Efficient Mixing Control Method for Puffed Product Production
By establishing a product database and calling history production records, and optimizing the mixing parameters with the adaptation analysis function, the problems of poor proportion control and stability in the production of puffed products are solved, efficient mixing control is achieved, and product quality and production stability are improved.
Patent Information
- Application Number
- CN202411253756.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-09
AI Technical Summary
Traditional methods are difficult to achieve accurate proportion control and stable production process in the production of puffed products, resulting in poor product quality and production stability.
By establishing a product database, performing process analysis, calling historical production records for step tunability evaluation, establishing a benchmark mixing chain, configuring adaptation analysis functions, generating a mixing solution set, and optimizing mixing parameters to achieve accurate mixing control.
Achieve more precise mixing steps impact quantification, improve the flexibility and responsiveness of the production process, and find the optimal mixing solution that takes into account both production efficiency and product quality.
Smart Images

Figure CN118761687B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of mixing control, and in particular to a high-efficiency mixing control method for producing puffed products. Background Art
[0002] Puffed products are a type of food that expands raw materials under high temperature and high pressure conditions, such as puffed cereals and puffed corn. These products are widely used in food processing and production. However, traditional methods often rely on experience or simple trial and error to optimize the mixing process, which makes it difficult to achieve precise ratio control and stability of the production process. In addition, traditional mixing scheme optimization methods are limited by parameter settings and a single optimization goal, making it difficult to fully consider the comprehensive optimization of product quality and production efficiency, resulting in poor final product quality. Summary of the invention
[0003] The present application provides an efficient mixing control method for the production of puffed products, aiming to solve the technical problem that the prior art often relies on experience or simple trial and error methods to optimize the mixing process, making it difficult to achieve precise ratio control and stable production control, resulting in poor production quality and production stability.
[0004] The present application discloses an efficient mixing control method for the production of puffed products, the method comprising: obtaining a product database of puffed products, the product database comprising target ratio information and an original mixing process; performing process analysis of the original mixing process in the product database to establish a step decomposition result; calling historical production records of puffed products, inputting the historical production records into an adjustment analysis network, performing step adjustability evaluation on the step decomposition result, and generating a step adjustability evaluation result; establishing a benchmark mixing chain according to the step adjustability evaluation result, and performing insertion traversal analysis with the benchmark mixing chain to establish an insertion solution space, and establishing a mixing scheme set based on the insertion solution space and the benchmark mixing chain with the target ratio as the control target; configuring an adaptation analysis function, performing adaptation analysis on the mixing scheme set with the adaptation analysis function to generate an adaptation analysis result, the adaptation analysis function comprising a quality analysis function and a consumption analysis function; configuring mixing parameters according to the adaptation analysis result, and performing mixing control of puffed product production according to the mixing parameters and the adaptation analysis result.
[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0006] A product database is established based on the target ratio information and the original mixing process, and the process analysis of the original mixing process in the product database is performed to establish detailed step decomposition results. This can more accurately quantify the impact and correlation of each mixing step, thereby optimizing the overall mixing process; historical production records are called, and the adjustment analysis network is used to evaluate the step adjustability. This method combines a large amount of historical data and can comprehensively evaluate the impact of each step on the quality and efficiency of the final product, thereby improving the flexibility and responsiveness of the production process; a benchmark mixing chain is established based on the step adjustability evaluation results, and the benchmark mixing chain is used for insertion traversal analysis to establish an insertion solution space, and an adaptation analysis function is configured to optimize the mixing scheme set. This method utilizes the comprehensive evaluation of the adaptation analysis function and integrates the weight factors of quality analysis and consumption analysis. It can effectively find the optimal mixing scheme while taking into account production efficiency and 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 efficient mixing control method for producing puffed products is provided for the embodiments of the present application;
[0009] Figure 2 A schematic flow chart of the adjustability evaluation results of the generation steps in the efficient mixing control method for the production of puffed products is provided for the embodiments of the present application. DETAILED DESCRIPTION
[0010] The embodiments of the present application provide an efficient mixing control method for the production of puffed products, thereby solving the technical problem that the prior art often relies on experience or simple trial and error methods to optimize the mixing process, making it difficult to achieve precise ratio control and stable production control, resulting in poor production quality and production stability.
