Automatic batching system for hotpot condiment processing

Through automatic ingredients system and machine learning model, the problems of low processing efficiency of traditional hot pot bases and difficult to meet personalized needs are solved, efficient and accurate production and personalized customization of hot pot bases are achieved, and market competitiveness and customer satisfaction are enhanced. Spend.

CN118949812APending Publication Date: 2024-11-15SICHUAN CHUANWEILI FOOD CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411011229.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Traditional hot pot base processing relies on manual operations, resulting in inefficiency and inaccurate ingredients ratios, making it difficult to meet personalized needs, affecting quality and market competitiveness.

Method used

Develop an automatic ingredient system for processing hot pot bases, combining automatic ingredient control module, automatic ingredient execution unit, hot pot base satisfaction data acquisition module and machine learning-based processing technology prediction model to achieve automated ingredient and process optimization.

Benefits of technology

It realizes full automation of the ingredients process, ensures the accuracy and consistency of ingredients, reduces labor costs, improves product quality and market competitiveness, and can adjust the process in real time according to customer feedback to meet personalized needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118949812A_ABST
    Figure CN118949812A_ABST
Patent Text Reader

Abstract

The invention discloses an automatic batching system for hotpot condiment processing, and relates to the technical field of hotpot condiment processing, and the automatic batching system comprises an automatic batching control module, an automatic batching execution unit, a hotpot condiment satisfaction degree data acquisition module, and a pre-trained hotpot condiment processing technology prediction model based on machine learning. The hotpot condiment satisfaction data acquisition module is used for acquiring hotpot condiment satisfaction data from a customer terminal through a wireless communication network; the hotpot condiment processing technology prediction model is used for predicting the latest hotpot condiment processing technology data according to the hotpot condiment satisfaction degree data; and the automatic batching control module is used for controlling the automatic batching execution unit to complete automatic batching in the hotpot condiment processing process according to the latest hotpot condiment processing technology data. Through automatic and intelligent technical means, the production efficiency and the product quality are improved, personalized customization and dynamic response to market demands are realized, and the market competitiveness is enhanced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of hot pot soup base processing, and in particular to an automatic batching system for hot pot soup base processing. Background Art

[0002] As a traditional Chinese dining form, hot pot is loved by consumers for its rich and diverse ingredients and rich flavor. The taste and texture of hot pot largely depends on the quality of hot pot soup base. Therefore, the processing of hot pot soup base has become a key factor affecting the taste and quality of hot pot. However, in the traditional hot pot soup base processing, the steps of adding, mixing and weighing ingredients mostly rely on manual operation, which is not only inefficient, but also easily leads to inaccurate ingredient ratios, thus affecting the quality and consistency of the hot pot soup base.

[0003] With the development of modern science and technology, automation and intelligent technologies have gradually been applied in various fields and have achieved remarkable results. The food processing industry is no exception. By introducing automation and intelligent technologies, production efficiency and product quality can be effectively improved, and labor costs and operating errors can be reduced. However, automatic batching systems for hot pot base processing are still relatively rare on the market, especially intelligent batching systems that can be customized based on customer feedback. Therefore, it is of great practical significance and market value to develop a system that can realize the automated processing of hot pot base and optimize the process based on customer satisfaction data.

[0004] In the prior art, the production of hot pot soup base generally includes the following steps: ingredient preparation, ingredient weighing, ingredient addition, ingredient mixing and packaging. Traditional hot pot soup base processing mainly relies on manual operation, which is not only inefficient, but also prone to errors in the process of adding and mixing ingredients, resulting in unstable quality and taste of hot pot soup base. In addition, the traditional hot pot soup base production method is difficult to quickly adjust according to market demand and customer feedback, and cannot meet the personalized needs of different customers, affecting the market competitiveness and brand value of enterprises.

[0005] In order to solve the above problems, modern food processing technology has gradually introduced automated equipment and intelligent control systems, replacing manual operations with mechanical equipment to achieve automatic addition and mixing of ingredients, thereby improving production efficiency and product quality. However, most traditional automated batching systems only have basic automation functions, making it difficult to achieve intelligent process optimization and dynamic adjustment, and unable to make real-time adjustments and optimizations based on market demand and customer feedback, making it difficult to meet the modern market's demand for high-quality and personalized products. Summary of the invention

[0006] The present invention provides an automatic batching system for hot pot soup base processing, which can directly and automatically obtain customer satisfaction feedback data on the hot pot soup base, and predict and optimize the processing technology through a machine learning model to achieve dynamic adjustment and personalized customization of the process.

