An intelligent production capacity optimization management system for fermented tofu production equipment

Abstract

The present invention discloses an intelligent production capacity optimization and management system for fermented tofu production equipment, comprising: a data acquisition module for forming a standardized production data set; a data preprocessing module for generating a preprocessed production data set; a dynamic optimization target construction module for establishing a dynamic production capacity optimization objective function based on the preprocessed production data set; a global optimization calculation module for performing global search and optimization of the dynamic production capacity optimization objective function using an improved ant colony optimization algorithm to generate a preliminary optimization solution containing candidate solutions for equipment operating parameters; a local optimization calculation module for generating a final dynamic production capacity optimization solution using the preliminary optimization solution as input; and an operation control module for the final dynamic production capacity optimization solution, thereby achieving intelligent production capacity optimization management of fermented tofu production equipment. The present invention effectively improves the production capacity optimization efficiency of fermented tofu production equipment under complex operating conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of fermented bean curd production, and in particular to an intelligent production capacity optimization management system for fermented bean curd production equipment. Background Art

[0002] Fermented tofu is a soy product made from fermented tofu blocks. Its distinctive flavor and texture are favored by some consumers. Its production process involves several key parameters, including equipment operating status, ambient temperature and humidity, raw material supply, and fermentation time. However, in traditional management of fermented tofu production equipment, operating parameters are often fixed or controlled by simple rules, making real-time optimization and adjustment difficult in dynamic production environments. This results in large fluctuations in production capacity, low resource utilization, and even compromised final product quality.

[0003] At present, most fermented tofu production lines still rely on manual experience or preset parameters for production control. Although this method can maintain a certain production efficiency in a stable production environment, it is difficult for traditional methods to ensure that the equipment always operates in the optimal state when faced with complex factors such as raw material fluctuations, environmental changes and equipment aging. In the production of fermented foods, subtle changes in ambient temperature and humidity may directly affect the stability of the fermentation process, and traditional control methods lack the ability to adjust these dynamic factors in real time, resulting in reduced production efficiency, unstable product quality, and even food safety issues.

[0004] In recent years, some food processing companies have begun to introduce intelligent optimization algorithms to improve production control. Common approaches include production parameter optimization models based on genetic algorithms and particle swarm optimization. However, existing optimization methods face the following challenges in practical application: First, traditional optimization algorithms converge slowly in complex multi-objective optimization problems and are prone to falling into local optima. This results in insufficient real-time adjustment of equipment parameters, making it difficult to rapidly respond to dynamic changes in fermented tofu production. Second, existing optimization methods lack specific design considerations for the specific characteristics of fermented tofu production processes and fail to fully account for industry-specific factors such as raw material supply fluctuations and fermentation time, making it difficult for optimization results to effectively guide actual production operations. Furthermore, existing systems often use offline optimization modes, which prevent real-time adjustments during production, resulting in significant deviations between the optimization solutions and the actual production environment.

[0005] Therefore, there is an urgent need for a dynamic capacity optimization method combined with an intelligent optimization algorithm that can perform global search and local optimization based on real-time production data, and realize adaptive adjustment of the operating parameters of fermented tofu production equipment to improve production efficiency, reduce energy consumption, and ensure the stability of product quality. Summary of the Invention

[0006] One purpose of the present invention is to propose an intelligent production capacity optimization management system for fermented bean curd production equipment, which effectively improves the production capacity optimization efficiency of fermented bean curd production equipment under complex working conditions.

[0007] An intelligent fermented bean curd production equipment capacity optimization management system according to an embodiment of the present invention includes the following modules:

[0008] The data acquisition module is used to collect the original production data set of the fermented bean curd production equipment during the production process and perform standardization processing to form a standardized production data set;

[0009] The data preprocessing module is used to preprocess the standardized production data set. The preprocessing includes data cleaning, outlier detection, and data normalization to generate a preprocessed production data set.

[0010] Dynamic optimization target building module, used to establish dynamic capacity optimization target function based on pre-processed production data set;

[0011] The global optimization calculation module is used to perform global search optimization on the dynamic production capacity optimization objective function using an improved ant colony optimization algorithm. The operating parameters of the fermented tofu production equipment are used as the ant search path to generate a preliminary optimization solution containing candidate solutions for the equipment operating parameters.

[0012] A local optimization calculation module is used to use the preliminary optimization plan as input, and adopt a mutation adaptive differential evolution algorithm to perform local optimization and fine-tuning on the candidate solutions of equipment operating parameters to generate a final dynamic production capacity optimization plan;

[0013] The operation control module is used to apply the final dynamic production capacity optimization plan to the fermented tofu production equipment, adjust the equipment operating parameters of the fermented tofu production equipment in real time, and realize the optimized management of the production capacity of the intelligent fermented tofu production equipment.

[0014] Another object of the present invention is to provide an intelligent fermented bean curd production equipment capacity optimization management method, which is applied to an intelligent fermented bean curd production equipment capacity optimization management system, comprising the following steps:

[0015] S1 collects the original production data set of fermented tofu production equipment in the production process and standardizes it to form a standardized production data set;

[0016] S2. Preprocess the standardized production dataset to generate a preprocessed production dataset;

[0017] S3. Establish a dynamic capacity optimization objective function based on the preprocessed production dataset;

[0018] S4. Performing a global search optimization on the dynamic capacity optimization objective function using an improved ant colony optimization algorithm to generate a preliminary optimization solution containing candidate solutions for equipment operating parameters;

[0019] S5. Using the preliminary optimization plan as input, a mutation-adaptive differential evolution algorithm is employed to locally optimize and fine-tune the candidate solutions for equipment operating parameters to generate a final dynamic capacity optimization plan.

[0020] S6. Applying the final dynamic capacity optimization solution to the fermented bean curd production equipment, periodically repeating steps S1 to S5, to achieve intelligent optimized management of the production capacity of the fermented bean curd production equipment.

