Adaptive parameter control method for fish fillet air-drying production line
The adaptive parameter control method for fish fillet air-drying uses sensors and prediction models to optimize drying parameters, addressing inconsistent quality issues by ensuring stable and efficient air-drying through real-time environmental and fillet data integration.
Patent Information
- Application Number
- CN202411449023.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-17
- Publication Date
- 2024-12-27
- Estimated Expiration
- 2044-10-17
AI Technical Summary
The traditional air-drying process for fish fillets relies on operator experience, leading to inconsistent quality due to lack of real-time monitoring and adaptive control of drying parameters, resulting in over-drying or under-drying and increased production costs.
An adaptive parameter control method using sensors and a prediction model to optimize infrared heater and fan settings based on real-time environmental data and fish fillet characteristics, ensuring stable air-drying quality.
Ensures consistent air-drying quality by accurately adjusting parameters in response to environmental conditions, improving efficiency and reducing variability in fish fillet drying.
Smart Images
Figure CN118980228B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field related to data processing, and in particular to an adaptive parameter control method for a fish fillet air-drying production line. Background Art
[0002] Due to its low cost, hot air drying has become the most common dehydration process for air-dried fish fillets. However, the existing traditional air-drying process relies on the operator's experience to set the temperature and time to complete the drying process, resulting in poor stability in the air-dried fish fillet quality, which may lead to over-drying or under-drying; the traditional air-drying process cannot monitor the actual state of the working area environment in real time, and cannot adjust the operation of equipment such as heaters and fans in real time, which affects the stability of the air-dried fish fillet quality and may increase production costs; the traditional air-drying process only considers the quality of a single indicator after air-drying, such as the dry-wet weight ratio, and needs to rely on the operator's experience to monitor and adjust the drying parameters. It is easily affected by the operator's skills and subjective judgment, and cannot guarantee the stability of the air-dried fish fillet quality. Summary of the invention
[0003] The present application aims to solve the technical problem that it is difficult to ensure the stability of the quality of air-dried fish fillets in the prior art by providing an adaptive parameter control method for a fish fillet air-drying production line.
[0004] The present application discloses an efficient mixing control method for the production of puffed products, the method comprising: the method is applied to an intelligent control platform of an air-drying production line, the control platform is communicatively connected with a temperature and humidity sensor, a floor scale, an infrared heater and a circulating fan, the method comprising: receiving a delivered target fish fillet, obtaining basic information of the target fish fillet, performing a quality analysis on the target fish fillet, and if the quality analysis result is greater than a preset quality score threshold, delivering the target fish fillet to a floor scale of the air-drying production line; obtaining the initial weight of the target fish fillet and the initial environmental information of the air-drying area in the fish fillet air-drying production line through the floor scale and the temperature and humidity sensor, the initial environmental information comprising the initial temperature and the initial humidity; The air-drying quality prediction model is input with initial environmental information into the air-drying quality prediction model to obtain a multidimensional key parameter space set that affects the air-drying quality of the target fish fillet, wherein the air-drying quality prediction model is constructed based on the key process of the air-drying production line, and the key process of the air-drying production line is based on the process analysis of the traditional air-drying process for executing the target fish fillet, and is established by obtaining the step decomposition results; combined with the working environment information of the air-drying area, key parameters are optimized in the multidimensional key parameter space set, and the working environment information is the temperature and humidity monitored in real time in the air-drying area during the air-drying process; the optimization results are sent to the intelligent control platform, and the infrared heaters and circulating fans on the air-drying production line are automatically and adaptively controlled to complete the dehydration and air-drying of the target fish fillet.
