Intelligent mutton grinding and puree extraction integrated system

Abstract

The invention discloses an intelligent mutton milling and puree extraction integrated system, and relates to the technical field of intelligent control, and the intelligent mutton milling and puree extraction integrated system is characterized in that a control center collects multi-modal data through a multi-modal sensor, screens out crushed raw materials through a decision tree model, and transmits the crushed raw materials to a self-adaptive crushing unit; feeding the crushed raw materials into a self-adaptive crushing unit, dynamically adjusting the parameters of a crushing cutter by using a PID control algorithm, and calculating the crushing power to obtain crushed mutton powder; the intelligent batching unit receives the crushed mutton powder and controls and adjusts the ingredient proportion for batching according to dynamic fuzzy logic to generate a uniform mixture; the uniform mixture is conveyed to a temperature control grinding unit, the temperature is adjusted in real time through a Newton method, and grinding of the uniform mixture is completed; and after the uniform mixture is ground, extracting and separating primary pulp through centrifugal force, optimizing by using Kalman filtering, and automatically packaging the optimized primary pulp. By using a PID control algorithm, real-time optimization and efficient control of the crushing process are realized.

Description

Technical Field

[0001] The invention relates to the field of intelligent control technology, in particular to an intelligent mutton powder grinding and puree extraction integrated system. Background Art

[0002] With the continuous development of the food industry, the application of automation and intelligence in food processing is becoming more and more extensive; traditional mutton processing technology mainly relies on manual operation and simple mechanical equipment, and has problems such as low efficiency and unstable product quality. In recent years, the application of machine learning and artificial intelligence technology in food processing has gradually increased. Through sensor data collection, algorithm optimization and intelligent control, processing efficiency and product quality can be significantly improved; in addition, the development of centrifugal extraction technology and dynamic fuzzy logic control technology has also provided new solutions for food processing, which can better meet consumers' requirements for food quality and safety.

[0003] However, existing food processing technologies still have some shortcomings. First, most systems lack accurate screening of raw material quality, resulting in unstable product quality in subsequent processing links. Second, parameter adjustments during the crushing process mostly rely on fixed settings and cannot respond to changes in raw material characteristics in real time, resulting in high energy consumption and low efficiency. In addition, during the batching and extraction process, the existing ones usually lack intelligent control methods and cannot be dynamically adjusted according to real-time data, which affects the purity and nutritional content of the final product. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an integrated system of intelligent mutton powder grinding and puree extraction to solve the problem of intelligent processing efficiency.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The present invention provides an intelligent integrated system for mutton powder grinding and raw pulp extraction, which comprises:

[0008] Screening module: the control center collects multimodal data through multimodal sensors, selects the crushed raw materials through the decision tree model and transmits them to the adaptive crushing unit;

[0009] The adjustment module, the crushed raw materials enter the adaptive crushing unit, use the PID control algorithm to dynamically adjust the crushing tool parameters, calculate the crushing power, and obtain the crushed mutton powder;

[0010] In the mixing module, the intelligent batching unit receives the crushed mutton powder and adjusts the proportion of ingredients according to dynamic fuzzy logic control to generate a uniform mixture;

[0011] The regulating module transfers the uniform mixture to the temperature-controlled grinding unit, adjusts the temperature in real time using Newton's method, and completes the grinding of the uniform mixture;

[0012] The extraction module, after completing the grinding of the uniform mixture, extracts and separates the original puree through centrifugal force, optimizes it using Kalman filtering, and automatically packages the optimized original puree.

[0013] As a preferred solution of the intelligent integrated system for mutton grinding and puree extraction described in the present invention, the multimodal sensor includes an infrared spectroscopy analysis sensor, a high-resolution camera, a pressure sensor and a vibration sensor.

[0014] As a preferred solution of the intelligent integrated system for mutton grinding and puree extraction of the present invention, the multimodal data collection includes the following steps:

[0015] Collect mutton fat content data, mutton moisture content data and mutton protein content data through infrared spectrum analysis sensor;

[0016] The high-resolution camera is used to collect the fat distribution data on the surface of mutton, the color data of mutton, the color data of mutton, the shape and size data of mutton, and the texture feature data of mutton;

[0017] Collect pressure data through pressure sensors;

[0018] Vibration data is collected through vibration sensors.