[0021] Optionally, the S1 includes the following steps:

[0022] S11. Collect the original production data set of the fermented tofu production equipment during the production process:

[0023]

[0024] in, is the original production dataset, Indicates the Production data, is the total amount of production data collected, The device operation status data, is the ambient temperature and humidity data, Supply data for raw materials, is the fermentation time data;

[0025] S12. Set the production data collection time window based on the operating cycle of the fermented tofu production equipment, standardize the equipment operation status production data, define standardized equipment operation status data, perform outlier detection on the ambient temperature and humidity data, perform shortage analysis on the raw material supply data and calculate the raw material supply stability. If the raw material supply stability exceeds the set threshold, an early warning signal is issued, and the fermentation time data is classified and processed to calculate the distribution density of fermentation time for different batches;

[0026] S13. Combine the standardized processing results of S12 to form a standardized production data set .

[0027] Optionally, S2 includes the following steps:

[0028] S21. Standardized production dataset Perform data cleaning, eliminate invalid production data, and define the production data set after data cleaning;

[0029] S22. Perform outlier detection on the production data set after data cleaning, calculate the mean and standard deviation of the production data, and define outlier detection rules ,like Indicates that the production data is abnormal and needs to be eliminated or corrected:

[0030]

[0031] in, Represents the production data after data cleaning, and are the mean and standard deviation of the production data of this category, is the set outlier detection threshold;

[0032] S23. Normalize the production data that has passed the outlier detection to obtain a normalized production data set, and form a preprocessed production data set .

[0033] Optionally, S3 includes the following steps:

[0034] S31. Produce data sets based on preprocessing Classify the states corresponding to equipment capacity, unit capacity energy consumption, product quality stability, ambient temperature and humidity, raw material supply stability, and fermentation time, and set the following classification rules:

[0035] Equipment capacity status:

[0036] High capacity state: the actual capacity of the fermented tofu production equipment exceeds 90% of the expected capacity standard; medium capacity state: the actual capacity of the fermented tofu production equipment is between 60% and 90% of the expected capacity standard; low capacity state: the actual capacity of the fermented tofu production equipment is less than 60% of the expected capacity standard;

[0037] Energy consumption per unit capacity:

[0038] Low energy consumption state: energy consumption per unit of production capacity is lower than 80% of the expected energy consumption standard; standard energy consumption state: energy consumption per unit of production capacity is between 80% and 120% of the expected energy consumption standard; high energy consumption state: energy consumption per unit of production capacity exceeds 120% of the expected energy consumption standard;

[0039] Product quality stability status:

[0040] Stable state: product quality fluctuation rate does not exceed 10%; moderately stable state: product quality fluctuation rate is between 10% and 20%; unstable state: product quality fluctuation rate exceeds 20%;

[0041] Ambient temperature and humidity conditions:

[0042] Suitable temperature and humidity state: the ambient temperature and humidity are both within the set optimal range; Fluctuating temperature and humidity state: the ambient temperature and humidity fluctuate between the optimal range Abnormal temperature and humidity state: The ambient temperature and humidity fluctuate beyond the optimal range above;

[0043] Raw material supply stability status:

[0044] Stable supply status: the raw material supply fluctuation rate is less than 5%, meeting production needs; slight shortage status: the raw material supply fluctuation rate is between 5% and 15%; severe shortage status: the raw material supply fluctuation rate exceeds 15%, which will cause production interruption;

[0045] Fermentation time status:

[0046] Standard fermentation state: the fermentation time is kept within the set standard time. Slight deviation: fermentation time fluctuation is between 5% and 15%; Abnormal fermentation: fermentation time fluctuation exceeds 15%, affecting product quality;

[0047] S32. Based on the classification rules of S31, establish a dynamic capacity optimization objective function The dynamic capacity optimization objective function aims to maximize equipment capacity, minimize energy consumption, and maintain product quality stability:

[0048] ;

[0049] in, is the equipment capacity status per unit time, is the energy consumption per unit capacity per unit time, is the product quality stability state per unit time, is the ambient temperature and humidity state per unit time, is the raw material supply stability state per unit time, is the fermentation time per unit time, are the corresponding weight coefficients respectively.

[0050] Optionally, the S4 includes the following steps:

[0051] S41. Optimize the objective function based on dynamic production capacity Define the ant individuals in the ant colony as a set of candidate operating parameter state combinations for fermented tofu production equipment and determine the initial pheromone concentration The initial heuristic information factor of ant search is set in combination with the equipment capacity status, unit capacity energy consumption status, product quality stability status, environmental temperature and humidity status, raw material supply stability status and fermentation time status of fermented tofu production equipment. ;

[0052] S42. Constructing an ant colony optimization state space, defining the state transition process of individual ants as the dynamic adjustment process of the operating parameters of candidate equipment for fermented tofu production equipment between different production states, and the position of individual ants in the state space represents a set of equipment operating parameter states of the fermented tofu production equipment;

[0053] S43. Design of dynamic heuristic information factor based on ant colony optimization state space , the dynamic heuristic information factor is dynamically adjusted by real-time feedback of production data:

[0054]

[0055] in, For the moment From the status To status The dynamic heuristic information factor of Current status The dynamic capacity optimization objective function of the equipment is calculated in real time. Target state The corresponding expected dynamic capacity optimization objective function is, is the sensitivity adjustment coefficient determined dynamically based on fermented tofu production data;

[0056] S44. Based on the initial pheromone concentration, a dynamic pheromone concentration update mechanism based on feedback adaptation is designed. The update rules are:

[0057]

[0058] in, It represents the dynamic pheromone concentration after the update at iteration t+1, which is used to indicate the quality of the transition from state k to state j and affects the subsequent ants' choice of the operating parameter state of the fermented tofu production equipment. is the dynamic evaporation coefficient, Indicates the fermented tofu production equipment at the current moment The difference between the capacity fluctuation and the previous capacity fluctuation, is the preset maximum permissible threshold for production capacity fluctuation. represents the contribution of the yth ant to the pheromone concentration based on its travel path at time t, and is used to enhance the operating parameter state path that performs better than the preset value in the dynamic capacity optimization of fermented tofu production equipment. M is the total number of ants;

[0059] S45. The ant colony selects a state based on a dynamic state transition probability function, which is defined as:

[0060]

[0061] in, Indicates that the ant is at time From the status Transfer to state The probability of represents the dynamic pheromone concentration after the update at iteration t+1, which is used to indicate the quality of the transition from state k to state e. For the moment From the status To status The dynamic heuristic information factor of 、 is the weight parameter, 、 Both are real-time feedback control factors, defined as real-time evaluation indicators of the effect of current state transfer on the capacity improvement of fermented tofu production equipment. is the set of transferable states;

[0062] S46. Repeat steps S42 to S45 until a preset convergence condition is reached, and generate a preliminary optimization solution including candidate solutions for equipment operating parameters.