[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0006] Obtaining the basic information of the target fish fillet helps to ensure the stability of the air-drying quality of the fish fillet; by using the floor scale and temperature and humidity sensors to obtain the initial weight of the target fish fillet and the initial environmental information of the air-drying area in the fish fillet air-drying production line, the influence of the air-drying production line environment and the initial weight of the target fish fillet on the air-drying quality of the fish fillet can be quantified more accurately; the initial weight, basic information and initial environmental information of the target fish fillet are input into the air-drying quality prediction model to obtain a multidimensional key parameter space set that affects the air-drying quality of the target fish fillet. Combined with a large amount of historical data, it is possible to comprehensively evaluate the influence of the key parameters of each step on the stability of the air-drying quality of the fish fillet, thereby improving the flexibility and responsiveness of the air-drying process; combined with the working environment information of the air-drying area, the key parameters are optimized in the multidimensional key parameter space set, and the optimization results are sent to the intelligent control platform to automatically and adaptively adjust the parameters of the infrared heater and the circulating fan on the air-drying production line to complete the dehydration and air-drying of the target fish fillet. According to the real-time environmental information of the air-drying area, the optimal working parameters of the heater and the fan on the air-drying production line can be accurately found to ensure the stability of the air-drying of the fish fillet. The method achieves a more efficient, more precise and more controllable fish fillet air-drying adaptability parameter regulation process, thereby improving product quality.
[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Figure 1 A schematic flow chart of an adaptive parameter control method for a fish fillet air-drying production line is provided for an embodiment of the present application. DETAILED DESCRIPTION
[0009] The embodiments of the present application provide an adaptive parameter control method for a fish fillet air-drying production line, thereby solving the technical problem that it is difficult in the prior art to accurately control the changes in the quality of air-dried fish fillets and the air-drying rate, resulting in poor quality and production stability of air-dried fish fillets.
[0010] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.
[0011] like Figure 1 As shown, the embodiment of the present application provides an adaptive parameter control method for a fish fillet air-drying production line, the method comprising:
[0012] Receive the delivered target fish fillet, obtain basic information of the target fish fillet, perform quality analysis on the target fish fillet, and if the quality analysis result is greater than a preset quality score threshold, deliver the target fish fillet to the air-drying production line weighing station.
[0013] Specifically, start the relevant data reading program or interface and connect to the database, file or sensor that stores the target fish fillet related information. Regarding the basic information of the target fish fillet, clarify the basic information of the target fish fillet, including the type, size, quantity, appearance and other basic information of the target fish fillet, as well as the basic information of the characteristics such as freshness and fat content. During the reading process, the acquired data is preliminarily checked for validity and format conversion to ensure the accuracy and availability of the data, prepare for subsequent processing and analysis, and help achieve accurate and stable adaptive parameter control of the fish fillet air-drying production line.
[0014] The initial weight of the target fish fillet and the initial environmental information of the air-drying area in the fish fillet air-drying production line are obtained respectively by the floor scale and the temperature and humidity sensor, wherein the initial environmental information includes the initial temperature and the initial humidity.
[0015] Specifically, ensure that the sensor has been properly installed in the air-drying work area of the air-drying production line and can monitor the temperature and humidity in the air-drying area in real time. Store the collected temperature and humidity data in the database and record the timestamp for subsequent analysis. The temperature and humidity in the air-drying work area of the air-drying production line not only affect the speed of air-drying, but also directly affect the quality of the final air-dried fish fillets. Excessive temperature may destroy some heat-sensitive nutrients, such as vitamins and certain unsaturated fatty acids, making the fish fillets too dry, hard and brittle, and losing their proper taste. Too low humidity may cause the fish fillets to be over-dried, and too high humidity may cause the fish fillets to be difficult to dry or even moldy. In actual operation, it is necessary to find a balance between temperature and humidity to ensure that the fish fillets can be effectively dried and maintain good quality. By measuring the initial weight of the fish fillets, its original moisture content can be determined, which is one of the important indicators to ensure the quality of air-drying of the fish fillets, thereby ensuring the consistency of the air-drying process and the uniformity of the final product. Taking the target fish fillet initial weight as one of the input parameters of the air-drying quality prediction model, combined with the basic information of the fish fillet and the initial environmental information of the air-drying working area, the quality of the final product can be predicted more accurately and the key parameter set of the air-drying production line can be extracted.
[0016] The initial weight, basic information and initial environmental information of the target fish fillet are input into the air-drying quality prediction model to obtain a multidimensional key parameter space set that affects the air-drying quality of the target fish fillet, wherein the air-drying quality prediction model is constructed based on the key process of the air-drying production line, and the key process of the air-drying production line is based on the process analysis of the traditional air-drying process for executing the target fish fillet, and is established by obtaining the step decomposition results.