[0019] As a preferred solution of the intelligent integrated system for mutton grinding and puree extraction of the present invention, the crushed raw materials are screened out by the decision tree model and transmitted to the adaptive crushing unit, and the specific steps are as follows:

[0020] The mutton fat content data, mutton moisture content data and mutton protein content data collected by the infrared spectrum analysis sensor and the mutton surface fat distribution data, mutton color data, mutton color data, mutton shape and size data and mutton texture feature data collected by the high-resolution camera are aggregated to generate a comprehensive data set;

[0021] Preprocess the data in the comprehensive dataset;

[0022] The decision tree model is trained by recursive splitting to divide the comprehensive data set into multiple subsets to form classification results;

[0023] Use cross-validation to evaluate the decision tree model, and input the preprocessed comprehensive data set into the trained decision tree model to perform actual classification judgment and generate classification results;

[0024] The decision tree model judges the quality of mutton based on the classification results;

[0025] For qualified mutton, the qualified mutton is conveyed to the adaptive crushing unit through the conveying equipment, and for unqualified mutton, it is directly discarded;

[0026] The qualified mutton is transported to the adaptive crushing unit as the crushing raw material.

[0027] As a preferred solution of the intelligent mutton powder grinding and raw pulp extraction integrated system of the present invention, the crushed raw material enters the adaptive crushing unit, and the crushing tool parameters are dynamically adjusted using the PID control algorithm to calculate the crushing power to obtain the crushed mutton powder. The specific steps are:

[0028] The hardness of the mutton pulverization is monitored in real time in the adaptive pulverization unit through the real-time pressure, average pressure, pressure fluctuation and pressure intensity in the collected pressure data;

[0029] The texture of lamb mince was monitored in real time through the vibration frequency, amplitude, acceleration and vibration pattern in the collected vibration data;

[0030] Using PID control algorithm, the speed and strength of the crushing tool can be adjusted in real time by adjusting the proportional control item, integral control item and differential control item;

[0031] In the process of adjusting the crushing tool, the crushing power is calculated based on the feedback of pressure data and vibration data, and the expression is:

[0032]

[0033] Among them, P is the crushing power, D is the density of mutton, L is the tool moving distance, μ is the friction coefficient, is the rotation speed, R is the tool radius, v is the feed speed, E is the input energy, f is the friction force, and H is the material hardness;

[0034] According to the obtained crushing power, the speed and strength of the crushing tool are adjusted to generate crushed mutton powder;

[0035] The ground mutton powder is conveyed to the intelligent batching unit.

[0036] As a preferred solution of the intelligent integrated system for mutton grinding and puree extraction of the present invention, the PID control algorithm is used to adjust the speed and strength of the crushing tool in real time by adjusting the proportional control item, the integral control item and the differential control item. The specific steps are as follows:

[0037] Define the ideal values ​​of crushing tool speed and force as crushing set points;

[0038] The pressure data and the vibration data are taken as actual measured values, and the error value is obtained according to the difference between the actual measured value and the crushing set value;

[0039] According to the error value pairing adjustment ratio as a proportional control item, the speed and strength of the initial crushing tool are adjusted;

[0040] Accumulate the current error value and the historical error value to obtain the integral control parameter as the integral control item to adjust the speed and strength of the current crushing tool;

[0041] Calculate the difference between the current error value and the historical error value, obtain the differential control parameter as the differential control item, and predict the future error trend;

[0042] The proportional control term, the integral control term and the differential control term are aggregated to generate a control signal;

[0043] According to the control signal, the speed and strength of the crushing tool are adjusted in real time.

[0044] As a preferred solution of the intelligent mutton powder grinding and puree extraction integrated system of the present invention, the intelligent batching unit receives the crushed mutton powder, adjusts the proportion of ingredients according to dynamic fuzzy logic control, and generates a uniform mixture. The specific steps are as follows:

[0045] Install a weight sensor in the intelligent batching unit to collect weight data and transmit the weight data to the control center;

[0046] Dynamic fuzzy logic control is used in combination with weight data to dynamically adjust the proportion of each component in the weight data;

[0047] The control center adjusts the proportion of ingredients through dynamic fuzzy logic control, regulates the amount of each ingredient, and generates a uniform mixture;

[0048] During the batching process, the weight data is filtered by Kalman filtering to remove noise and outliers;

[0049] The control center further adjusts the proportion of each component based on the filtered weight data.