[0007] The technical solution adopted by the present invention is as follows:

[0008] The present invention provides an automatic batching system for hot pot soup base processing, comprising an automatic batching control module, an automatic batching execution unit, a hot pot soup base satisfaction data collection module, and a pre-trained hot pot soup base processing technology prediction model based on machine learning; the hot pot soup base satisfaction data collection module is used to collect hot pot soup base satisfaction data from customer terminals through a wireless communication network; the hot pot soup base processing technology prediction model is used to predict the latest hot pot soup base processing technology data according to the hot pot soup base satisfaction data; the automatic batching control module is used to control the automatic batching execution unit to complete automatic batching in the hot pot soup base processing process according to the latest hot pot soup base processing technology data.

[0009] As a further description of the above technical solution: the automatic batching execution unit includes a number of sequentially numbered batching storage bins for loading various ingredients one by one, and an intelligent weighing module and an automatic batching discharge module arranged in the batching storage bins; the automatic batching discharge module is used to automatically discharge the ingredients in the batching storage bins into the batching mixing chamber; the intelligent weighing module is used to weigh the ingredients in the batching storage bins.

[0010] As a further description of the above technical solution: the batching and mixing chamber is provided with an automatic stirrer, and the automatic stirrer is used to mix the ingredients in the batching and mixing chamber.

[0011] As a further description of the above technical solution: it also includes an automatic flushing system for flushing the ingredient mixing chamber and the ingredient storage bin.

[0012] As a further description of the above technical solution: according to one or more items in the hot pot soup base satisfaction data, the corresponding customers are divided into multiple categories, and each category of customers is matched with a kind of hot pot soup base processing data; there are multiple automatic ingredient proportioning execution units, and each automatic ingredient proportioning execution unit is used to automatically proportion ingredients according to its corresponding latest hot pot soup base processing data; the hot pot soup base processing technology prediction model can predict a plurality of latest hot pot soup base processing data corresponding to each automatic ingredient proportioning execution unit according to the hot pot soup base satisfaction data.

[0013] As a further description of the above technical solution: the hot pot soup base processing technology prediction model generates the latest hot pot soup base processing technology data at least once a month.

[0014] As a further description of the above technical solution: the hot pot base processing technology prediction model is built based on a multi-layer perceptron, uses mean square error as the loss function for training, and adopts back propagation and gradient descent to update parameters.

[0015] As a further description of the above technical solution: the loss function used to train the hot pot base processing technology prediction model is constructed based on the overall satisfaction loss, taste satisfaction loss, spiciness satisfaction loss, freshness satisfaction loss, sweetness satisfaction loss, aroma satisfaction loss, color satisfaction loss and texture satisfaction loss assigned different weights.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1) The system realizes full automation of the batching process through the automatic batching control module and the automatic batching discharge module, which reduces manual intervention and improves production efficiency; the intelligent weighing module monitors the weight of the ingredients in real time to ensure that the amount of each ingredient added is accurate, reducing the time for manual weighing and adjustment; the ingredient storage bin and the automatic discharge module accurately control the amount and order of each ingredient added to ensure the consistency of the ingredients of each batch of hot pot soup base and improve product quality; the automatic agitator ensures that the ingredients are evenly mixed in the ingredient mixing chamber to avoid taste differences caused by uneven mixing, thereby improving the overall quality of the hot pot soup base.

[0018] 2) Automated batching and intelligent control systems reduce dependence on manual operations, reduce labor costs, and improve production efficiency. Accurate batching and mixing control reduces ingredient waste and repeated production, optimizes resource utilization, and reduces raw material costs.

[0019] 3) The system adopts advanced automation and intelligent technologies, which improves the production process level of hot pot soup base and enhances the company's technological competitiveness; by providing high-quality, personalized hot pot soup base products, it enhances brand value and market recognition, and wins more customers and market share for the company.