[0011] 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.
[0012] like Figure 1 As shown, the embodiment of the present application provides an efficient mixing control method for the production of puffed products, the method comprising:
[0013] A product database of puffed products is obtained, wherein the product database includes target ratio information and original mixing process.
[0014] Puffed products refer to foods that expand raw materials and form specific structures through high temperature, high pressure or other methods, such as popcorn and potato chips. The target ratio information and original mixing process are obtained from the production records, industry standards and other data sources within the enterprise. The target ratio information includes the raw materials required for the production of various puffed products and their specific proportions, and the original mixing process includes traditional mixing steps and their parameters, including mixing order, mixing time, temperature, etc. The obtained target ratio information and original mixing process are integrated to obtain a product database.
[0015] Perform process analysis of the original mixing process in the product database and establish step decomposition results.
[0016] The original mixing process is extracted from the product database, and each mixing step is preliminarily analyzed. For example, the original process is decomposed into steps such as mixing, heating, and cooling, and the specific operations and parameters of each step are obtained. Each step is further refined. For example, the mixing step can be further decomposed into sub-steps such as feeding, stirring, and testing, and the specific operations and parameters of each sub-step are obtained. For example, the parameters in the stirring step include stirring speed, stirring time, stirring temperature, etc. After process analysis, the step decomposition results are formed, and these results are used for subsequent analysis and optimization.
[0017] The historical production records of the puffed product are called, and the historical production records are input into the adjustment analysis network to perform step adjustability evaluation on the step decomposition results, and generate step adjustability evaluation results.
[0018] The historical production records of puffed products recorded in the past production process are retrieved and the adjustment analysis network is called. The network adopts a multi-layer neural network model, which contains multiple processing layers, and each layer performs different analysis tasks. The step adjustability evaluation is to evaluate the degree of influence of the production process steps on the overall process and product quality during the adjustment process, so as to determine which steps can be adjusted and how to adjust them.
[0019] The first-layer processing network of the adjustment analysis network is called to perform the minimum impact evaluation of the adjustment of each step in the historical production record, and the first adjustability evaluation result is established. This layer is used to evaluate the minimum impact of each step on the overall process during the adjustment process; the second-layer processing network of the adjustment analysis network is called to perform the balanced impact evaluation of each step in the historical production record, and the second adjustability evaluation result is established. This layer is used to evaluate the impact of each step on the process balance during the adjustment process. The first adjustability evaluation result and the second adjustability evaluation result are weighted calculations, and the minimum adjustment impact and balanced impact of each step are combined to generate the final step adjustability evaluation result.
[0020] Furthermore, if Figure 2As shown, the inputting of the historical production records into the adjustment analysis network, performing step adjustability evaluation on the step decomposition results, and generating step adjustability evaluation results also includes:
[0021] The first layer processing network in the adjustment analysis network is called to perform the minimum impact evaluation of the adjustment of each step in the historical production record, and establish a first adjustability evaluation result; the second layer processing network in the adjustment analysis network is called to perform the balanced impact evaluation of each step in the historical production record, and establish a second adjustability evaluation result; the first adjustability evaluation result and the second adjustability evaluation result are subjected to attention weighted calculation to generate a step adjustability evaluation result.
[0022] Historical production records include detailed data for each step in the production process of puffed products, including time, temperature, pressure, mixing ratio, etc. The minimum impact assessment is used to evaluate the impact of fine-tuning each step in the production process on the entire production process and the final product. The goal is to find the adjustment method with the least impact on production.
[0023] The historical production records are input into the first-layer processing network, and the key features of each production step are extracted through the first-layer processing network. Based on the extracted features, the impact of each step on the overall production process and the final product when adjusted is evaluated. Specifically, the adjustment feasibility of each production step is evaluated to find the step that is easiest to adjust and has the least impact on the production process, including the difficulty of adjustment, the impact on other steps, the impact on the quality of the final product, etc. Exemplarily, fine-tuning of each step is simulated in the model to obtain its impact on the entire production process. Based on the simulation results, the impact of the adjustment on the production process and the quality of the final product is evaluated, and the step with the least impact on the overall production process when adjusted is found. The adjustment impact results of each step are summarized to generate a first adjustability evaluation result for subsequent adjustment analysis.