[0050] As a preferred solution of the intelligent integrated system for mutton grinding and puree extraction of the present invention, the control center adjusts the proportion of ingredients through dynamic fuzzy logic control, adjusts the amount of each ingredient, and generates a uniform mixture. The specific steps are:

[0051] Define the dynamic fuzzy logic control input as weight data and the output as the ingredient ratio adjustment value;

[0052] The input weight data is fuzzified and divided into three fuzzy sets: low, medium and high;

[0053] A fuzzy rule base is established based on historical multimodal data and weight data to define the proportion of each ingredient;

[0054] If the ingredient weight is low, the amount of ingredients added is increased;

[0055] If the weight of the ingredient is medium, the amount of ingredients put in is reduced;

[0056] If the ingredient weight is high, the ingredient dosage remains unchanged;

[0057] According to the input weight data, the proportion of ingredients is dynamically adjusted. The expression is:

[0058]

[0059] Among them, T x (t+1) is the dynamic adjustment value of the batching ratio x at time t+1, t is time, x is the batching ratio, T x (t) is the current value of the batching ratio x at time t, γ is the adjustment coefficient, e x (t) is the error value of the batching ratio x at time t;

[0060] Match the corresponding ingredient ratio according to the dynamically adjusted ingredient ratio;

[0061] The matched ingredient ratio is converted into a specific ingredient dosage adjustment value by weighted average method;

[0062] The control center adjusts the value of the specific ingredient dosage and issues dosage instructions to generate a uniform mixture.

[0063] As a preferred solution of the intelligent integrated system for mutton grinding and puree extraction of the present invention, wherein: the uniform mixture is transmitted to the temperature-controlled grinding unit, and the temperature is adjusted in real time by Newton's method to complete the grinding of the uniform mixture, the specific steps are as follows:

[0064] A temperature sensor is installed in the temperature-controlled grinding unit to monitor the temperature of the grinding mill, generate temperature data, and transmit the temperature data to the control center;

[0065] The control center receives the temperature data, sets the upper and lower temperature limits, and automatically adjusts the temperature of the temperature-controlled grinding unit;

[0066] If the temperature exceeds the set upper limit, the heating power is reduced;

[0067] If the temperature is lower than the set lower limit, the heating power is increased;

[0068] The control center sets the target temperature, observes the difference between the current temperature and the target temperature, and generates a temperature error;

[0069] Through the iterative calculation of Newton's method, the temperature optimization adjustment strategy is generated to obtain the heat transfer amount, which is expressed as:

[0070]

[0071] Where O is the amount of heat transferred, m is the mass of the material, h is the specific heat capacity of the material, ΔW is the temperature change, β is the correction factor, Y is the fluid velocity, A is the radius of the fluid path, g is the gravitational acceleration, ∈ is the thermal conductivity, and ΔS is the height difference;

[0072] The control center adjusts the temperature-controlled grinding unit equipment according to the heat transfer amount obtained by Newton's method iteration;

[0073] The uniform mixture is ground into powder and conveyed to the adjusted temperature-controlled grinding unit to complete the uniform mixture grinding.

[0074] As a preferred solution of the intelligent mutton powder grinding and puree extraction integrated system of the present invention, wherein: the puree is separated by centrifugal force extraction, optimized by Kalman filtering, and the optimized puree is automatically packaged, the specific steps are:

[0075] Extract and separate the puree and solid particles in the homogeneous mixture through the centrifugal force after high-speed rotation;

[0076] The PSO algorithm is used to define the speed and time of the centrifugal force as a combined particle;

[0077] Initialize the combined particles and randomly set the position of each particle in the combined particles;

[0078] Calculate the fitness value of each particle in the combined particle, find the position of each particle and the position of the combined particle group;

[0079] According to the position of each particle found and the position of the combined particle group, the position of each particle in the initialized combined particle is updated;

[0080] The final centrifugal force speed and time are found through PSO algorithm iteration and input into the centrifugal force extraction unit;

[0081] The quality of the extracted puree was evaluated by weighted average method, and the extraction parameters were optimized based on the feedback;

[0082] The extracted puree is filtered through a fine filter membrane, and the filtration is optimized using a Kalman filter to obtain a purified puree;

[0083] The purified puree is conveyed to the automatic packaging unit for filling, sealing and labeling.