[0020] 4) Through the hot pot soup base satisfaction data collection module, the system can directly obtain customer satisfaction feedback data, and optimize and adjust the ingredient process through machine learning models to achieve personalized customization and meet the taste needs of different customers; the hot pot soup base processing technology prediction model based on machine learning can adjust the ingredient sequence, time and amount in real time according to customer feedback, quickly respond to changes in market demand, and improve customer satisfaction; the hot pot soup base processing technology prediction model inputs multi-dimensional customer satisfaction data (such as taste, aroma, spiciness, saltiness, umami, color, texture and overall satisfaction), comprehensively considers customer feedback from all aspects, and ensures The comprehensiveness and accuracy of process optimization uses advanced machine learning algorithms (such as multi-layer neural networks) and big data training models to accurately capture the complex relationship between customer satisfaction and processing technology, thereby improving the accuracy and reliability of process optimization. The model dynamically predicts and adjusts the processing technology of hot pot soup base based on real-time customer satisfaction data to ensure that each batch of hot pot soup base can meet the personalized needs of current customers and improve customer satisfaction and loyalty. Through model prediction, different parameters such as ingredient sequence, time and quantity can be flexibly adjusted to form a variety of ingredient schemes to meet the different tastes and needs of hot pot soup base products on the market.

[0021] 5) The loss function comprehensively considers multiple objectives such as taste error, aroma error, spiciness error, saltiness error, umami error, color error, texture error and overall satisfaction error. Through the multi-objective optimization algorithm, the processing technology of hot pot base is comprehensively evaluated and optimized to ensure that customer satisfaction in all dimensions is improved; the weights of different objectives in the loss function can be adjusted according to actual needs, flexibly adapt to changes in demand from different markets and customers, and improve the applicability and flexibility of the model; through optimization algorithms such as gradient descent, the loss function is minimized, the model parameters are gradually adjusted, and the prediction accuracy and training effect of the model are improved; the multi-layer neural network model uses the back propagation algorithm to calculate the gradient of the loss function according to the gradient of the loss function The weights and biases of each layer are adjusted based on the degree information to ensure that the model gradually converges to the optimal solution during the training process; the design of the loss function can accurately feedback the changes in customer satisfaction after the process parameters are adjusted, and guide the model to gradually optimize process parameters such as ingredient sequence, time and amount, thereby improving the accuracy of process prediction; the loss function evaluates and feedbacks the optimization effect of the current process in real time, and supports the model to make dynamic adjustments based on the latest customer satisfaction data to ensure that each batch of hot pot soup base can meet the latest customer needs; through continuous training and optimization, the loss function can guide the model to adaptively adjust process parameters to cope with changes in demand from different customers and markets, thereby improving the flexibility and adaptability of hot pot soup base processing.

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the embodiments of the present invention are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 It is a schematic flowchart of the working process of the automatic batching system for hot pot soup base processing described in the embodiment;

[0025] Figure 2 is a schematic structural diagram of the automatic batching execution unit described in the embodiment;

[0026] Figure 3 It is a schematic diagram of the flow of the automatic batching control module controlling the automatic batching execution unit in the embodiment. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0028] Please refer to Figure 1 The embodiment of the present invention provides an automatic batching system for hot pot soup base processing, including an automatic batching control module, an automatic batching execution unit, a hot pot soup base satisfaction data collection module, a microprocessor and a local memory.

[0029] The local memory stores the hot pot soup base material processing technology data predicted last time and the hot pot soup base material processing technology prediction model updated last time. The microprocessor can load the hot pot soup base material processing technology prediction model updated last time from the local memory and obtain the hot pot soup base material processing technology data predicted last time.

[0030] The working process of the automatic batching system for hot pot soup base processing described in the embodiment of the present invention is as follows:

[0031] 1. The hot pot soup base satisfaction data collection module collects hot pot soup base satisfaction data from customer terminals through a wireless communication network.

[0032] Among them, the hot pot base satisfaction data collection module is responsible for collecting customer satisfaction feedback on hot pot base.