[0063] Optionally, the S5 includes the following steps:

[0064] S51. Using the preliminary optimization plan as input, construct a candidate solution set ,in Indicates the candidate solutions, is the total number of candidate solution data;

[0065] S52. For each candidate solution , design adaptive mutation operation and generate mutation vector :

[0066]

[0067] in, 、 and Candidate solution sets A non-repeated candidate solution randomly selected from , For the Candidate solutions in the iteration The adaptive mutation factor when For the The production feedback factor of a candidate solution represents the candidate solution relative to the expected dynamic capacity optimization objective function. The degree of deviation, is the production feedback influence coefficient, is the preset candidate solution;

[0068] S53. Based on mutation vector and candidate solutions Perform adaptive crossover operation to generate test vectors , where the first Quantity The generation rules are:

[0069]

[0070] in, For the interval A randomly generated number inside, For the Candidate solutions in the iteration Adaptive crossover probability when ;

[0071] S54. For each candidate solution The corresponding test vector Calculate the dynamic capacity optimization objective function of candidate solutions separately Dynamic capacity optimization objective function with test vector , adopting the greedy selection mechanism, if the candidate solution dynamic capacity optimization objective function value Better than the test vector dynamic capacity optimization objective function , then update the candidate solution , otherwise keep the candidate solution constant;

[0072] S55. Repeat the local optimization iteration process from S52 to S54 until the preset termination condition is reached and the final dynamic capacity optimization plan is output. The dynamic capacity optimization scheme is the optimal operating parameter state of the fermented tofu production equipment obtained after local optimization.

[0073] The beneficial effects of the present invention are:

[0074] (1) The present invention adopts an improved ant colony optimization algorithm in the process of dynamic capacity optimization. By constructing the state space of the operating parameters of fermented bean curd production equipment, the ant search path is defined as the adjustment process of the production equipment operating parameters. In addition, a dynamic heuristic information factor and a feedback adaptive pheromone update mechanism are introduced in the ant colony search process. The dynamic heuristic information factor is adaptively adjusted according to the real-time data of fermented bean curd production, so that individual ants are more inclined to search for parameter combinations close to the target optimal capacity. The feedback adaptive pheromone update mechanism enhances the algorithm's responsiveness to drastic changes in the production environment by introducing a dynamic evaporation coefficient adjustment strategy based on capacity fluctuations, thereby preventing the ant colony optimization from falling into local optimality, improving the global search capability, and effectively improving the capacity optimization efficiency of fermented bean curd production equipment under complex working conditions.

[0075] (2) The present invention introduces a mutation-adaptive differential evolution algorithm to perform local optimization on the equipment operating parameters to improve the accuracy of the optimization results. By designing a mutation adjustment strategy based on the production feedback factor, the mutation factor and the crossover probability can be dynamically adjusted according to the degree of deviation of the current optimization solution relative to the target optimal solution. When the production capacity of the production equipment deviates from the target value by a large margin, the mutation factor is increased to enhance the search range and prevent local optimality. When the production capacity of the production equipment gradually approaches the optimal solution, the mutation factor is reduced to improve the local fine-tuning ability. In addition, the dynamic adjustment of the crossover probability ensures the stability of individual updates, making the optimization process more accurate and effectively improving the adaptive ability of the fermented tofu production equipment to changes in the production environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0077] Figure 1 This is a flow chart of an intelligent fermented bean curd production equipment capacity optimization management method proposed by the present invention. DETAILED DESCRIPTION

[0078] The present invention will now be described in further detail with reference to the accompanying drawings, which are simplified schematic diagrams that illustrate the basic structure of the present invention in a schematic manner.

[0079] An intelligent fermented bean curd production equipment capacity optimization management system, including the following modules:

[0080] The data acquisition module is used to collect the original production data set of the fermented bean curd production equipment during the production process and perform standardization processing to form a standardized production data set;

[0081] The data preprocessing module is used to preprocess the standardized production data set. The preprocessing includes data cleaning, outlier detection, and data normalization to generate a preprocessed production data set.

[0082] Dynamic optimization target building module, used to establish dynamic capacity optimization target function based on pre-processed production data set;

[0083] The global optimization calculation module is used to perform global search optimization on the dynamic production capacity optimization objective function using an improved ant colony optimization algorithm. The operating parameters of the fermented tofu production equipment are used as the ant search path to generate a preliminary optimization solution containing candidate solutions for the equipment operating parameters.

[0084] A local optimization calculation module is used to use the preliminary optimization plan as input, and adopt a mutation adaptive differential evolution algorithm to perform local optimization and fine-tuning on the candidate solutions of equipment operating parameters to generate a final dynamic production capacity optimization plan;

[0085] The operation control module is used to apply the final dynamic production capacity optimization plan to the fermented tofu production equipment, adjust the equipment operating parameters of the fermented tofu production equipment in real time, and realize the optimized management of the production capacity of the intelligent fermented tofu production equipment.

[0086] refer to Figure 1 The present invention also provides an intelligent fermented bean curd production equipment capacity optimization management method, which is applied to the above-mentioned intelligent fermented bean curd production equipment capacity optimization management system, comprising the following steps:

[0087] S1 collects the original production data set of fermented tofu production equipment in the production process and standardizes it to form a standardized production data set;

[0088] S2. Preprocess the standardized production dataset to generate a preprocessed production dataset;

[0089] S3. Establish a dynamic capacity optimization objective function based on the preprocessed production dataset;

[0090] S4. Use an improved ant colony optimization algorithm to perform global search optimization on the dynamic capacity optimization objective function, generating a preliminary optimization solution that includes candidate solutions for equipment operating parameters.