[0084] The beneficial effects of the present invention are as follows: the present invention realizes real-time optimization and efficient control of the pulverizing process by using a PID control algorithm. The PID control algorithm monitors and adjusts the speed and force of the pulverizing tool in real time by adjusting the proportional, integral and differential parameters to ensure the consistency and stability of the pulverizing effect. By collecting pressure and vibration data, the pulverizing power can be accurately calculated and adjusted according to the calculation results. This real-time adjustment mechanism not only improves the pulverizing efficiency, but also significantly reduces energy consumption and reduces resource waste. At the same time, the PID control algorithm can respond to changes in raw material characteristics and automatically adjust the pulverizing parameters to ensure the uniformity of the mutton powder after pulverization, thereby improving product quality and enhancing overall adaptability and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0086] Figure 1 This is a schematic diagram of the integrated system for intelligent mutton grinding and puree extraction in Example 1.

[0087] Figure 2 This is a flow chart for dynamically adjusting the proportion of each ingredient in Example 1. DETAILED DESCRIPTION

[0088] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0089] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0090] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0091] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides an intelligent mutton grinding and puree extraction integrated system, comprising the following steps:

[0092] In the screening module, the control center collects multimodal data through multimodal sensors, selects the crushed raw materials through the decision tree model and transmits them to the adaptive crushing unit.

[0093] Furthermore, the control center integrates the adaptive crushing unit, intelligent batching unit, temperature-controlled grinding unit, centrifugal extraction unit and automatic packaging unit to achieve centralized control, real-time optimization and efficient operation of intelligent mutton grinding and puree extraction, ensuring the automation, intelligence and efficiency of the entire production process. The control center uniformly manages and dispatches the operations of each unit to ensure that each unit works according to the predetermined process and time sequence;

[0094] The fat content, moisture content and protein content of mutton are collected by infrared spectroscopy analysis sensor;

[0095] The high-resolution camera is used to collect the fat distribution on the surface of the mutton, the color of the mutton, the shape and size of the mutton, and the texture characteristics of the mutton;

[0096] Pressure data was collected through pressure sensors (the texture and hardness changes of lamb during the grinding process);

[0097] Vibration data (vibration between the cutter and the lamb during the pulverization process) was collected through vibration sensors;

[0098] Select high-quality mutton, remove unnecessary parts such as bones and fascia, and cut the mutton into pieces of appropriate size to facilitate the subsequent crushing process;

[0099] The processed mutton is sent to the multimodal sensor for detection through a conveyor device;

[0100] The mutton was scanned using a near-infrared spectrometer to detect its internal fat content, moisture distribution and protein content. The near-infrared spectrum can penetrate the mutton and provide a high-precision internal component analysis.

[0101] The infrared spectroscopy analysis sensor is able to penetrate the mutton and detect its internal fat content, which is achieved by measuring the absorption peak position and intensity of the near-infrared spectrum;

[0102] The protein content in mutton can be analyzed by infrared spectroscopy sensor, which is calculated based on the absorption peak of protein at a specific wavelength;

[0103] In addition to fat, water and protein, infrared spectroscopy sensors can also detect other components such as carbohydrates and mineral content;

[0104] The surface image of the mutton (surface fat distribution, color and luster, shape and size) is captured by a high-resolution camera, and the surface fat distribution and luster are analyzed by combining image processing technology;

[0105] A high-resolution camera can capture images of the surface of mutton, and analyze the distribution of fat on the surface through image processing technology (specifically, a high-resolution camera captures images of the surface of mutton, performs denoising, grayscale, edge detection and texture analysis on the surface image, separates the fat area and the lean meat area, calculates the fat distribution ratio, and generates a fat distribution map);

[0106] The images captured by the high-resolution camera can be used to analyze the color and hue of the lamb to determine the freshness and quality of the meat;

[0107] High-resolution cameras can also measure the shape and size of the mutton to ensure that the cut mutton meets the requirements of subsequent processing;

[0108] The pressure sensor is used to detect the changes in texture and hardness of the mutton during the crushing process, helping to dynamically adjust the speed and force of the crushing tool. The pressure sensor ensures that the crushing effect matches the characteristics of the raw material by monitoring the pressure changes;

[0109] The vibration sensor is used to monitor the vibration between the cutter and the mutton during the crushing process, and further optimize the crushing process. The vibration sensor ensures the stability and effect of the crushing process by monitoring the vibration.