[0024] Balance impact assessment is used to evaluate the impact of each production step on the balance of the overall production process when it is adjusted. The purpose is to ensure the stability and consistency of the production process and avoid local adjustments from having adverse effects on the overall process.
[0025] The historical production records are input into the second-layer processing network, and each production step is subjected to in-depth feature extraction through the second-layer processing network to extract associated features. The impact of the adjustment of each step on the balance of the entire production process is analyzed based on the associated features to ensure the stability and consistency of the adjusted production process. For example, fine-tuning of each step is simulated in the model to obtain its impact on the balance of the overall production process. Based on the simulation results, the impact of the adjustment on the stability and consistency of the production process is evaluated to find the step with the least impact on the balance of the overall production process when making adjustments. The balance impact results of each step are summarized to generate a second adjustability evaluation result for subsequent adjustment analysis.
[0026] Different weights are assigned to each evaluation result based on the actual needs and objectives of the production process. For example, if balance is more important to production, the weight of the second adjustability evaluation result can be set higher. The first adjustability evaluation result and the second adjustability evaluation result are weighted and summed according to the set weights to generate a comprehensive step adjustability evaluation result.
[0027] A benchmark mixing chain is established according to the adjustability evaluation results of the steps, and an insertion traversal analysis is performed on the benchmark mixing chain to establish an insertion solution space. Based on the insertion solution space and the benchmark mixing chain, a mixing scheme set is established with the target ratio as the control target.
[0028] According to the step adjustability evaluation results, select the steps with high adjustability and less impact on the production process as the benchmark steps. Arrange the selected benchmark steps in accordance with the process sequence to form a benchmark mixing chain. The benchmark mixing chain is the basis for optimization and adjustment.
[0029] Determine which steps can be inserted or adjusted in the benchmark mixing chain, 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 benchmark mixing chain, perform traversal analysis, evaluate the impact of each insertion position on the target ratio and the overall process, record all feasible insertion combinations, and form an insertion solution space. The insertion solution space includes all feasible mixing process solutions for subsequent optimization.
[0030] The target ratio is the target raw material and its specific target ratio for the production of puffed products. It is obtained according to actual needs to ensure product quality and consistency. Taking the target ratio as the control target, evaluate the realization of the target ratio for each solution in the inserted solution space, select solutions that meet the target ratio requirements from the inserted solution space, and establish a mixed solution set, which includes all solutions that meet the target ratio and are process-feasible.
[0031] Furthermore, the insert traversal analysis is performed on the benchmark hybrid chain to establish an insert solution space, and further includes:
[0032] The benchmark step in the benchmark mixed chain is used as the matching step, and the active step association matching of the non-benchmark mixed chain is performed; an association threshold is set, and if the association matching result meets the association threshold, the order constraint of the corresponding benchmark step and the active step is established; according to all the order constraints, the insertion traversal analysis is completed to establish the insertion solution space.
[0033] The benchmark steps in the benchmark mixing chain are used as matching steps. The non-benchmark mixing chain refers to other production step sequences containing similar but not identical steps or sequences, which are used to compare and optimize the production process. Each benchmark step in the benchmark mixing chain is associated and matched with the corresponding active step in the non-benchmark mixing chain. The purpose of matching is to identify steps in the non-benchmark mixing chain that have similar effects to the benchmark steps for subsequent analysis and comparison.
[0034] The association matching method includes functional comparison, for example, comparing the functions and operations of the benchmark steps and the active steps to confirm whether they can achieve similar production effects; it also includes effect evaluation, for example, analyzing the actual effect and impact of each step to determine whether it is suitable for the corresponding step in the benchmark mixing chain; it also includes sequence constraints, for example, according to the sequence requirements in the production process, confirming whether the active steps can be matched with the benchmark steps in the same sequence.
[0035] Through association matching, each benchmark step is mapped to a suitable activity step to form an association relationship between the benchmark step and the non-benchmark mixed chain activity step. The association relationship between each benchmark step and the activity step is evaluated to obtain the matching compliance.