[0017] Specifically, according to the type of target fish fillets, the corresponding air-drying quality standards and original air-drying processes are selected from the product database, the original air-drying process corresponding to the target fish fillets is extracted, and each air-drying step is preliminarily analyzed. For example, the original process is decomposed into steps such as preheating the air-drying chamber, turning on the infrared heater and running the circulating fan for heating, air-drying and cooling, and the specific operations and parameters of each step are obtained. Each step is further refined. For example, the heating and air-drying step can be further decomposed into sub-steps such as placing fish fillets, heating and air-drying, and testing. The specific operations and parameters of each sub-step are obtained. For example, the parameters in the heating and air-drying step include hot air speed, air-drying time, air-drying temperature, etc. The step decomposition results are formed after process analysis, and these results are used for subsequent analysis and optimization. Based on the process analysis of the air-drying process, the key parameters that have a significant impact on the quality of air-drying of the target fish fillets in each step are determined, including preheating process parameters and hot air air-drying process parameters. The preheating process parameters include preheating time, preheating temperature, preheating wind force, preheating air humidity and other parameters; the hot air air-drying process parameters include air-drying temperature, heating speed, air-drying wind force, air-drying humidity and other parameters.
[0018] Constructing the key process of the air-drying production line means determining a standardized production process and converting each step of the air-drying process into specific operations on the production line. This process will serve as the basis for optimization and adjustment. The key sequence includes all necessary air-drying steps, and each step has a clear operating guide and time frame. When constructing the key sequence of the air-drying production line, consider product quality standards, production efficiency goals, and the capabilities of existing equipment and resources. In order to more accurately adjust the adaptive parameters, it is necessary to collect the adjustment range of each device parameter in the air-drying production line, which can include parameters corresponding to multiple devices on the air-drying production line, such as equipment energy consumption parameters and equipment operation parameters. The adjustment range corresponding to the equipment energy consumption parameters can be from room temperature to the maximum heating capacity of the equipment, from the lowest power level of the equipment to the highest power level, and the heating time ranges from a few minutes to a few hours, depending on the heating task to be completed; the adjustment range corresponding to the equipment operation parameters can be the adjustment range of the equipment operation parameters determined based on historical data to ensure that the equipment will not fail to operate normally due to too low parameters, and to prevent equipment damage or increased energy consumption due to too high parameters.
[0019] Set up a prediction model, and perform predictive analysis on the status of the air-drying production line based on real-time monitoring data and air-drying quality standards. The prediction model can use machine learning, statistical analysis and other methods to build a robust model. Predict the operating status of the air-drying production line based on historical data and real-time monitoring data, including power demand, heat demand, wind demand, power loss, etc. According to the prediction results of the air-drying production line status, extract the key parameter sets that have a great impact on the final air-drying quality standards in each key process, and generate a multi-dimensional key parameter space set. This space set contains various parameter combinations to achieve optimal control of the air-drying production line, ensuring that the air-drying production line can achieve the optimal balance between efficiency and product quality under different air-drying environments.
[0020] In combination with the working environment information of the air-drying area, key parameter optimization is performed in the multi-dimensional key parameter space set, and the working environment information is the real-time temperature and humidity in the air-drying area during the air-drying process.
[0021] Specifically, confirming the state prediction and timing adaptive optimization strategy of the air-drying production line provides basic data and optimization direction for the dual optimization of temperature and humidity in the working area of the air-drying production line. Based on the monitored state of the air-drying production line, the system automatically adjusts the weight factors of the efficiency and quality strategies to determine the most appropriate optimization direction at the moment. Next, the system uses the timing adaptive optimization strategy to optimize the "key parameter set" and "key sequence of the air-drying production line" at the same time. During this process, the system will continuously try different parameter combinations and evaluate their impact on efficiency and quality. In each iteration, the system checks whether the newly proposed parameter combination meets the preset adjustable space. Only solutions that meet the constraints will be accepted and further evaluated.
[0022] The optimization results are sent to the intelligent control platform, and the infrared heaters and circulating fans on the air-drying production line are automatically and adaptively adjusted to complete the dehydration and air-drying of the target fish fillets.