[0110] The mutton fat content data, mutton moisture content data and mutton protein content data collected by the infrared spectrum analysis sensor and the mutton surface fat distribution data, mutton color data, mutton color data, mutton shape and size data and mutton texture feature data collected by the high-resolution camera are aggregated to generate a comprehensive data set;

[0111] Preprocess the data in the comprehensive data set, specifically, remove random noise in the data through mean filtering;

[0112] Use moving average method to reduce data fluctuations;

[0113] Use the mean to fill in missing data or remove records with a large number of missing values;

[0114] Use the standard deviation method to identify and handle outliers;

[0115] Transform the data into a standard normal distribution so that the mean of the data is 0 and the standard deviation is 1;

[0116] Reduce the dimensionality of high-dimensional data and extract the main components;

[0117] Aggregate data from different sources to generate comprehensive data sets;

[0118] Convert continuous data into discrete data, such as dividing age groups into different groups;

[0119] Convert categorical variables into binary vector form;

[0120] The decision tree model is trained by recursive splitting to divide the comprehensive data set into multiple subsets, and the splitting points are selected according to the principle of maximizing information gain to form the classification results;

[0121] Use cross-validation to evaluate the accuracy of the decision tree model to ensure that the decision tree model can accurately classify, and input the preprocessed comprehensive data set into the trained decision tree model to perform actual classification judgment and generate classification results;

[0122] The decision tree model determines whether the quality of the mutton meets the requirements based on the classification results. Only the mutton that meets the requirements will be sent to the adaptive crushing unit;

[0123] The qualified standards are: compliance with the quality and safety standards of fat (such as 10%-20%), moisture (such as 60%-70%), protein content (such as 15%-25%), color and luster (uniform color, in line with the normal color of mutton), surface fat distribution (such as uniform fat distribution, no obvious fat accumulation), shape and size (such as regular shape, size within the standard range), texture (clear and uniform texture), etc.

[0124] For qualified mutton, the qualified mutton is conveyed to the adaptive crushing unit through the conveying equipment, and for unqualified mutton, it is directly discarded;

[0125] Specifically, the CART classification and regression tree algorithm was selected as the framework of the decision tree model;

[0126] Starting from the root node, select a feature (such as fat content, color, etc.) as a split node;

[0127] Divide the comprehensive dataset into multiple subsets based on this feature (e.g., fat content, color, etc.);

[0128] Repeat this process on each subset, continue to select features (such as fat content, color, etc.) and split subsets until the stopping condition is met (such as reaching the maximum depth, the minimum amount of split node data, etc.);

[0129] In each splitting process, the feature that maximizes information gain or minimizes the Gini coefficient is selected as the splitting point to ensure that the data purity of the subset increases after each split;

[0130] Each node in the decision tree model corresponds to the splitting of a feature, and the leaf node corresponds to a classification result;

[0131] Each leaf node represents the final classification result, i.e., the quality of the mutton;

[0132] Use the cross-validation method to evaluate the constructed decision tree model to ensure its accuracy and stability;

[0133] Divide the comprehensive data set into a training set and a validation set, repeat the training and validation multiple times, and calculate the average performance indicators of the decision tree model, such as accuracy, precision, recall, etc.;

[0134] The preprocessed comprehensive data set is input into the trained decision tree model, and each data point is classified by the decision tree model to judge the quality of the mutton;

[0135] According to the classification results of the decision tree model, judge whether the mutton is qualified or not;

[0136] For qualified mutton, it is conveyed to the adaptive crushing unit through a conveying device for further processing;

[0137] For unqualified mutton, discard it directly;

[0138] The qualified mutton is transported to the adaptive crushing unit as a crushing raw material;

[0139] It should be noted that multimodal data is collected through multimodal sensors, and a decision tree model is used to screen the crushing raw materials to ensure that only high-quality mutton is selected. Infrared spectroscopy, high-resolution cameras, pressure sensors and vibration sensors jointly provide accurate data support to dynamically adjust the crushing tool parameters to improve the crushing effect and consistency.

[0140] The adjustment module, the crushed raw materials enter the adaptive crushing unit, use the PID control algorithm to dynamically adjust the crushing tool parameters, calculate the crushing power, and obtain the crushed mutton powder.