[0033] In this embodiment, the hot pot base satisfaction data includes overall satisfaction, taste satisfaction, spiciness satisfaction, freshness satisfaction, sweetness satisfaction, aroma satisfaction, color satisfaction and texture satisfaction. Each type of satisfaction has a corresponding score of no more than 10 points given by the customer. The overall satisfaction is used to comprehensively evaluate the satisfaction with the hot pot base. The taste satisfaction is used to evaluate the overall taste experience of the hot pot base. The spiciness satisfaction is used to evaluate whether the spiciness of the hot pot base meets expectations. The freshness satisfaction is used to evaluate whether the freshness and aroma of the hot pot base are rich. The sweetness satisfaction is used to evaluate whether the sweetness of the hot pot base is moderate. The aroma satisfaction is used to evaluate whether the aroma of the hot pot base is pleasant. The color satisfaction is used to evaluate whether the color of the hot pot base is satisfactory. The texture satisfaction is used to evaluate whether the taste of the hot pot base is delicate and the viscosity is appropriate.

[0034] Table 1 is an example of hot pot base satisfaction data.

[0035] Table 1

[0036]

[0037] Most customer terminals are payment phones of customers. Therefore, in this embodiment, the customer terminal is a mobile phone. Merchants can create WeChat public accounts, which can be service accounts or subscription accounts, depending on the frequency of interaction with customers and functional requirements. Using the development interface of WeChat public accounts, the hot pot base material satisfaction data survey form (including overall satisfaction, taste satisfaction, spiciness satisfaction, freshness satisfaction, sweetness satisfaction, aroma satisfaction, color satisfaction and texture satisfaction, as well as corresponding option descriptions and scoring options) is embedded in the menu of the public account, and customers can participate in the survey by clicking on the menu.

[0038] After customers consume, the official account can push a link to the questionnaire to encourage them to provide feedback. It should be noted that it is easy to obtain hot pot base satisfaction data, and customers can be motivated to obtain it through incentives, such as rewarding lucky draws or giving consumer coupons.

[0039] After the customer completes and submits the questionnaire, the data will be uploaded to the hot pot base satisfaction data collection module.

[0040] In this embodiment, the wireless communication network is the key to connecting the customer terminal and the automatic batching system, which can be a 4G or 5G network, or wireless wifi.

[0041] 2. The microprocessor loads the hot pot soup base processing technology prediction model updated by the last training, and obtains the hot pot soup base processing technology data predicted last time. The hot pot soup base processing technology prediction model predicts the latest hot pot soup base processing technology data based on the hot pot soup base satisfaction data and the hot pot soup base processing technology data predicted last time.

[0042] In this embodiment, the hot pot base processing technology prediction model is constructed based on a multi-layer perceptron, uses mean square error (MSE) as a loss function for training, and adopts back propagation and gradient descent for parameter update.

[0043] Before inputting the hot pot soup base material satisfaction data into the hot pot soup base material processing technology prediction model, the hot pot soup base material satisfaction data needs to be preprocessed, specifically:

[0044] 1) Data cleaning, removal of outliers and missing values:

[0045] X clean ={x∈X|isValid(x)}

[0046] Among them, X is the original data set, X clean is the cleaned data set, x is a single data in the data set, and isValid(x) is a function that determines whether the data is valid.

[0047] 2) Data standardization: standardize each satisfaction score to the interval [0, 1]:

[0048]

[0049] Among them, X normalized is the standardized data set, X min is the minimum value in the data set, X max is the maximum value in the data set.

[0050] 3) Feature engineering to generate new features:

[0051]

[0052] Among them, feature i is the i-th feature, N is the number of samples in the data set, x ij is the j-th eigenvalue of the ith sample.

[0053] In this embodiment, the architecture of the hot pot base processing technology prediction model includes an input layer, multiple hidden layers and an output layer.

[0054] Input Layer:

[0055]

[0056] Where N is the number of samples in the dataset, and d is the number of features for each sample.

[0057] The lth hidden layer:

[0058] H (l) =ReLU(W (l-1) H (l-1) +b (l-1) )

[0059] Among them, H (l) is the output of the lth hidden layer, W (l-1) is the weight matrix of the l-1th layer, H (l-1) is the output of the l-1th layer, b (l-1) is the bias vector of the l-1th layer, ReLU is the activation function, H (0) =X.

[0060] Output layer:

[0061]

[0062] in, represents the latest predicted hot pot base processing data, W (L) is the weight matrix of the output layer, H (L) is the output of the last hidden layer, b (L) is the bias vector of the output layer.