[0023] Specifically, in order to improve the efficiency and quality of the air-drying process of air-dried fish products, it is necessary to optimize the temperature and humidity in the working area of the air-drying production line. The results, including the optimal set of key parameters, are sent to the intelligent control platform of the air-drying production line. The intelligent control platform is used to control the key operating parameters of the infrared heaters and circulating fans on the air-drying production line. The infrared heaters and circulating fans are adaptively adjusted to coordinate the drying efficiency and drying quality to ensure that the infrared heaters and circulating fans are always in the optimal state during the drying process of the target fish fillets, meeting the high quality and high efficiency requirements of the target fish fillets.
[0024] Furthermore, the target fish fillet is subjected to quality analysis, and if the quality analysis result is greater than a preset quality score threshold, the target fish fillet is transported to a weighing scale of an air-drying production line, including:
[0025] Connect to the product database to obtain basic information of the target fish fillet, including the type, size, appearance, freshness and fat content of the target fish fillet;
[0026] Input the acquired basic information into the quality analysis model, perform quality analysis, and obtain quality analysis results;
[0027] If the quality analysis result is greater than the preset quality score threshold, the target fish fillet is transported to the air-drying production line scale.
[0028] Specifically, start the relevant data reading program or interface and connect to the database, file or sensor storing the target fish fillet related information. As for the basic information of the target fish fillet, clarify the basic information of the target fish fillet, including the type, size, weight, quantity and other basic information of the target fish fillet, as well as the basic information of the characteristics such as freshness index, initial moisture content, fat content, etc. During the reading process, perform preliminary validity check and format conversion on the acquired data to ensure the accuracy and availability of the data, and prepare for subsequent processing and analysis.
[0029] Furthermore, the quality analysis model includes:
[0030] Read the historical information of the target fish fillet and extract the historical basic information required for quality analysis;
[0031] Evaluate the degree of influence of historical basic information on the air-dried quality of target fish fillets, determine the scoring range and standard for each historical basic information, and assign weights to them;
[0032] Multiply each historical basic information score by its corresponding weight and sum it up to get the target fish fillet quality analysis.
[0033] score.
[0034] Specifically, the fish fillet quality analysis model is constructed based on the historical basic information of the target fish fillets, and a series of objective and subjective evaluation methods can be used for comprehensive evaluation. Freshness can be selected as a measure of quality analysis because it directly affects the taste, nutritional value and food safety of the product. Freshness can be measured in many aspects, including sensory characteristics (such as appearance, smell, texture) and physical and chemical properties (such as pH value, volatile basic nitrogen TVB-N content, etc.). According to the degree of influence of each indicator on freshness, a weight is assigned to it. The physical and chemical characteristics have the greatest impact on the air-dried quality and match a larger weight. A specific scoring range and standard are set for each indicator. TVB-N content and microbial load: set a threshold according to international or national standards. If the threshold is exceeded, the batch of target fish fillets will be transported to the weighing scale of the air-drying production line.
[0035] Furthermore, the key processes of the air-drying production line include:
[0036] Read the air-dried quality standard of the target fish fillet, which is the quality mark achieved by the target fish fillet after air drying, including: the range of air-dryness, water content, dry-to-wet weight ratio and loss ratio;
[0037] Read the historical control records of the air-drying production line, explore the production line factors that directly affect the air-drying quality standards, and explore the frequency at which the standards meet the preset standards;
[0038] A cause-effect analysis is conducted on the factors affecting the air-drying production line, the constraint relationship of the factors affecting the production line is obtained, and the key processes of the air-drying production line are established.
[0039] Specifically, the production line influencing factors are obtained according to the air-drying quality standard of the target fish fillets, including the process steps and process parameter sets that affect the air-drying quality standard. The process steps with high adjustability and greater impact on the air-drying quality of the target fish fillets are selected as key steps. The key steps are sorted according to the process constraints to form the key processes of the air-drying production line. The key processes of the air-drying production line are the basis for optimization and adjustment.
[0040] Determine which steps can be inserted or adjusted in the key processes of the air-drying production line, set insertion rules and constraints based on actual production conditions and process requirements, such as the position where each step can be inserted, the process constraints after insertion, etc. According to the insertion rules and constraints, insert other steps one by one in the key processes of the air-drying production line, conduct traversal analysis, establish an insertion connection space, evaluate the impact of each insertion position on the air-drying quality of the target fish fillet, retain the insertion combinations with large parameter adjustability and greater impact on the air-drying quality, and form the key processes of the air-drying production line for subsequent optimization.