[0141] Furthermore, the hardness of the mutton pulverization is monitored in real time in the adaptive pulverization unit through the real-time pressure (related to the material hardness), average pressure, pressure fluctuation and pressure intensity in the collected pressure data;

[0142] The texture of minced lamb was monitored in real time through the vibration data collected, including vibration frequency (related to rotation speed), amplitude, acceleration (related to input energy and friction), and vibration pattern;

[0143] Using PID control algorithm, by adjusting the proportional control item, integral control item and differential control item, the speed and strength of the crushing tool can be adjusted in real time to ensure that the crushing effect matches the characteristics of the raw materials;

[0144] Specifically, according to the target effect of mutton crushing, the ideal values ​​of crushing tool speed and force are defined as crushing setting values;

[0145] Specifically, the appropriate crushing parameters are determined based on the characteristics of the mutton, such as density, hardness, and moisture content. These characteristics can be obtained through early near-infrared spectroscopy (NIR) and high-resolution camera data analysis;

[0146] According to the specifications and capabilities of the crushing equipment, such as the stroke of the cutter, the radius of the cutter, the rotation speed, etc., set reasonable speed and force target values. The equipment manufacturer usually provides a recommended operating parameter range;

[0147] Use historical crushing data and experience to set crushing target values ​​by referring to parameters that have had good crushing effects in the past. These data can be obtained through data recording and analysis;

[0148] In the actual production process, multiple tests are carried out to adjust the speed and strength of the crushing cutter, observe the crushing effect and equipment load, and gradually optimize the crushing target value. During the test, various parameters and crushing effects need to be recorded for further correction and optimization;

[0149] Ensure that the set crushing target value is within the safe operating range of the equipment, and maximize production efficiency and minimize energy consumption while ensuring the crushing effect;

[0150] Assume that based on raw material characteristics, equipment specifications and historical data, the following target values ​​are initially set:

[0151] Speed ​​target value: 2000RPM (revolutions per minute);

[0152] Force target value: 50N (Newton);

[0153] The pressure data and vibration data are taken as the actual measured values, and the error value is obtained according to the difference between the actual measured value and the crushing target value, which is expressed as:

[0154] e(t) = GQ - QZ;

[0155] Among them, e(t) is the error value, GQ is the crushing target value, QZ is the actual measured value, t is the time, and Q is the indicated value, which is used to distinguish the set value from the actual measured value;

[0156] According to the error value, the adjustment ratio is used as a proportional control item to adjust the speed and strength of the initial crushing tool. The proportional control item is obtained by multiplying the current error with a fixed proportional constant. The expression is:

[0157] B(t)=K b ×e(t);

[0158] Among them, B(t) is the proportional control term, K b is the proportional gain, b is the index of the ratio, and e(t) is the error value;

[0159] The current error value and the historical error value are accumulated to obtain the integral control parameter as the integral control item to adjust the speed and strength of the current crushing tool. The integral control item is obtained by multiplying the accumulated error with a fixed integral constant. The expression is:

[0160]

[0161] Among them, I(t) is the integral control term, K i is the integral gain, i is the index of the integral, is the cumulative error, Δt is the time interval;

[0162] Calculate the difference between the current error value and the historical error value (that is, calculate the rate of error change), obtain the differential control parameter as the differential control item, and predict the future error trend. The expression is:

[0163]

[0164] Among them, C(t) is the differential control term, K j is the differential gain, j is the differential sign, is the error change rate, de(t) is the small change in error, and dt is the small change in time t;

[0165] The proportional control term, integral control term and differential control term are combined to generate a control signal, which is expressed as:

[0166] U(t)=B(t)+I(t)+C(t);

[0167] Among them, U(t) is the control signal, which is used to adjust the speed and strength of the crushing tool;

[0168] According to the control signal, the speed and strength of the crushing tool are adjusted in real time;

[0169] If the control signal indicates that the rotation speed of the crushing tool needs to be increased or decreased, the rotation speed is adjusted;

[0170] If the control signal indicates that the pressure applied by the crushing tool needs to be increased or decreased, the pressure is adjusted;

[0171] In the process of adjusting the crushing tool, the crushing power is calculated based on the feedback of pressure data and vibration data, and the expression is:

[0172]

[0173] Among them, P is the crushing power, D is the density of mutton, which is obtained through experimental tests. The density of mutton is measured using equipment such as a density meter. L is the tool movement distance, which is a crushing equipment parameter and is provided by the technical specifications of the equipment. μ is the friction coefficient, which is obtained from the technical information provided by the equipment manufacturer. g is the acceleration of gravity, which is a physical constant, usually 9.81m / s2. n is the rotation speed, which is obtained according to the vibration frequency. The vibration frequency is measured by a vibration sensor, and then the rotation speed of the equipment is adjusted to match the measured frequency. R is the tool radius, which is a physical parameter of the crushing equipment and is provided by the technical specifications of the equipment. v is the feed speed, which is obtained by setting the feed speed through the equipment control center and monitoring the feeding process in real time. E is the input energy. The energy input during the crushing process is deduced based on the acceleration data measured by the vibration sensor. f is the friction force, which is calculated based on the vibration amplitude and frequency measured by the vibration sensor. H is the material hardness. The hardness of mutton is deduced based on the real-time pressure data measured by the pressure sensor.