[0091] S5. Using the preliminary optimization plan as input, the adaptive differential evolution algorithm is used to locally optimize and fine-tune the candidate solutions for equipment operating parameters to generate the final dynamic capacity optimization plan.

[0092] S6. Apply the final dynamic capacity optimization solution to the fermented tofu production equipment and continuously adjust the equipment operating parameters through a real-time feedback mechanism to achieve intelligent capacity optimization management.

[0093] First, the optimized production parameters are uploaded to the production control system. The system will adjust the operating parameters of the production equipment according to the optimization plan. During the production process, the data acquisition module continuously monitors the equipment operating status, ambient temperature and humidity, raw material supply stability and fermentation process, and inputs real-time data into the optimization system to ensure that the production equipment always operates in the optimal state. If fluctuations in ambient temperature and humidity, raw material supply or equipment status are detected, the optimization system will readjust the optimization plan based on the new production data and update the equipment operating parameters in real time.

[0094] To prevent abnormalities from disrupting the production process, the system has an early warning mechanism. When it detects any of the following abnormalities (such as insufficient raw material supply, abnormal ambient temperature and humidity, abnormal equipment operation, or fermentation time deviation), the system automatically adjusts parameters or issues an alarm prompting manual intervention.

[0095] Insufficient raw material supply: If the raw material supply fluctuation rate exceeds 15%, the system will reduce the production rate, reduce raw material consumption, and send a request for replenishment of raw materials to the procurement system in advance.

[0096] Abnormal ambient temperature and humidity: If the ambient temperature and humidity exceed the set range by ±10%, the system will automatically adjust the air conditioning system or turn on auxiliary temperature control equipment to ensure a stable fermentation process.

[0097] Abnormal equipment operation: If the equipment power, pressure, and speed deviate from the normal range, the system will automatically reduce the equipment load or adjust the working mode to prevent the equipment from being overloaded and damaged.

[0098] Fermentation time deviation: If the fermentation time exceeds the set range, the system adjusts the stirring frequency and temperature control parameters to correct the deviation of the fermentation process and ensure stable product quality.

[0099] Ultimately, the optimization system of the present invention achieves dynamic capacity optimization of fermented tofu production equipment through intelligent closed-loop control, maximizes production efficiency, minimizes energy consumption, and ensures consistency of product quality.

[0100] In this embodiment, S1 includes the following steps:

[0101] S11. Collect the original production data set of the fermented tofu production equipment during the production process:

[0102]

[0103] in, is the original production dataset, Indicates the Production data, is the total amount of production data collected, The device operation status data, is the ambient temperature and humidity data, Supply data for raw materials, is the fermentation time data;

[0104] S12. Set the production data collection time window based on the operating cycle of the fermented tofu production equipment, standardize the equipment operation status production data, define standardized equipment operation status data, perform outlier detection on the ambient temperature and humidity data, perform shortage analysis on the raw material supply data and calculate the raw material supply stability. If the raw material supply stability exceeds the set threshold, an early warning signal is issued, and the fermentation time data is classified and processed to calculate the distribution density of fermentation time for different batches;

[0105] S13. Combine the standardized processing results of S12 to form a standardized production data set .

[0106] In this embodiment, S2 includes the following steps:

[0107] S21. Standardized production dataset Perform data cleaning, eliminate invalid production data, and define the production data set after data cleaning;

[0108] S22. Perform outlier detection on the production data set after data cleaning, calculate the mean and standard deviation of the production data, and define outlier detection rules ,like Indicates that the production data is abnormal and needs to be eliminated or corrected:

[0109]

[0110] in, Represents the production data after data cleaning, and are the mean and standard deviation of the production data of this category, is the set outlier detection threshold;

[0111] S23. Normalize the production data that has passed the outlier detection to obtain a normalized production data set, and form a preprocessed production data set .

[0112] This implementation ensures data accuracy and consistency, improving the reliability of optimization calculations, by performing data cleaning, outlier detection, and normalization on standardized production data sets. Outlier detection utilizes a dynamic threshold method based on mean and standard deviation, adaptively identifying abnormal data points during the production process. This prevents optimization calculations from being affected by noisy data, improves model stability and the credibility of optimization results, enables more precise adjustment of equipment operating parameters, and ensures the stability of the production process.

[0113] In this embodiment, S3 includes the following steps:

[0114] S31. Produce data sets based on preprocessing Classify the states corresponding to equipment capacity, unit capacity energy consumption, product quality stability, ambient temperature and humidity, raw material supply stability, and fermentation time, and set the following classification rules:

[0115] (1) Equipment production capacity status:

[0116] High capacity state: the actual capacity of the fermented tofu production equipment exceeds 90% of the expected capacity standard; medium capacity state: the actual capacity of the fermented tofu production equipment is between 60% and 90% of the expected capacity standard; low capacity state: the actual capacity of the fermented tofu production equipment is less than 60% of the expected capacity standard;

[0117] (2) Energy consumption per unit of production capacity:

[0118] Low energy consumption state: energy consumption per unit of production capacity is lower than 80% of the expected energy consumption standard; standard energy consumption state: energy consumption per unit of production capacity is between 80% and 120% of the expected energy consumption standard; high energy consumption state: energy consumption per unit of production capacity exceeds 120% of the expected energy consumption standard;

[0119] (3) Product quality stability status:

[0120] Stable state: product quality fluctuation rate does not exceed 10%; moderately stable state: product quality fluctuation rate is between 10% and 20%; unstable state: product quality fluctuation rate exceeds 20%;

[0121] (4) Ambient temperature and humidity conditions:

[0122] Suitable temperature and humidity state: the ambient temperature and humidity are both within the set optimal range; Fluctuating temperature and humidity state: the ambient temperature and humidity fluctuate between the optimal range Abnormal temperature and humidity state: The ambient temperature and humidity fluctuate beyond the optimal range above;

[0123] (5) Raw material supply stability status:

[0124] Stable supply status: the raw material supply fluctuation rate is less than 5%, meeting production needs; slight shortage status: the raw material supply fluctuation rate is between 5% and 15%; severe shortage status: the raw material supply fluctuation rate exceeds 15%, which will cause production interruption;

[0125] (6) Fermentation time status:

[0126] Standard fermentation state: the fermentation time is kept within the set standard time. Slight deviation: fermentation time fluctuation is between 5% and 15%; Abnormal fermentation: fermentation time fluctuation exceeds 15%, affecting product quality;

[0127] S32. Based on the classification rules of S31, establish a dynamic capacity optimization objective function The dynamic capacity optimization objective function aims to maximize equipment capacity, minimize energy consumption, and maintain product quality stability:

[0128]

[0129] in, is the equipment capacity status per unit time, is the energy consumption per unit capacity per unit time, is the product quality stability state per unit time, is the ambient temperature and humidity state per unit time, is the raw material supply stability state per unit time, is the fermentation time per unit time, are the corresponding weight coefficients respectively.