[0041] The air-drying quality standard is used to ensure the quality and consistency of the air-drying of the target fish fillet. Taking the air-drying quality standard as the control target, the quality impact of each step in the inserted solution space on the air-drying quality standard is evaluated, and the steps with large parameter adjustability and greater impact on the air-drying quality that meet the requirements of the air-drying quality standard are selected from the inserted solution space, and the air-drying production line benchmark sequence is established. The air-drying production line benchmark sequence includes the key processes of the air-drying production line and all steps that meet the air-drying quality standard and have feasible processes.
[0042] Furthermore, the air-drying quality prediction model includes:
[0043] Digital modeling is performed on the key processes of the air-drying production line to establish a digital model, wherein the digital model is constructed through historical data and mathematical modeling;
[0044] Obtaining historical data sets of each step in the key process of the air-drying production line, performing multivariate application scenario simulation through the digital model, extracting key parameters of the air-drying production line based on the simulation results, constructing a key parameter set, and establishing a weight coefficient of the key parameter set, wherein the simulation results meet the air-drying quality standard of the target fish fillet;
[0045] Perform adaptation analysis and reconstruction of constraint weights through weight coefficients and adaptation analysis functions;
[0046] The key parameter space of each key parameter is constructed based on the constraint weight to form a multi-dimensional key parameter space set.
[0047] Specifically, the actual use data sets of each step in the key process of the air-drying production line are collected, including different heating speeds, air-drying temperatures, air-drying wind speeds, air-drying time, and air-drying humidity, and the performance of resource consumption that changes with the adjustment of process parameters, including evaluation indicators such as time consumption, power consumption, and energy consumption. Based on the established digital model, the collected actual use data sets are simulated through simulation software, and the simulation results are analyzed to extract the process parameters that affect the air-drying quality standard of the target fish fillet. According to the simulation results, which process parameters have the greatest impact on the air-drying quality standard of the target fish fillet, and the parameter adjustable range is large, so as to determine the key parameter set. According to the importance of the key parameter set, a weight coefficient is established for each parameter, and the weight coefficient reflects the contribution of different parameters to maintaining the stability of the air-drying quality standard. By statistically analyzing historical data, the normal range of the air-drying quality standard is set in combination with production practice. In combination with multiple application scenarios, the initial environment of the air-drying production line is built by reading the historical initial weight, historical basic information, and historical initial environmental information of the target fish fillet, simulating the air-drying heating process, and extracting the key parameter set of the air-drying production line. Combining expert knowledge and actual production experience, a reasonable range of key sensitive parameters is set, and a network search is used to generate a series of parameter combinations within the defined parameter range. These parameter combinations are predicted using the trained model to determine which parameter ranges have the greatest impact on the air-drying quality standards and are identified as key parameters. The relationship between the key parameter range and the air-drying quality standards is analyzed, and the optimal parameter range is determined based on the optimization objectives (such as maximizing air-dryness, minimizing moisture content, etc.).
[0048] A digital model was established, and the regression model was trained using historical data and validated. The regression coefficient was used to quantify the influence of each process parameter on the air-drying quality standard, identify the distribution characteristics of the air-drying quality standard in each application scenario, extract the process parameters of each application scenario whose distribution characteristics fall within the normal range, and use the multivariate regression method to perform sensitivity analysis. According to the absolute value of the sensitivity coefficient, the adaptation analysis function of the key parameters to the air-drying quality standard was identified.
[0049] Furthermore, the adaptation analysis function includes:
[0050] Constructing a sensitive parameter prediction model for air-drying quality, wherein the sensitive parameter prediction model is a multi-layer fully connected neural network structure, and each network layer is configured with a regression analysis algorithm whose dependent variable is an air-drying quality standard;
[0051] Read historical data and determine key parameter set samples as training samples;
[0052] Based on the training samples, supervised training is performed on the model architecture to generate the sensitive parameter prediction model;
[0053] Based on the training samples, the key parameter set samples are verified, and the model convergence test and retraining are performed until the convergence conditions are met, and the constructed sensitive parameter prediction model is obtained;
[0054] The key parameters are input into a sensitive parameter prediction model, the sensitivity of the key parameter set to the air-drying quality standard is analyzed, and an adaptation analysis function is obtained.