[0174] According to the obtained crushing power, the speed and strength of the crushing tool are adjusted to generate crushed mutton powder;

[0175] Specifically, the calculated actual crushing power is compared with the preset crushing target value;

[0176] Calculate the difference value (or error value) to determine the degree of deviation between the current crushing process and the target;

[0177] If the actual crushing power is lower than the crushing target value, it means that the crushing process is not efficient enough and the speed (rotation speed) of the crushing tool needs to be increased;

[0178] The control center increases the rotation speed of the crushing tool to make it close to the crushing target value;

[0179] If the actual crushing power is higher than the crushing target value, it means that the crushing process may be excessive and the pressure (force) applied by the crushing tool needs to be reduced;

[0180] The pressure applied by the crushing cutter is reduced through the control center to make it close to the crushing target value;

[0181] Use PID control algorithm to calculate the required proportional control term, integral control term and differential control term according to the current error value;

[0182] Summarize the proportional control term, the integral control term and the differential control term to generate a comprehensive control signal;

[0183] According to the comprehensive control signal, the speed and force of the crushing tool are adjusted;

[0184] Convey the ground mutton powder to the intelligent batching unit;

[0185] It should be noted that by using the PID control algorithm, the speed and force of the crushing tool can be monitored and adjusted in real time to ensure that the crushing process matches the characteristics of the raw materials, collect pressure data and vibration data in real time, calculate and adjust the crushing power, improve the crushing efficiency and consistency, reduce energy consumption, and ensure that the generated crushed mutton powder is of high quality and uniform.

[0186] In the mixing module, the intelligent batching unit receives the crushed mutton powder and adjusts the proportion of ingredients according to dynamic fuzzy logic control to generate a uniform mixture.

[0187] Furthermore, a weight sensor is installed in the intelligent batching unit to collect weight data (monitor the weight of each component in the crushed mutton powder) and transmit the weight data to the control center;

[0188] Use dynamic fuzzy logic control combined with weight data to dynamically adjust the proportion of each component in the weight data to ensure the accuracy and consistency of the mixture;

[0189] The control center adjusts the proportion of ingredients through dynamic fuzzy logic control, adjusts the amount of each ingredient, achieves precise batching, ensures that the amount of each batching process meets the preset formula ratio, and generates a uniform mixture;

[0190] Specifically, the dynamic fuzzy logic control input is defined as weight data, and the output is the ingredient ratio adjustment value;

[0191] The input weight data is fuzzified and divided into three fuzzy sets: low, medium and high;

[0192] A fuzzy rule base is established based on historical multimodal data and weight data to define the proportion of each ingredient;

[0193] If the ingredient weight is low, the amount of ingredients added is increased;

[0194] If the weight of the ingredient is medium, the amount of ingredients put in is reduced;

[0195] If the ingredient weight is high, the ingredient dosage remains unchanged;

[0196] According to the input weight data, the proportion of ingredients is dynamically adjusted. The expression is:

[0197]

[0198] Among them, T x (t+1) is the dynamic adjustment value of the ingredient ratio x at time t+1. This is the new ingredient ratio adjusted according to the current ingredient ratio and error. t is the time, x is the ingredient ratio, and it is used to distinguish the ingredient ratios of different components (for example, mutton powder, seasoning, additives, etc.). x(t) is the current value of the batching ratio x at time t, which is the existing batching ratio before adjustment. γ is the adjustment coefficient, which is used to control the adjustment step size. x (t) is the error value of the batching ratio x at time t, which represents the difference between the actual batching ratio and the preset target ratio. It is the standardized form of the error and is used to ensure that the adjustment amount is within a reasonable range;

[0199] Match the corresponding ingredient ratio according to the dynamically adjusted ingredient ratio;

[0200] The matched ingredient ratio is converted into a specific ingredient dosage adjustment value by weighted average method;

[0201] The control center issues a dosing instruction based on the specific ingredient dosing amount adjustment value to generate a uniform mixture (the dosing instruction is issued to the dosing device of each component through the control center, and the opening and dosing speed of the dosing device are adjusted in real time. The dosing unit puts each component in accordance with the instruction ratio and mixes them in the mixing unit to generate a uniform mixture);

[0202] During the batching process, the weight data is filtered through Kalman filtering to remove noise and outliers to ensure the accuracy and stability of the data;

[0203] The control center further accurately adjusts the proportion of each ingredient based on the filtered weight data to ensure the uniformity and consistency of the mixture;

[0204] It should be noted that the dynamic fuzzy logic control and weight sensor of the intelligent batching unit can monitor and dynamically adjust the batching ratio of each component in real time to ensure the accuracy and consistency of the mixture. Kalman filtering further ensures the accuracy of the data. This achieves accurate batching, generates a uniform mixture, improves product quality, and reduces errors and waste.