[0130] The above formula constructs the dynamic capacity optimization objective function for fermented tofu production equipment. This function comprehensively considers multiple optimization objectives, including production efficiency, energy consumption control, product quality stability, ambient temperature and humidity, raw material supply stability, and fermentation time. By maximizing equipment capacity, minimizing energy consumption, and controlling production environmental variables, production continuity and product consistency are ensured. The optimization goal is to maximize capacity while minimizing the uncertainty introduced by energy consumption and other variables, thereby ensuring the stability of the fermented tofu production process.

[0131] This implementation constructs a dynamic capacity optimization objective function that comprehensively considers factors such as production capacity, energy consumption, product quality stability, ambient temperature and humidity, raw material supply, and fermentation time. It then achieves a multi-objective balance through weight adjustment. By employing hierarchical classification rules to quantify each influencing factor, the optimization objective becomes more adaptable and targeted, improving the accuracy and stability of the optimization calculations and ensuring efficient operation of production equipment under varying operating conditions. This results in increased production capacity, reduced energy consumption, and consistent product quality.

[0132] In this embodiment, S4 includes the following steps:

[0133] S41. Optimize the objective function based on dynamic production capacity Define the ant individuals in the ant colony as a set of candidate operating parameter state combinations for fermented tofu production equipment and determine the initial pheromone concentration The initial heuristic information factor of ant search is set in combination with the equipment capacity status, unit capacity energy consumption status, product quality stability status, environmental temperature and humidity status, raw material supply stability status and fermentation time status of fermented tofu production equipment. ;

[0134] S42. Constructing an ant colony optimization state space, defining the state transition process of individual ants as the dynamic adjustment process of the operating parameters of candidate equipment for fermented tofu production equipment between different production states, and the position of individual ants in the state space represents a set of equipment operating parameter states of the fermented tofu production equipment;

[0135] S43. Design of dynamic heuristic information factor based on ant colony optimization state space , the dynamic heuristic information factor is dynamically adjusted by real-time feedback of production data:

[0136]

[0137] in, For the moment From the status To status The dynamic heuristic information factor of Current status The dynamic capacity optimization objective function of the equipment is calculated in real time. Target state The corresponding expected dynamic capacity optimization objective function is, is the sensitivity adjustment coefficient determined dynamically based on fermented tofu production data;

[0138] The above formula defines the dynamic heuristic information factor used by ants as they search for paths between different states during the ACO process. This allows them to prioritize optimal solutions, improving optimization efficiency and accelerating convergence. This formula enhances the ACO's global search capabilities, enabling it to prioritize optimal parameter combinations in dynamic environments, speeding up convergence and avoiding local optima.

[0139] S44. Based on the initial pheromone concentration, a dynamic pheromone concentration update mechanism based on feedback adaptation is designed. The update rules are:

[0140]

[0141] in, It represents the dynamic pheromone concentration after the update at iteration t+1, which is used to indicate the quality of the transition from state k to state j and affects the subsequent ants' choice of the operating parameter state of the fermented tofu production equipment. is the dynamic evaporation coefficient, Indicates the fermented tofu production equipment at the current moment The difference between the capacity fluctuation and the previous capacity fluctuation, is the preset maximum permissible threshold for production capacity fluctuation. represents the contribution of the yth ant to the pheromone concentration based on its travel path at time t, and is used to enhance the operating parameter state path that performs better than the preset value in the dynamic capacity optimization of fermented tofu production equipment. M is the total number of ants;

[0142] The above formula is used to dynamically update pheromone concentration during the ant colony optimization process, ensuring that pheromone concentration adjustments can adapt to the dynamic changes in the fermented tofu production process and enhance the ability to guide high-quality solutions. This ensures that pheromone concentration updates can be dynamically adjusted based on production capacity fluctuations, thereby enhancing the algorithm's adaptability to changing environments.

[0143] S45. The ant colony selects a state based on the dynamic state transition probability function, which is defined as:

[0144]

[0145] in, Indicates that the ant is at time From the status Transfer to state The probability of represents the dynamic pheromone concentration after the update at iteration t+1, which is used to indicate the quality of the transition from state k to state e. For the moment From the status To status The dynamic heuristic information factor of 、 is the weight parameter, 、 Both are real-time feedback control factors, defined as real-time evaluation indicators of the effect of current state transfer on the capacity improvement of fermented tofu production equipment. is the set of transferable states;

[0146] The above formula defines the state transition probability of ants between different states during the ant colony optimization process, enabling individual ants to optimize the search path based on historical information and real-time data.

[0147] S46. Repeat steps S42 to S45 until a preset convergence condition is reached, and generate a preliminary optimization solution including candidate solutions for equipment operating parameters.

[0148] This implementation utilizes an improved ant colony optimization algorithm during dynamic capacity optimization. By constructing a state space for the operating parameters of fermented tofu production equipment, the ants' search path is defined as the process of adjusting these parameters. Furthermore, a dynamic heuristic information factor and a feedback-adaptive pheromone update mechanism are incorporated into the ant colony search process. The dynamic heuristic information factor is adaptively adjusted based on real-time fermented tofu production data, enabling individual ants to more readily search for parameter combinations that approach the target optimal capacity. The feedback-adaptive pheromone update mechanism enhances the algorithm's responsiveness to drastic changes in the production environment by introducing a dynamic evaporation coefficient adjustment strategy based on capacity fluctuations. This prevents the ant colony optimization algorithm from falling into local optimality, improves global search capabilities, and effectively enhances the efficiency of capacity optimization for fermented tofu production equipment under complex operating conditions.