[0055] Specifically, the historical database is read, and the sample data of the key parameter set is used as a training sample. The training sample is the basic data for training and verifying the sensitive parameter prediction model for air-drying quality. The parameters of the sensitive parameter prediction model for air-drying quality are adjusted through the training samples, so that it can better adapt to identifying the parameters that can most affect the air-drying quality standards, and correctly input the key parameter set into the corresponding multi-task learning model for multiple sensitive analysis layers for processing. By calculating the degree of influence of each input feature on each output layer, the parameters that have the greatest impact on the air-drying quality standard can be identified. Use the training samples to verify the sensitive analysis model and test the performance, including convergence tests, to ensure that the sensitive analysis model gradually stabilizes during the training process. If the sensitive analysis model does not meet the convergence conditions during the verification process, retraining is performed. This involves performing the training steps again and adjusting the sensitive analysis model parameters until the model meets the predetermined performance and convergence conditions.
[0056] Furthermore, a key parameter space of each key parameter is constructed based on the constraint weight, including:
[0057] Establishing a mapping relationship between each key parameter in the key parameter set and each device on the air-drying production line;
[0058] Traversing the device parameters to extract adjustment ranges and generate first adjustable ranges for each key parameter;
[0059] Using the constraint weight as a limiting factor, a parameter adjustable identification network layer is used to identify the first adjustable range of each key parameter, and a second adjustable range of each key parameter is generated;
[0060] According to the second adjustable range of each key parameter, a parameter space of each key parameter is formed, and a multi-dimensional key parameter space set is constructed.
[0061] Specifically, analyze each key process on the air-drying production line, such as heaters, humidity regulators, fans, etc., clarify their corresponding key parameters, and establish mapping relationships. Based on the equipment specifications, determine a preliminary adjustable range for each key parameter, namely the first adjustable range. According to the simulation results of multiple application scenarios and the requirements of the air-drying process, define the constraint weight for each key parameter, and the weight reflects the priority and restrictions when adjusting the parameter. Through the parameter adjustable identification network layer, consider the constraint weight, identify and correct the first adjustable range, generate a second adjustable range that is more in line with actual production needs, and generate the key parameter space of the key parameter based on the second adjustable range. Each dimension represents the adjustable range of a key parameter. The parameter adjustable identification network layer is a neural network model in machine learning that can continuously perform self-iterative optimization. The parameter adjustable identification network layer is obtained through training data sets and supervision data sets. Each set of training data in the training data set includes the first adjustable range of key parameters and weight limit factors. The supervision data set is the supervision data that corresponds one-to-one with the training data set.
[0062] Construct a parameter-adjustable recognition network layer: input each set of training data in the training data set into the parameter-adjustable recognition network layer, and use the supervisory data corresponding to this set of training data to perform supervisory adjustment on the output of the parameter-adjustable recognition network layer. When the output result of the parameter-adjustable recognition network layer is consistent with the supervisory data, the training of the current set is completed. When all the training data in the training data set are trained, the training of the parameter-adjustable recognition network layer is completed.
[0063] In order to ensure the accuracy of the parameter-adjustable recognition network layer, the parameter-adjustable recognition network layer can be tested using a test data set. For example, the test accuracy can be set to 80%. When the test accuracy of the test data set meets 80%, the parameter-adjustable recognition network layer is constructed.
[0064] Furthermore, in combination with the working environment information of the air-drying area, key parameter optimization is performed in the multi-dimensional key parameter space set, including:
[0065] Establishing synchronization constraints of key parameter sets of the air-drying production line based on the real-time temperature and humidity;
[0066] By synchronously constraining the key parameter set of the air-drying production line, the optimization in the multi-dimensional key parameter space set is performed, and parameter adaptation is adjusted, and the fitness value is recalculated according to the parameter adaptation adjustment result;
[0067] Calculate the key parameter fitness values based on all updated fitness values.
