor of the first chromatographic peak based on the tailing peak height; If the tailing factor is 1, the first chromatographic peak is a symmetrical chromatographic peak; if the tailing factor is not 1, the first chromatographic peak is an asymmetrical chromatographic peak.
6. The method according to claim 1, characterized in that The method comprises: Build a quality traceability database; Based on the quality performance results, trace the raw material information; The raw material information, the production attribute information, the nutritional component index, and the quality performance result are spliced to generate a quality traceability record, which is stored in the quality traceability number.
7. A fruit and vegetable freeze-drying control system based on multi-stage quality detection, characterized in that: The system is used to implement the steps of the fruit and vegetable freeze-drying control method based on multi-stage quality detection according to any one of claims 1 to 6, comprising: A module for collecting basic attribute information of freeze-dried fruits and vegetables, which is used to collect basic attribute information of freeze-dried fruits and vegetables to be tested, wherein the basic attribute information includes appearance attribute information, production attribute information, and physical property attribute information; A first sample-to-be-detected acquisition module, the first sample-to-be-detected acquisition module is used to perform visual inspection based on the appearance attribute information to obtain a first sample to be detected; A connection chromatography module, wherein the connection chromatography module is used to connect to a high performance liquid chromatography module, and based on the high performance liquid chromatography module, a nutrient component analysis is performed on the first sample to be detected to obtain a nutrient component chromatogram, wherein the nutrient component chromatogram comprises N chromatographic peaks, where N is an integer greater than 0; A nutrient index group acquisition module, the nutrient index group acquisition module is used to quantify the nutrient components of the N chromatographic peaks to obtain a nutrient index group; A first market demand generation module, the first market demand generation module is used to set a time window, collect freeze-dried fruit and vegetable preferences based on the time window, and generate a first market demand; A quality detection model construction module, the quality detection model construction module is used to decompose the production attribute information, obtain the production process parameters of the freeze-dried fruits and vegetables to be detected, and construct a quality detection model based on the production process parameters, wherein the quality detection model construction method includes: Based on the production process parameters, retrieve and obtain historical physical property attribute information, historical nutritional component index groups, and historical market demand under the production process parameters; Based on a neural network, the quality detection model is constructed by combining the historical physical property attribute information, the historical nutritional component index group, and the historical market demand; A quality performance result generating module, the quality performance result generating module is used to input the physical property attribute information, the nutritional component index group, and the first market demand into the quality detection model, perform quality detection, and generate a quality performance result; A process parameter optimization control module is used to calculate the standard compliance rate deviation if the standard compliance rate of the quality performance result is lower than the quality threshold, search the optimization strategy library based on the standard compliance rate deviation, obtain the optimization strategy, and optimize the production process parameters based on the optimization strategy.
Citation Information
Patent Citations
Quality detection method and comprehensive quality evaluation method for fresh-eating Lycium L. fruit
CN107246892A
Food flavor detection method and device, and electronic equipment
CN111695825A
Intelligent production control method and system for Chinese wolfberry flavor type Chinese wolfberry wine
CN114879615A