[0149] In this embodiment, S5 includes the following steps:

[0150] S51. Using the preliminary optimization plan as input, construct a candidate solution set ,in Indicates the candidate solutions, is the total number of candidate solution data;

[0151] S52. For each candidate solution , design adaptive mutation operation and generate mutation vector :

[0152]

[0153] in, 、 and Candidate solution sets A non-repeated candidate solution randomly selected from , For the Candidate solutions in the iteration The adaptive mutation factor when For the The production feedback factor of a candidate solution represents the candidate solution relative to the expected dynamic capacity optimization objective function. The degree of deviation, is the production feedback influence coefficient, is the preset candidate solution;

[0154] The above formula is used for mutation operations in the local optimization phase. By adjusting the mutation factor based on production feedback, the ability to explore optimal solutions is improved. This enhances the adaptability of the local optimization phase and improves the robustness of the algorithm in optimizing the production of fermented tofu production equipment.

[0155] S53. Based on mutation vector and candidate solutions Perform adaptive crossover operation to generate test vectors , where the first Quantity The generation rules are:

[0156]

[0157] in, For the interval A randomly generated number inside, For the Candidate solutions in the iteration Adaptive crossover probability when ;

[0158] S54. For each candidate solution The corresponding test vector Calculate the dynamic capacity optimization objective function of candidate solutions separately Dynamic capacity optimization objective function with test vector , adopting the greedy selection mechanism, if the candidate solution dynamic capacity optimization objective function value Better than the test vector dynamic capacity optimization objective function , then update the candidate solution , otherwise keep the candidate solution constant;

[0159] S55. Repeat the local optimization iteration process from S52 to S54 until the preset termination condition is reached and the final dynamic capacity optimization plan is output. The dynamic capacity optimization scheme is the optimal operating parameter state of the fermented tofu production equipment obtained after local optimization.

[0160] This embodiment introduces a mutation-adaptive differential evolution algorithm to perform local optimization on the equipment operating parameters to improve the accuracy of the optimization results. By designing a mutation adjustment strategy based on the production feedback factor, the mutation factor and crossover probability can be dynamically adjusted according to the degree of deviation of the current optimization solution relative to the target optimal solution. When the production capacity of the production equipment deviates from the target value by a large margin, the mutation factor is increased to enhance the search range and prevent local optimality. When the production capacity of the production equipment gradually approaches the optimal solution, the mutation factor is reduced to improve the local fine-tuning capability. In addition, the dynamic adjustment of the crossover probability ensures the stability of individual updates, the optimization process is more accurate, and the adaptability of the fermented tofu production equipment to changes in the production environment is effectively improved.

[0161] Example 1:

[0162] This embodiment is carried out in the automated production workshop of a fermented bean curd production enterprise. The enterprise mainly produces rice sauce fermented bean curd, with a daily production capacity of about 5,000 bottles. Since the production process of fermented bean curd involves multiple links, including tofu raw material processing, pickling, desalination, cooking, koji making, and fermentation, the equipment operating parameters need to be dynamically adjusted according to the raw material quality, ambient temperature and humidity, production batch and process requirements. However, under the traditional production mode, the enterprise mainly relies on fixed parameter settings and manual intervention to adjust the equipment operating parameters, resulting in the following problems:

[0163] The equipment operating parameters are difficult to adapt to the dynamic factors of raw material supply fluctuations and changes in ambient temperature and humidity, resulting in low capacity utilization.

[0164] Due to the large number of steps in the production process, manual adjustment of parameters is prone to errors, resulting in fluctuations in product quality.

[0165] Traditional optimization methods lack global optimization and real-time adaptive adjustment mechanisms, resulting in high energy consumption of production equipment and difficult cost control.

[0166] To address the above issues, this embodiment dynamically optimizes the operating parameters of production equipment to increase production capacity, reduce energy consumption, and ensure consistency in product quality.

[0167] In this experiment, the fermented tofu production process was optimized according to the following steps:

[0168] The system first deploys multiple sensors, including equipment operation status sensors, ambient temperature and humidity sensors, raw material supply monitoring equipment, and a fermentation time tracking system, to collect data from the production process in real time, including:

[0169] Equipment operating status data (operating power, speed, and pressure parameters of production equipment);

[0170] Environmental temperature and humidity data (real-time temperature and humidity in fermentation rooms and production workshops);

[0171] Raw material supply data (supply and inventory of soybeans, rice, and auxiliary materials);

[0172] Fermentation time data (fermentation time distribution of different batches of fermented tofu).

[0173] After data collection, data cleaning and standardization are first performed to eliminate abnormal data, fill missing data, and normalize different parameters to make them have the same scale to ensure the effectiveness of the optimization algorithm.

[0174] This embodiment constructs a dynamic capacity optimization objective function based on a standardized production data set, where the optimization objectives include:

[0175] Maximize the unit time output of production equipment;

[0176] Minimize energy consumption per unit of production capacity and reduce energy consumption by adjusting production parameters;

[0177] Ensure product quality stability, control environmental temperature and humidity fluctuations, and fermentation time deviations to avoid product quality fluctuations.

[0178] After the objective function is established, it is input into the optimization algorithm for solution.

[0179] This embodiment uses an improved ant colony optimization algorithm to perform a global search on the objective function.

[0180] The state of an individual ant is represented as a combination of equipment operating parameters.

[0181] Enhance the search capability for high-quality solutions through heuristic information factors.

[0182] An adaptive pheromone update mechanism is adopted, and adjustments are made in combination with production capacity fluctuation factors to ensure the stability of the search process. After 50 iterations, a preliminary optimization plan is generated.

[0183] For the preliminary optimization scheme, the mutation adaptive differential evolution algorithm is further used for local optimization:

[0184] Dynamically adjust the variation factor based on the production feedback factor to improve the convergence speed of optimization.