[0036] The association threshold is a pre-set standard used to determine whether the association matching result between the reference step and the active step meets the requirements. It is set according to the needs and technical requirements of the production process. If the association matching result between a reference step and the active step meets the set association threshold, that is, the matching degree meets the preset threshold, then the corresponding sequence constraint can be established. The sequence constraint indicates that these steps are executed in a specific order to ensure the stability of the production process and product quality.
[0037] The order constraints established between all benchmark steps and active steps are summarized and inserted into the traversal analysis. The insertion traversal analysis is a method used to explore and generate possible production step sequences to achieve production goals under the premise of considering the order constraints. Multiple possible step sequences are generated through the insertion traversal analysis, each of which meets the pre-set order constraints, so as to establish the insertion solution space. The insertion solution space refers to the set of all possible step sequences that meet the order constraints, which can be used as a reference for the hybrid control process.
[0038] An adaptation analysis function is configured, and the adaptation analysis function is used to perform adaptation analysis on the hybrid solution set to generate an adaptation analysis result, wherein the adaptation analysis function includes a quality analysis function and a consumption analysis function.
[0039] The adaptation analysis function includes a quality analysis function and a consumption analysis function. The quality analysis function is used to evaluate the performance of the mixing scheme in terms of product quality, including adverse reactions, puffed food safety, miscibility, and uniformity. The consumption analysis function is used to evaluate the performance of the mixing scheme in terms of resource consumption, including time consumption, power consumption, energy consumption, and other evaluation indicators. The specific function formulas are described in detail in subsequent steps and will not be repeated here.
[0040] For each scheme in the hybrid scheme set, its quality analysis score and consumption analysis score are calculated respectively, and the quality analysis score and the consumption analysis score are weighted and integrated to obtain the comprehensive analysis score of each scheme. The schemes in the hybrid scheme set are sorted according to the comprehensive analysis score, and the scheme with the best comprehensive analysis score is selected as the optimal hybrid scheme to obtain the adaptation analysis result.
[0041] Furthermore, the configuration adaptation analysis function further includes:
[0042] Configure the quality analysis function as follows:
[0043] ;
[0044] ;
[0045] ;
[0046] ; ;
[0047] in, Characterization quality analysis score, For adverse reactions, To ensure the safety of puffed food, For miscibility, Characterize uniformity, They are weight factors for adverse reactions, food safety, miscibility and uniformity. is the number of adverse events, is the total mixed batch, is the index of adverse events, Characterization The severity of the adverse events, Characterizes the maximum possible severity of an adverse event. For batches that have passed safety testing, The index of the batch that passed the security check. Characterization The quality scores of the batches that passed the safety test, For batches that do not meet safety testing requirements, The index of the batch that does not meet the safety test. Characterization The severity of the batches that did not meet safety testing, The maximum possible severity of a batch that does not meet safety testing, is the number of raw material types, Index of raw material types, Indicates The solubility of the raw materials, Indicates The quality of the raw materials, Indicates The quality of the raw materials, Index of raw material types, The number of adverse events involved in the miscibility adverse event assessment. is the index of adverse events, For the The severity of the adverse event, The maximum possible severity of the adverse event, is the standard deviation of the mixed material concentration, is the total number of samples, is the sample index, For the The concentration value of each sample, is the average value of sample concentration, is the number of adverse events of uniformity, is the uniformity adverse event index, Characterization The severity of adverse events was uniform. is the maximum severity of adverse events of uniformity; and the adaptation analysis function is configured according to the quality analysis function.
[0048] Specifically, configure the quality analysis function as follows:
[0049] ;
[0050] ;
[0051] ;
[0052] ;
[0053] ;
[0054] This quality analysis function evaluates the performance of the mixing scheme in terms of product quality, including indicators such as adverse reactions, puffed food safety, miscibility and uniformity. Among them, adverse reactions refer to non-standard phenomena that occur in the production process or in the product, such as quality defects, foreign matter mixing, etc. Puffed food safety refers to whether there are safety hazards in the product during consumption, such as harmful substance residues, etc. Miscibility refers to the degree of mutual solubility between different raw materials, and uniformity refers to the uniform distribution of the mixed materials in the final product. Among them, the weight factors of adverse reactions, puffed food safety, miscibility and uniformity are used to determine the degree of influence of each quality indicator on the overall quality analysis score.