[0068] Specifically, a time-series adaptive optimization strategy is established to establish a fitness function for each item in the air-drying quality standard. The fitness function integrates factors such as efficiency and stability, and can determine the fitness of the key parameter set according to the temperature, humidity and factor weights of the air-drying production line working area. By analyzing the requirements of the temperature and humidity of the air-drying production line working area for key parameters, the key parameters of the air-drying production line are optimized. This optimization process is carried out for the decomposed situation under various states of the air-drying production line working area, and the goal is to maximize the fitness of the system under different states. The results of the control parameter optimization are analyzed in a collaborative optimization process with temperature and humidity. This analysis process integrates the key parameters of each device under different states of the production line working area, as well as the synergy and influence between the devices. According to the results of the collaborative optimization analysis, the final control parameter optimization results are obtained. This result covers the system fitness, stability and efficiency under different production environments, thereby determining the optimal control parameters of each device. These results reflect the performance of the system under different schemes, and the optimal optimization scheme is selected accordingly to achieve the best control and operation of the system.
[0069] In summary, the adaptive parameter control method for the fish fillet air-drying production line provided in the embodiment of the present application has the following technical effects:
[0070] Obtaining the basic information of the target fish fillet helps to ensure the stability of the air-drying quality of the fish fillet; by using the floor scale and temperature and humidity sensors to obtain the initial weight of the target fish fillet and the initial environmental information of the air-drying area in the fish fillet air-drying production line, the influence of the air-drying production line environment and the initial weight of the target fish fillet on the air-drying quality of the fish fillet can be quantified more accurately; the initial weight, basic information and initial environmental information of the target fish fillet are input into the air-drying quality prediction model to obtain a multidimensional key parameter space set that affects the air-drying quality of the target fish fillet. Combined with a large amount of historical data, it is possible to comprehensively evaluate the influence of the key parameters of each step on the stability of the air-drying quality of the fish fillet, thereby improving the flexibility and responsiveness of the air-drying process; combined with the working environment information of the air-drying area, the key parameters are optimized in the multidimensional key parameter space set, and the optimization results are sent to the intelligent control platform to automatically and adaptively adjust the parameters of the infrared heater and the circulating fan on the air-drying production line to complete the dehydration and air-drying of the target fish fillet. According to the real-time environmental information of the air-drying area, the optimal working parameters of the heater and the fan on the air-drying production line can be accurately found to ensure the stability of the air-drying of the fish fillet. The method achieves a more efficient, more precise and more controllable fish fillet air-drying adaptability parameter regulation process, thereby improving product quality.
[0071] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. An adaptive parameter control method for a fish fillet air-drying production line, characterized in that: The method is applied to an intelligent control platform of an air-drying production line, wherein the control platform is communicatively connected with a temperature and humidity sensor, a floor scale, an infrared heater, and a circulating fan, and the method comprises: Receive the delivered target fish fillet, obtain basic information of the target fish fillet, perform quality analysis on the target fish fillet, and if the quality analysis result is greater than a preset quality score threshold, deliver the target fish fillet to the weighing station of the air-drying production line; The initial weight of the target fish fillet and the initial environmental information of the air-drying area in the fish fillet air-drying production line are respectively obtained by the floor scale and the temperature and humidity sensor, wherein the initial environmental information includes the initial temperature and the initial humidity; Input the initial weight, basic information and initial environmental information of the target fish fillet into the air-drying quality prediction model to obtain a multidimensional key parameter space set that affects the air-drying quality of the target fish fillet, wherein the air-drying quality prediction model is constructed based on the key process of the air-drying production line, and the key process of the air-drying production line is based on the process analysis of the traditional air-drying process for executing the target fish fillet, and is established by obtaining the step decomposition results; In combination with the working environment information of the air-drying area, key parameter optimization is performed in the multi-dimensional key parameter space set, wherein the working environment information is the temperature and humidity monitored in real time in the air-drying area during the air-drying process; The optimization results are sent to the intelligent control platform, and the infrared heaters and circulating fans on the air-drying production line are automatically and adaptively adjusted to complete the dehydration and air-drying of the target fish fillets.
2. The adaptive parameter control method for the fish fillet air-drying production line according to claim 1, characterized in that: The target fish fillet is subjected to quality analysis, and if the quality analysis result is greater than a preset quality score threshold, the target fish fillet is transported to a weighing scale of an air-drying production line, including: Connect to the product database to obtain basic information of the target fish fillet, including the type, size, appearance, freshness and fat content of the target fish fillet; Input the acquired basic information into the quality analysis model, perform quality analysis, and obtain quality analysis results; If the quality analysis result is greater than the preset quality score threshold, the target fish fillet is transported to the air-drying production line scale.