[0063] The loss function used to train the hot pot base processing technology prediction model is:

[0064] L(W, b) = α1·L overall +α2·L taste +α3·L spiciness +α4·L flavor +α5·L sweetness +α6·L aroma +α7·L color +α8·L texture

[0065] Among them, α1, α2, α3, α4, α5, α6, α7, and α8 are the weights of overall satisfaction loss, taste satisfaction loss, spiciness satisfaction loss, freshness satisfaction loss, sweetness satisfaction loss, aroma satisfaction loss, color satisfaction loss, and texture satisfaction loss, respectively;

[0066] L overall is the overall satisfaction error,

[0067]

[0068] Where N is the number of samples, m is the number of process parameters, and Yoverall,ij is the true overall satisfaction of the jth process parameter of the i-th sample, is the predicted overall satisfaction of the jth process parameter of the i-th sample;

[0069] L taste is the taste satisfaction error,

[0070]

[0071] Among them, Y taste,ij is the actual taste satisfaction of the jth process parameter of the i-th sample, is the predicted taste satisfaction of the jth process parameter of the i-th sample;

[0072] L spiciness is the spiciness satisfaction error,

[0073]

[0074] Among them, Y spiciness,ij is the actual spiciness satisfaction of the jth process parameter of the i-th sample, is the predicted spiciness satisfaction of the jth process parameter of the i-th sample;

[0075] L flavor is the freshness satisfaction error,

[0076]

[0077] Among them, Y flavor,ij is the true freshness and aroma satisfaction of the jth process parameter of the i-th sample, is the predicted freshness and aroma satisfaction of the jth process parameter of the i-th sample;

[0078] L sweetness is the sweetness satisfaction error,

[0079]

[0080] Among them, Y sweetness,ij is the actual sweetness satisfaction of the jth process parameter of the i-th sample, is the predicted sweetness satisfaction of the jth process parameter of the i-th sample;

[0081] L aroma is the aroma satisfaction error,

[0082]

[0083] Among them, Y aroma,ij is the true aroma satisfaction of the jth process parameter of the i-th sample, is the predicted aroma satisfaction of the jth process parameter of the i-th sample;

[0084] L color is the color satisfaction error,

[0085]

[0086] Among them, Y color,ij is the true color satisfaction of the jth process parameter of the ith sample, is the predicted color satisfaction of the jth process parameter of the i-th sample;

[0087] L texture is the texture satisfaction error,

[0088]

[0089] Among them, Y texture,ij is the true texture satisfaction of the jth process parameter of the ith sample, is the predicted texture satisfaction of the jth process parameter of the ith sample.

[0090] The method of training the hot pot base processing technology prediction model includes:

[0091] 1) Initialization: Randomly initialize the weight W and bias b.

[0092] 2) Forward propagation, calculating the predicted processing parameters

[0093] 3) Calculate the loss and calculate the total loss L(W, b) through the loss function.

[0094] 4) Back propagation, calculating the gradient of the loss with respect to the weight W and bias b.

[0095] 5) Parameter update: use the Adam optimization algorithm to update weights and biases.

[0096]

[0097] Where η is the learning rate, is the gradient of the loss function with respect to the weight, is the gradient of the loss function with respect to the bias.

[0098] 6) Iteratively perform forward propagation, calculate loss, back propagation and parameter update until the loss function converges or reaches the maximum number of iterations, thus completing the training of the hot pot base processing technology prediction model.

[0099] After the training is completed, the pre-processed new hot pot base satisfaction data X can be used newTo make a prediction, first use the above-mentioned preprocessing method to process the hot pot base material satisfaction data X new After cleaning, standardization and feature engineering, forward propagation is performed to convert the preprocessed hot pot base satisfaction data X new Input the trained hot pot soup base processing technology prediction model to predict the latest hot pot soup base processing technology data

[0100] 3. Automatic batching control module based on the latest hot pot base processing data Control the automatic batching execution unit to complete the automatic batching of hot pot base ingredients during the processing.