[0055] The quality analysis function comprehensively calculates the above parameters and indicators, and provides an assessment of multiple quality aspects of the puffed product production process, which helps to optimize the production process and ensure the stability of product quality.
[0056] The quality analysis function is used as a part of the adaptation analysis function to construct the adaptation analysis function.
[0057] Furthermore, the method further comprises:
[0058] Configure the consumption analysis function as follows:
[0059] ;
[0060] in, Characterize the consumption analysis score, is the time consumption function, is the power consumption function, is the energy consumption function, They are weight factors for time consumption, power consumption, and energy consumption respectively; and an adaptation analysis function is configured according to the quality analysis function and the consumption analysis function.
[0061] Specifically, configure the consumption analysis function as follows:
[0062] ;
[0063] The consumption analysis function is used to evaluate the performance of the hybrid solution in terms of resource consumption, including time consumption, power consumption and energy consumption. The time consumption function is used to evaluate the time required for the production process, including the time consumption of each step and the time consumption of the entire production cycle; the power consumption function is used to evaluate the power consumption required in production equipment and processes; energy consumption is used to evaluate the total amount of energy consumed in the production process, including electricity, fuel or other forms of energy. The weight factors of time consumption, power consumption and energy consumption are used to determine the degree of influence of each consumption indicator on the overall consumption analysis score.
[0064] The purpose of this consumption analysis function is to comprehensively evaluate the time efficiency, power efficiency and cost efficiency in the production process through weight factors and specific consumption data. Through function calculation, the production process can be optimized, resource waste can be reduced, and production efficiency and economic benefits can be maximized.
[0065] Integrate the quality analysis function and the consumption analysis function, for example, set weight factors for the quality analysis score and the consumption analysis score, perform weighted summation of the quality analysis score and the consumption analysis score according to the weight factors, and establish an adaptation analysis function. The adaptation analysis function can provide a comprehensive score for each hybrid scheme to guide the selection of the optimal hybrid scheme.
[0066] The mixing parameters are configured according to the adaptation analysis results, and the mixing control of the puffed product production is performed according to the mixing parameters and the adaptation analysis results.
[0067] According to the adaptation analysis results, the mixing parameters are configured, including process parameters such as mixing ratio, mixing time, temperature, pressure, etc. The mixing parameters are input into the production control system as control parameters in the production process of the puffed product, and the production line is started in combination with the optimal mixing solution in the adaptation analysis results to execute the mixing process and complete the mixing control of the puffed product production.
[0068] Furthermore, configuring the mixing parameters according to the adaptation analysis result further includes:
[0069] A joint database for hybrid control is established, wherein the joint database includes factory data, encrypted data, and public data; a data trust evaluation is performed on the joint database, and a selection probability of data in the joint database is configured based on the data trust evaluation result; a hybrid control optimization is performed based on the adaptation analysis result according to the selection probability, and the optimization result is output as a hybrid parameter.
[0070] Factory data includes various real-time and historical data related to production facilities and production processes, such as the operating status of production equipment, production line output data, environmental conditions, etc. Factory data covers all stages of the production process and provides real-time production operation status and related parameters for hybrid control; encrypted data includes sensitive data that needs to be protected and access restricted, such as intellectual property rights, customer information, etc.; public data refers to data resources that can be publicly accessed and used, such as industry standards, technical specifications, market trends, etc. Integrate factory data, encrypted data and public data into a unified database to form a joint database.
[0071] Data trust evaluation aims to evaluate the quality, credibility and reliability of each data source or data set in the joint database. Data quality includes evaluating the accuracy, completeness, consistency and timeliness of the data. Data source credibility includes evaluating the credibility and trust of the data source, including data collection methods, collection equipment, and security during data transmission. Reliability includes evaluating the historical records and long-term stability of the data to determine whether the data is persistent and reliable.