[0185] Adopt crossover probability that is adaptively adjusted based on production data to improve the accuracy of the optimization solution.

[0186] After 100 iterations, the final dynamic capacity optimization plan is generated.

[0187] The optimized parameters are ultimately applied to production equipment, forming an intelligent closed-loop control system. During the production process, the system continuously monitors production data and makes real-time adjustments based on the optimization algorithm to keep the equipment in optimal operating condition.

[0188] In this experiment, we conducted comparative experiments using the method of this embodiment and the traditional method. The experiment lasted for 2 months and optimized 120 production batches in total.

[0189] The production capacity per unit time is increased by 21.4%. Since the traditional method uses fixed parameters, it is impossible to adjust the production parameters in time, resulting in low utilization of production equipment. However, this embodiment optimizes the equipment parameters in real time, making the production rhythm more stable and improving the production capacity per unit time.

[0190] The energy consumption per unit of production capacity is reduced by 20.2%. The traditional method has high energy consumption during the production process, mainly due to the lagging adjustment of production parameters. However, this embodiment reduces energy consumption and improves energy utilization by optimizing the operating status of the equipment.

[0191] The product quality fluctuation rate was reduced by 57.3%. In the traditional method, when the temperature and humidity fluctuated greatly or the fermentation time was not properly controlled, the product quality was prone to large fluctuations. However, this embodiment made the product quality more stable by optimizing the ambient temperature and humidity and the fermentation time.

[0192] The response time for equipment parameter adjustment is reduced by 75.0%. The traditional method relies on manual intervention to adjust parameters, which has a long response time and leads to a decrease in equipment operating efficiency. The optimization system of this embodiment can automatically adjust parameters in a short time, significantly shortening the response time.

[0193] Production costs were reduced by 13.6%. Due to the increase in unit production capacity, decrease in energy consumption and improved product quality stability, production costs were effectively controlled, ultimately reducing overall production costs by 13.6%.

[0194] This embodiment verifies the effectiveness of this embodiment in the dynamic production capacity optimization of fermented tofu production equipment. The experimental results show that the method of this embodiment has obvious advantages over the traditional method in production efficiency, energy consumption control, product quality stability and production cost. By combining the improved ant colony optimization algorithm with the mutation adaptive differential evolution algorithm, this embodiment successfully constructs an intelligent optimization system, which enables the fermented tofu production equipment to be dynamically adjusted according to real-time data, to maximize production capacity, minimize energy consumption, and ensure product quality stability, ultimately improving the intelligence level of the fermented tofu production industry.

[0195] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An intelligent fermented bean curd production equipment capacity optimization management system, characterized in that: include: The data acquisition module is used to collect the original production data set of the fermented bean curd production equipment during the production process and perform standardization processing to form a standardized production data set; The data preprocessing module is used to preprocess the standardized production data set. The preprocessing includes data cleaning, outlier detection, and data normalization to generate a preprocessed production data set. Dynamic optimization target building module, used to establish dynamic capacity optimization target function based on pre-processed production data set; The global optimization calculation module is used to perform global search optimization on the dynamic production capacity optimization objective function using an improved ant colony optimization algorithm. The operating parameters of the fermented tofu production equipment are used as the ant search path to generate a preliminary optimization solution containing candidate solutions for the equipment operating parameters. The global optimization calculation module includes: defining the ant individuals in the ant colony as a set of candidate operating parameter state combinations of fermented tofu production equipment according to the dynamic production capacity optimization objective function, determining the initial pheromone concentration , and combined with the equipment capacity status, unit capacity energy consumption status, product quality stability status, environmental temperature and humidity status, raw material supply stability status and fermentation time status of fermented tofu production equipment to set the initial heuristic information factor of ant search; An ant colony optimization state space was constructed, and the state transition process of individual ants was defined as the dynamic adjustment process of the operating parameters of candidate equipment in the fermented tofu production equipment between different production states. The position of individual ants in the state space represents a set of equipment operating parameter states of the fermented tofu production equipment. Design dynamic heuristic information factors based on ant colony optimization state space ,The dynamic heuristic information factor is dynamically adjusted by real-time feedback of production data; Design a dynamic pheromone concentration update mechanism based on feedback adaptation based on the initial pheromone concentration; The ant colony selects its state based on the dynamic state transition probability function; Repeat the process until the preset convergence condition is reached, and generate a preliminary optimization solution containing candidate solutions for equipment operating parameters; A local optimization calculation module is used to use the preliminary optimization plan as input, and adopt a mutation adaptive differential evolution algorithm to perform local optimization and fine-tuning on the candidate solutions of equipment operating parameters to generate a final dynamic production capacity optimization plan; The operation control module is used to apply the final dynamic capacity optimization plan to the fermented tofu production equipment, adjust the equipment operating parameters of the fermented tofu production equipment in real time, and realize the optimized management of the production capacity of the intelligent fermented tofu production equipment; The dynamic heuristic information factor is dynamically adjusted by real-time feedback from production data: ； in, For the moment From the status To status The dynamic heuristic information factor of Current status The dynamic capacity optimization objective function of the equipment is calculated in real time. Target state The corresponding expected dynamic capacity optimization objective function is, is the sensitivity adjustment coefficient determined dynamically based on fermented tofu production data; Dynamic pheromone concentration The update rules are: ； in, It represents the dynamic pheromone concentration after the update at iteration t+1, which is used to indicate the quality of the transition from state k to state j and affects the subsequent ants' choice of the operating parameter state of the fermented tofu production equipment. is the dynamic evaporation coefficient, Indicates the fermented tofu production equipment at the current moment The difference between the capacity fluctuation and the previous capacity fluctuation, is the preset maximum permissible threshold for production capacity fluctuation, represents the contribution of the yth ant to the pheromone concentration based on its travel path at time t, and is used to enhance the operating parameter state path that performs better than the preset value in the dynamic capacity optimization of fermented tofu production equipment. M is the total number of ants; The dynamic state transition probability function is defined as: ； in, Indicates that the ant is at time From the status Transfer to state The probability of represents the dynamic pheromone concentration after the update at iteration t+1, which is used to indicate the quality of the transition from state k to state e. For the moment From the status To status The dynamic heuristic information factor of 、 is the weight parameter, 、 Both are real-time feedback control factors, defined as real-time evaluation indicators of the effect of current state transfer on the capacity improvement of fermented tofu production equipment. is a set of transferable states.