[0101] The latest hot pot soup base processing technology data in Dedao After that, the control can be changed relative to the original hot pot base processing data, such as optimizing the order of adding ingredients, considering adding certain spices in advance, adjusting the amount of different ingredients, etc. If the spiciness score is low, the amount of chili can be increased, and the stirring time and speed can be adjusted according to the texture score.

[0102] Suppose customer feedback data shows that the freshness score is low, but the spiciness and overall taste scores are high. Based on this feedback, the model can predict: increase the amount of fresh and fragrant seasonings such as garlic and ginger; adjust the time of adding garlic and ginger, which may be earlier or later to improve the freshness; maintain or slightly adjust the amount of chili to maintain the overall spiciness. By continuously collecting and analyzing customer satisfaction data, the system can continuously optimize the recipe and processing technology of hot pot base and improve customer satisfaction.

[0103] In this embodiment, if Figure 2 As shown, the automatic batching execution unit includes a plurality of batching storage bins which are numbered in sequence and used to load various ingredients one by one, as well as an intelligent weighing module and an automatic batching discharge module arranged in the batching storage bins.

[0104] The ingredient storage bins are numbered sequentially for loading various ingredients, such as spices, seasonings, etc., one-to-one. The independent numbering of each ingredient storage bin helps to accurately control the order and amount of adding each ingredient.

[0105] The automatic ingredient discharge module is used to automatically discharge the ingredients in the ingredient storage bin into the ingredient mixing chamber. The automated discharge process reduces manual operations and improves the accuracy and efficiency of ingredient batching.

[0106] The intelligent weighing module is installed on each ingredient storage bin to monitor and weigh the weight of the ingredients in real time to ensure that the amount of each ingredient added is accurate and meets the formula requirements.

[0107] An automatic stirrer is provided in the ingredient mixing chamber. The automatic stirrer is used to mix the ingredients in the ingredient mixing chamber. All ingredients are mixed here to form the final hot pot base. The automatic stirrer ensures that the ingredients are mixed evenly to avoid taste differences caused by uneven mixing.

[0108] like Figure 3 As shown, the process of the automatic batching control module controlling the automatic batching execution unit includes:

[0109] 1) Read and analyze hot pot base processing data Get the specific addition parameters of each ingredient.

[0110] 2) Select the corresponding ingredient storage bin in turn according to the order and number of addition;

[0111] 3) The current weight of each ingredient is monitored in real time through the intelligent weighing module, and the weight to be discharged is calculated according to the added amount in the process data, and the target weight is set (target weight = initial weight - added amount);

[0112] 4) Start the automatic ingredient discharge module to discharge the ingredients in the corresponding ingredient storage bin into the ingredient mixing chamber, monitor the weighing data in real time, and when the target weight is reached, the automatic ingredient discharge module stops working and the ingredients stop being discharged.

[0113] 5) In the ingredient mixing room, start the automatic agitator to ensure that the ingredients are evenly mixed according to the hot pot base processing data The mixing time in the mixing bowl controls the stirring time.

[0114] 6) Repeat steps 2) to 5) until all ingredients are added according to the process data.

[0115] The control flow is as follows:

[0116] Assume hot pot base processing data as follows:

[0117] Ingredient 1: No. 1, addition amount 100g, addition time 0s, mixing time 30s;

[0118] Ingredient 2: No. 2, addition amount S0g, addition time 30s, mixing time 20s.

[0119] The control flow is:

[0120] Select ingredient storage bin 1 and calculate the target weight.

[0121] Dispense ingredient 1 until the weight is reduced by 100 g, start the stirrer and mix for 30 seconds.

[0122] Select ingredient storage bin 2 and calculate the target weight.

[0123] Dispense ingredient 2 until the weight is reduced by 50 g, start the stirrer and mix for 20 seconds.

[0124] Through the above steps, the automatic batching control module can efficiently and accurately control the automatic batching execution unit to complete the automatic batching process of hot pot soup base.

[0125] In an optional embodiment of the present invention, the corresponding customers can be divided into three categories according to one or more items in the hot pot soup base satisfaction data, and each category of customers is matched with a type of hot pot soup base processing technology data.

[0126] For example, by selecting the spiciness level in the hot pot base satisfaction data, customers can be divided into three categories: slightly spicy, medium spicy, and super spicy. The amount of chili ingredients in the hot pot base processing technology data corresponding to these three types of customers is different.