[0072] Based on the data trust evaluation results, corresponding selection probabilities are assigned to different data sources or data sets in the joint database. A higher selection probability is assigned to high-trust and high-quality data sources, so that they can be used preferentially for data analysis and decision support in the hybrid control process. A lower selection probability is assigned to low-trust or low-quality data sources, or they need to go through additional verification and correction steps before they can be used.
[0073] Based on the selection probability and adaptation analysis results, the optimization process of hybrid control is carried out, and optimization algorithms such as genetic algorithm, particle swarm algorithm, simulated annealing algorithm, etc. are used to find the optimal combination of hybrid parameters. The optimization process aims to achieve production goals such as maximizing production efficiency, optimizing product quality, and minimizing resource consumption. The optimization results are output as hybrid parameters, which include set values of various operating variables in the production process.
[0074] Furthermore, the method further comprises:
[0075] Establish an optimization evaluation function for iterative optimization; perform search evaluation of the iterative optimization process through the optimization evaluation function to generate a search evaluation result; generate a probability penalty constraint based on the search evaluation result and the iteration cycle; complete the hybrid control optimization after selecting probability optimization through the probability penalty constraint.
[0076] Determine the optimization goal of hybrid control, such as maximizing production efficiency, optimizing product quality, minimizing resource consumption, etc. According to the optimization goal, select the corresponding evaluation index to measure the performance of different hybrid parameter combinations. It can be a single index or a combination of multiple indexes. According to the actual situation, give appropriate weights to different evaluation indicators to reflect their importance to the final production results. Combined with the weights, the selected evaluation indicators are converted into mathematical functions, and an optimization evaluation function is established to evaluate the pros and cons of each hybrid parameter combination. Furthermore, various constraints in the production process are determined, such as production capacity, equipment limitations, quality standards, etc., and the optimization evaluation function is constrained by the constraints.
[0077] Using the selected optimization algorithm, such as genetic algorithm, particle swarm algorithm, simulated annealing algorithm, etc., at the beginning of the iterative optimization process, the mixed parameters are first initialized, and each mixed parameter combination is evaluated according to the optimization evaluation function. The mixed parameters are adjusted according to the evaluation results, and a new parameter combination is generated, and evaluated again. This process is iterated until the preset termination conditions are reached, such as reaching the maximum number of iterations, the objective function value converges, etc. The evaluation results generated by each evaluation include the objective function value corresponding to each parameter combination, reflecting the optimization effect and potential improvement space of each parameter combination.
[0078] According to the pros and cons of the evaluation results and the actual production needs, a probability penalty constraint is defined. The constraint can be a probability limit for certain adverse effects or high-risk events. The probability penalty constraint can reflect the probability of adverse events that need to be avoided or limited in the hybrid control process, such as quality problems, resource waste, etc. Combined with the impact of the iteration cycle, when generating the probability penalty constraint, dynamic adjustments are made based on the search evaluation results of the current iteration cycle, including evaluating the changes and stability of the evaluation results in different cycles during the optimization process, so as to adjust the strength and application scope of the constraint according to the search evaluation results in different cycles.
[0079] In the optimization algorithm, the probability of selecting the hybrid parameters is adjusted in combination with the probability penalty constraint to optimize the hybrid control process. The selection probability optimization refers to adjusting the probability of each hybrid parameter combination being selected according to the restriction conditions of the probability penalty constraint, so that the parameter combination that satisfies the constraint conditions is more likely to be selected during the optimization process. When executing the hybrid control optimization process, the search direction and strategy of the optimization algorithm are adjusted according to the probability penalty constraint. The algorithm will tend to select hybrid parameter combinations that not only perform well in target optimization but also do not exceed the probability penalty constraint.
[0080] Furthermore, the method further comprises:
[0081] The deviation monitoring threshold is configured according to the hybrid parameter and the adaptation analysis result; the monitoring and evaluation of the hybrid control process is performed through the deviation monitoring threshold, and an abnormal warning is reported according to the monitoring and evaluation result.
[0082] The deviation monitoring threshold is used to define the upper and lower limits of parameter deviations or abnormal conditions allowed during the production process. According to the specific mixing parameters and adaptation analysis results, the corresponding deviation monitoring threshold is set. For example, thresholds are set for the component concentration of the mixture, product quality indicators, energy consumption levels, etc., to ensure that these parameters remain within a reasonable range during the production process.