3. The adaptive parameter control method for the fish fillet air-drying production line according to claim 2, characterized in that: The quality analysis model comprises: Read the historical information of the target fish fillet and extract the historical basic information required for quality analysis; Evaluate the degree of influence of historical basic information on the air-dried quality of target fish fillets, determine the scoring range and standard for each historical basic information, and assign weights to them; The basic information scores of the target fish fillet are multiplied by their corresponding weights, and the sum is calculated to obtain the target fish fillet quality analysis score.
4. The adaptive parameter control method for the fish fillet air-drying production line according to claim 1, characterized in that: The key processes of the air-drying production line include: Read the air-dried quality standard of the target fish fillet, which is the quality mark achieved by the target fish fillet after air drying, including: the range of air-dryness, water content, dry-to-wet weight ratio and loss ratio; Read the historical control records of the air-drying production line, explore the production line factors that directly affect the air-drying quality standards, and explore the frequency at which the standards meet the preset standards; A cause-effect analysis is conducted on the factors affecting the air-drying production line, the constraint relationship of the factors affecting the production line is obtained, and the key processes of the air-drying production line are established.
5. The adaptive parameter control method for the fish fillet air-drying production line according to claim 4, characterized in that: The air-drying quality prediction model comprises: Digital modeling is performed on the key processes of the air-drying production line to establish a digital model, wherein the digital model is constructed through historical data and mathematical modeling; Obtaining historical data sets of each step in the key process of the air-drying production line, performing multivariate application scenario simulation through the digital model, extracting key parameters of the air-drying production line based on the simulation results, constructing a key parameter set, and establishing a weight coefficient of the key parameter set, wherein the simulation results meet the air-drying quality standard of the target fish fillet; Perform adaptation analysis and reconstruction of constraint weights through weight coefficients and adaptation analysis functions; The key parameter space of each key parameter is constructed based on the constraint weight to form a multi-dimensional key parameter space set.
6. The adaptive parameter control method for the fish fillet air-drying production line according to claim 5, characterized in that: The adaptation analysis function comprises: Constructing a sensitive parameter prediction model for air-drying quality, wherein the sensitive parameter prediction model is a multi-layer fully connected neural network structure, and each network layer is configured with a regression analysis algorithm whose dependent variable is an air-drying quality standard; Read historical data and determine key parameter set samples as training samples; Based on the training samples, supervised training is performed on the model architecture to generate the sensitive parameter prediction model; Based on the training samples, the key parameter set samples are verified, and the model convergence test and retraining are performed until the convergence conditions are met, and the constructed sensitive parameter prediction model is obtained; The key parameters are input into a sensitive parameter prediction model, the sensitivity of the key parameter set to the air-drying quality standard is analyzed, and an adaptation analysis function is obtained.
7. The adaptive parameter control method for the fish fillet air-drying production line according to claim 5, characterized in that: The key parameter space of each key parameter is constructed based on the constraint weights, including: Establishing a mapping relationship between each key parameter in the key parameter set and each device on the air-drying production line; Traversing the device parameters to extract adjustment ranges and generate first adjustable ranges for each key parameter; Using the constraint weight as a limiting factor, a parameter adjustable identification network layer is used to identify the first adjustable range of each key parameter, and a second adjustable range of each key parameter is generated; According to the second adjustable range of each key parameter, a parameter space of each key parameter is formed, and a multi-dimensional key parameter space set is constructed.
8. The adaptive parameter control method for the fish fillet air-drying production line according to claim 1, characterized in that: In combination with the working environment information of the air-drying area, key parameter optimization is performed in the multi-dimensional key parameter space set, including: Establishing synchronization constraints of key parameter sets of the air-drying production line based on the real-time temperature and humidity; By synchronously constraining the key parameter set of the air-drying production line, the optimization in the multi-dimensional key parameter space set is performed, and parameter adaptation is adjusted, and the fitness value is recalculated according to the parameter adaptation adjustment result; Calculate the key parameter fitness values based on all updated fitness values.