[0127] Similarly, the number of automatic batching execution units is also set to three. The hot pot soup base material processing technology prediction model predicts three latest hot pot soup base material processing technology data corresponding to the three automatic batching execution units based on the hot pot soup base material satisfaction data. The three automatic batching execution units can automatically batch ingredients according to their respective latest hot pot soup base material processing technology data.

[0128] In an optional embodiment of the present invention, the automatic ingredient system for hot pot base processing also includes an automatic flushing system for flushing the ingredient mixing chamber and the ingredient storage bin to ensure cleanliness and hygiene inside and outside the equipment, prevent the growth of bacteria and other harmful microorganisms, and meet the hygiene standards for food processing.

[0129] In an optional embodiment of the present invention, the hot pot soup base processing technology prediction model generates the latest hot pot soup base processing technology data at least once a month. Regularly updating the process data can reflect changes in market and customer demand, continuously optimize the hot pot soup base formula and production process, ensure that the hot pot soup base produced always meets the latest market trends and customer preferences (because many customers' preferences will change, so this system regularly updates the process data for timely adjustment), enhance the market competitiveness of the product, and optimize the production process by regularly updating the process data, improve production efficiency and output, and reduce production costs.

[0130] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An automatic batching system for hot pot soup base processing, characterized in that: It includes an automatic batching control module, an automatic batching execution unit, a hot pot soup base material satisfaction data collection module, and a pre-trained hot pot soup base material processing technology prediction model based on machine learning; the hot pot soup base material satisfaction data collection module is used to collect hot pot soup base material satisfaction data from customer terminals through a wireless communication network; the hot pot soup base material processing technology prediction model is used to predict the latest hot pot soup base material processing technology data based on the hot pot soup base material satisfaction data; the automatic batching control module is used to control the automatic batching execution unit to complete automatic batching in the hot pot soup base material processing process based on the latest hot pot soup base material processing technology data.

2. The automatic batching system for hot pot soup base processing according to claim 1, characterized in that: The automatic batching execution unit includes a plurality of batching storage bins which are numbered in sequence and used to load various ingredients one by one, and an intelligent weighing module and an automatic batching discharge module which are arranged in the batching storage bins; the automatic batching discharge module is used to automatically discharge the ingredients in the batching storage bins into the batching mixing chamber; the intelligent weighing module is used to weigh the ingredients in the batching storage bins.

3. The automatic batching system for hot pot soup base processing according to claim 2, characterized in that: The batching and mixing chamber is provided with an automatic stirrer, and the automatic stirrer is used to mix the ingredients in the batching and mixing chamber.

4. The automatic batching system for hot pot soup base processing according to claim 3, characterized in that: It also includes an automatic flushing system for flushing the ingredient mixing chamber and the ingredient storage bin.

5. The automatic batching system for hot pot soup base processing according to claim 1, characterized in that: According to one or more items in the hot pot soup base satisfaction data, the corresponding customers are divided into multiple categories, and each category of customers is matched with a kind of hot pot soup base processing technology data; there are multiple automatic ingredient proportioning execution units, and each automatic ingredient proportioning execution unit is used to automatically proportion ingredients according to its corresponding latest hot pot soup base processing technology data; the hot pot soup base processing technology prediction model can predict a plurality of latest hot pot soup base processing technology data corresponding to each automatic ingredient proportioning execution unit according to the hot pot soup base satisfaction data.

6. The automatic batching system for hot pot soup base processing according to claim 1, characterized in that: The hot pot soup base processing technology prediction model generates the latest hot pot soup base processing technology data at least once a month.

7. The automatic batching system for hot pot soup base processing according to claim 1, characterized in that: The hot pot base processing technology prediction model is built based on a multi-layer perceptron, uses mean square error as a loss function for training, and adopts back propagation and gradient descent to update parameters.

8. The automatic batching system for hot pot soup base processing according to claim 7, characterized in that: The loss function used to train the hot pot base processing technology prediction model is constructed based on the overall satisfaction loss, taste satisfaction loss, spiciness satisfaction loss, freshness satisfaction loss, sweetness satisfaction loss, aroma satisfaction loss, color satisfaction loss and texture satisfaction loss assigned different weights.