[0083] During the mixing control process, real-time data of key parameters are continuously monitored, such as the concentration of ingredients in the mixture, product quality indicators, energy consumption levels, etc. For each key parameter, the real-time data is compared with the set deviation monitoring threshold. If any parameter is detected to exceed the set monitoring threshold, an abnormal warning is automatically triggered to indicate which parameters have deviated. Based on the abnormal warning, the production operation team can immediately take necessary measures to adjust the mixing process to restore the parameters to the normal range.
[0084] In summary, the efficient mixing control method for puffed product production provided by the embodiments of the present application has the following technical effects:
[0085] A product database is established based on the target ratio information and the original mixing process, and the process analysis of the original mixing process in the product database is performed to establish detailed step decomposition results. This can more accurately quantify the impact and correlation of each mixing step, thereby optimizing the overall mixing process; historical production records are called, and the adjustment analysis network is used to evaluate the step adjustability. This method combines a large amount of historical data and can comprehensively evaluate the impact of each step on the quality and efficiency of the final product, thereby improving the flexibility and responsiveness of the production process; a benchmark mixing chain is established based on the step adjustability evaluation results, and the benchmark mixing chain is used for insertion traversal analysis to establish an insertion solution space, and an adaptation analysis function is configured to optimize the mixing scheme set. This method utilizes the comprehensive evaluation of the adaptation analysis function and integrates the weight factors of quality analysis and consumption analysis. It can effectively find the optimal mixing scheme while taking into account production efficiency and product quality.
[0086] 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 efficient mixing control method for the production of puffed products, characterized in that: The method comprises: Acquire a product database of puffed products, wherein the product database includes target ratio information and original mixing process; Performing process analysis of the original mixing process in the product database to establish a step decomposition result; Calling the historical production records of the puffed product, inputting the historical production records into the adjustment analysis network, the adjustment analysis network is a multi-layer neural network model, performing step adjustability evaluation of the step decomposition results, and generating step adjustability evaluation results, specifically including: calling the first layer processing network of the adjustment analysis network, executing the adjustment minimum impact evaluation of each step in the historical production records, and establishing a first adjustability evaluation result, the first layer processing network is used to evaluate the minimum impact of each step on the overall process during the adjustment process; calling the second layer processing network of the adjustment analysis network, executing the balanced impact evaluation of each step in the historical production records, and establishing a second adjustability evaluation result, the second layer processing network is used to evaluate the impact of each step on the process balance during the adjustment process, and performing attention weighted calculation on the first adjustability evaluation result and the second adjustability evaluation result to generate the step adjustability evaluation result; Establishing a benchmark mixed chain according to the step adjustability evaluation results, specifically including selecting steps with high adjustability and little impact on the production process as benchmark steps according to the step adjustability evaluation results, arranging the selected benchmark steps in process sequence to form a benchmark mixed chain, performing insertion traversal analysis on the benchmark mixed chain to establish an insertion solution space, and establishing a mixed solution set based on the insertion solution space and the benchmark mixed chain with the target ratio as the control target; Performing insertion traversal analysis with the benchmark hybrid chain to establish the insertion solution space also includes: Taking the benchmark step in the benchmark hybrid chain as the matching step, performing the associative matching of the active steps of the non-benchmark hybrid chain; Setting an association threshold, and if the association matching result satisfies the association threshold, establishing a sequence constraint between the corresponding reference step and the active step; Complete the insertion traversal analysis according to all the sequential constraints and establish the insertion solution space; configuring an adaptation analysis function, performing adaptation analysis on the mixed solution set with the adaptation analysis function, and generating an adaptation analysis result, wherein the adaptation analysis function includes a quality analysis function and a consumption analysis function, wherein the quality analysis function is used to evaluate the performance of the mixed solution in terms of product quality, including adverse reactions, puffed food safety, miscibility, and uniformity indicators, and the consumption analysis function is used to evaluate the performance of the mixed solution in terms of resource consumption, including time consumption, power consumption, and energy consumption indicators; The mixing parameters are configured according to the adaptation analysis results, and the mixing control of the puffed product production is performed according to the mixing parameters and the adaptation analysis results.