2. An intelligent fermented bean curd production equipment capacity optimization management method, applied to the intelligent fermented bean curd production equipment capacity optimization management system according to claim 1, characterized in that: The following steps are involved: S1 collects the original production data set of fermented tofu production equipment in the production process and standardizes it to form a standardized production data set; S2. Preprocess the standardized production dataset to generate a preprocessed production dataset; S3. Establish a dynamic capacity optimization objective function based on the preprocessed production dataset; S4. Performing a global search optimization on the dynamic capacity optimization objective function using an improved ant colony optimization algorithm to generate a preliminary optimization solution containing candidate solutions for equipment operating parameters; S5. Using the preliminary optimization plan as input, a mutation-adaptive differential evolution algorithm is employed to locally optimize and fine-tune the candidate solutions for equipment operating parameters to generate a final dynamic capacity optimization plan. S6. Applying the final dynamic capacity optimization solution to the fermented bean curd production equipment, periodically repeating steps S1 to S5, to achieve optimized management of the production capacity of the intelligent fermented bean curd production equipment.

3. The intelligent fermented bean curd production equipment capacity optimization management method according to claim 2, characterized in that: Said S1 comprises the following steps: S11. Collect the original production data set of fermented tofu production equipment during the production process ; S12. Set the production data collection time window based on the operating cycle of the fermented tofu production equipment, standardize the equipment operation status production data, define standardized equipment operation status data, perform outlier detection on the ambient temperature and humidity data, perform shortage analysis on the raw material supply data and calculate the raw material supply stability. If the raw material supply stability exceeds the set threshold, an early warning signal is issued, and the fermentation time data is classified and processed to calculate the distribution density of fermentation time for different batches; S13. Combine the standardized processing results of S12 to form a standardized production data set .

4. The intelligent fermented bean curd production equipment capacity optimization management method according to claim 3, characterized in that: The S2 comprises the following steps: S21. Standardized production dataset Perform data cleaning, eliminate invalid production data, and define the production data set after data cleaning; S22. Perform outlier detection on the production dataset after data cleaning; S23. Normalize the production data that has passed the outlier detection to obtain a normalized production data set, and form a preprocessed production data set .

5. The intelligent fermented bean curd production equipment capacity optimization management method according to claim 4, characterized in that: The S3 includes the following steps: S31. Produce data sets based on preprocessing Classify the corresponding states of equipment capacity, energy consumption per unit capacity, product quality stability, ambient temperature and humidity, raw material supply stability, and fermentation time respectively; S32. Based on the classification rules of S31, establish a dynamic capacity optimization objective function to maximize equipment capacity, minimize energy consumption, and maintain product quality stability. .

6. The intelligent fermented bean curd production equipment capacity optimization management method according to claim 5, characterized in that: The classification rules of S31 include: Equipment capacity status: High capacity state: the actual capacity of the fermented tofu production equipment exceeds 90% of the expected capacity standard; medium capacity state: the actual capacity of the fermented tofu production equipment is between 60% and 90% of the expected capacity standard; low capacity state: the actual capacity of the fermented tofu production equipment is less than 60% of the expected capacity standard; Energy consumption per unit capacity: Low energy consumption state: energy consumption per unit of production capacity is lower than 80% of the expected energy consumption standard; standard energy consumption state: energy consumption per unit of production capacity is between 80% and 120% of the expected energy consumption standard; high energy consumption state: energy consumption per unit of production capacity exceeds 120% of the expected energy consumption standard; Product quality stability status: Stable state: product quality fluctuation rate does not exceed 10%; moderately stable state: product quality fluctuation rate is between 10% and 20%; unstable state: product quality fluctuation rate exceeds 20%; Ambient temperature and humidity conditions: Suitable temperature and humidity state: the ambient temperature and humidity are both within the set optimal range; Fluctuating temperature and humidity state: the ambient temperature and humidity fluctuate between the optimal range Abnormal temperature and humidity state: The ambient temperature and humidity fluctuate beyond the optimal range above; Raw material supply stability status: Stable supply status: the raw material supply fluctuation rate is less than 5%, meeting production needs; slight shortage status: the raw material supply fluctuation rate is between 5% and 15%; severe shortage status: the raw material supply fluctuation rate exceeds 15%, which will cause production interruption; Fermentation time status: Standard fermentation state: the fermentation time is kept within the set standard time. Slight deviation: fermentation time fluctuation is between 5% and 15%; Abnormal fermentation: fermentation time fluctuation exceeds 15%, affecting product quality.

7. The intelligent fermented bean curd production equipment capacity optimization management method according to claim 6, characterized in that: The S5 comprises the following steps: S51. Using the preliminary optimization plan as input, construct a candidate solution set ,in Indicates the candidate solutions, is the total number of candidate solution data; S52. For each candidate solution , design adaptive mutation operation and generate mutation vector ; S53. Based on mutation vector and candidate solutions Perform adaptive crossover operation to generate test vectors ; S54. For each candidate solution The corresponding test vector Calculate the dynamic capacity optimization objective function of candidate solutions separately Dynamic capacity optimization objective function with test vector , adopting the greedy selection mechanism, if the candidate solution dynamic capacity optimization objective function value Better than the test vector dynamic capacity optimization objective function , then update the candidate solution , otherwise keep the candidate solution constant; S55. Repeat the local optimization iteration process from S52 to S54 until the preset termination condition is reached and the final dynamic capacity optimization plan is output. The dynamic capacity optimization scheme is the optimal operating parameter state of the fermented tofu production equipment obtained after local optimization.

8. The intelligent fermented bean curd production equipment capacity optimization management method according to claim 7, characterized in that: The test vector No. Quantity The generation rule is The amount is , for ; For the interval A randomly generated number inside, For the Candidate solutions in the iteration Adaptive crossover probability when .

Images