2. The efficient mixing control method for puffed product production according to claim 1, characterized in that: The configuration adaptation analysis function also includes: Configure the quality analysis function as follows: ; in, Characterization quality analysis score, For adverse reactions, To ensure the safety of puffed food, For miscibility, Characterize uniformity, They are weight factors for adverse reactions, food safety, miscibility and uniformity. is the number of adverse events, is the total mixed batch, is the index of adverse events, Characterization The severity of the adverse events, Characterizes the maximum possible severity of an adverse event. For batches that have passed safety testing, The index of the batch that passed the security check. Characterization The quality scores of the batches that passed the safety test, For batches that do not meet safety testing requirements, The index of the batch that does not meet the safety test. Characterization The severity of the batches that did not meet safety testing, The maximum possible severity of a batch that does not meet safety testing, is the number of raw material types, Index of raw material types, Indicates The solubility of the raw materials, Indicates The quality of the raw materials, Indicates The quality of the raw materials, Index of raw material types, The number of adverse events involved in the miscibility adverse event assessment. is the index of adverse events, For the The severity of the adverse event, The maximum possible severity of the adverse event, is the standard deviation of the mixed material concentration, is the total number of samples, is the sample index, For the The concentration value of each sample, is the average value of sample concentration, is the number of adverse events of uniformity, is the uniformity adverse event index, Characterization The severity of adverse events was uniform. The maximum severity of the adverse event for uniformity; Configure the adaptation analysis function according to the quality analysis function.
3. The efficient mixing control method for puffed product production according to claim 2, characterized in that: The method further comprises: Configure the consumption analysis function as follows: ; in, Characterize the consumption analysis score, is the time consumption function, is the power consumption function, is the energy consumption function, are the weight factors of time consumption, power consumption, and energy consumption respectively; Configure the adaptation analysis function according to the quality analysis function and the consumption analysis function.
4. The efficient mixing control method for puffed product production according to claim 1, characterized in that: The step of inputting the historical production records into the adjustment analysis network, performing step adjustability evaluation on the step decomposition results, and generating step adjustability evaluation results also includes: Calling the first layer processing network in the adjustment analysis network, performing the adjustment minimum impact evaluation of each step in the historical production record, and establishing the first adjustability evaluation result; Calling the second layer processing network of the adjustment analysis network to perform the equilibrium impact evaluation of each step in the historical production record and establish a second adjustability evaluation result; A weighted calculation is performed on the first adjustability evaluation result and the second adjustability evaluation result to generate a step adjustability evaluation result.
5. The efficient mixing control method for puffed product production according to claim 1, characterized in that: The configuring of the mixing parameters according to the adaptation analysis result also includes: Establishing a joint database of hybrid control, the joint database including factory data, encrypted data and public data; Performing a data trust evaluation on the joint database, and configuring a selection probability of data in the joint database based on the data trust evaluation result; A hybrid control optimization is performed on the adaptation analysis result according to the selection probability, and the optimization result is output as a hybrid parameter.
6. The efficient mixing control method for puffed product production according to claim 5, characterized in that: The method further comprises: Establish an optimization evaluation function for iterative optimization; Perform search evaluation of the iterative optimization process through the optimization evaluation function to generate a search evaluation result; Generate probability penalty constraints based on search evaluation results and iteration cycles; After the probability optimization is performed through the probability penalty constraint, the hybrid control optimization is completed.
7. The efficient mixing control method for puffed product production according to claim 1, characterized in that: The method further comprises: Configure deviation monitoring thresholds based on hybrid parameters and adaptation analysis results; The hybrid control process is monitored and evaluated by using the deviation monitoring threshold, and an abnormal warning is issued according to the monitoring and evaluation result.
Citation Information
Patent Citations
Production process parameter recommendation method, related equipment, program product and storage medium
CN115204474A
Method and system for optimizing production process of steel structural member
CN116842768A
Rubber glove process parameter optimization method and system based on data mining
CN118365215A
Method and device for providing, retrieving and using a data element
EP3557350A1