[0205] The regulating module transmits the uniform mixture to the temperature-controlled grinding unit, and uses Newton's method to adjust the temperature in real time to complete the grinding of the uniform mixture.

[0206] Furthermore, a temperature sensor is installed in the temperature-controlled grinding unit to monitor the temperature of the grinding, generate temperature data, and transmit the temperature data to the control center;

[0207] The control center receives the temperature data, sets the upper and lower limits of the temperature (for example, the temperature range is set to 40°C to 60°C, 40°C is the lower limit temperature, 60°C is the upper limit temperature, and a target temperature is determined therefrom, generally taking the middle value of the temperature range, such as 50°C), and automatically adjusts the temperature of the temperature-controlled grinding unit;

[0208] If the temperature exceeds the set upper limit, the heating power is reduced. For example, if an electric heater is used, the heating power can be reduced or increased by adjusting the current or voltage to reduce or increase the cooling intensity, such as starting or speeding up the cooling fan, or starting a water cooling device;

[0209] If the temperature is lower than the set lower limit, the heating power is increased, for example, by increasing the current or voltage to increase the power of the electric heater or the cooling intensity is reduced, for example, by reducing the speed of the cooling fan or shutting down the cooling device;

[0210] The control center continuously monitors temperature changes and makes adjustments based on real-time temperature data;

[0211] In the event of equipment failure in the temperature-controlled grinding unit, staff responded to the equipment failure by manually adjusting the temperature;

[0212] In special cases, operators can also manually adjust temperature settings to deal with special situations, such as equipment failure or special process requirements;

[0213] The control center sets the target temperature, observes the difference between the current temperature and the target temperature, and generates a temperature error. The smaller the error, the closer the current temperature is to the target temperature; the larger the error, the greater the degree of deviation of the current temperature from the target temperature.

[0214] Through the iterative calculation of Newton's method, the temperature optimization adjustment strategy is generated to obtain the heat transfer amount, which is expressed as:

[0215]

[0216] Wherein, O is the amount of heat transferred, which is the total amount of heat transferred during the grinding process, m is the mass of the material, which is the total mass of the material processed during the grinding process, h is the specific heat capacity of the material, which represents the amount of heat required to raise the unit temperature per unit mass of the material, ΔW is the temperature change, which is the change in the temperature of the material during the grinding process, P is the crushing power, β is the correction factor, which is used to correct the error in the calculation, Y is the fluid velocity, which is the speed of cooling or heating the fluid during the grinding process, A is the radius of the fluid path, which is the radius of the fluid path in the grinding equipment, g is the acceleration of gravity, ∈ is the thermal conductivity, which is the thermal conductivity of the material, which represents the ability of the material to conduct heat, ΔS is the height difference, which is the height change of the material during the grinding process, and t is time;

[0217] Specifically, the Newton method iteration will analyze the trend of temperature change and predict the direction and magnitude of future temperature changes. Based on these predictions, the control center will determine an optimal adjustment strategy, such as increasing or decreasing the heating power, or adjusting the working intensity of the cooling equipment to correct the temperature deviation problem;

[0218] The control center adjusts the temperature-controlled grinding unit equipment (heating or cooling equipment) according to the heat transfer amount obtained by the iteration of Newton's method. For example, if the temperature is predicted to rise too quickly, the control center will reduce the heating power or increase the cooling force in advance; if the temperature is predicted to drop too quickly, the control center will increase the heating power or reduce the cooling force in advance);

[0219] The uniform mixture is ground into powder and conveyed to the adjusted temperature-controlled grinding unit to complete the uniform mixture grinding, and the uniform mixture is ground into powder and conveyed to the centrifugal extraction unit.

[0220] It should be noted that by using Newton's method to adjust the temperature in real time in the temperature-controlled grinding unit, combined with temperature sensor monitoring and automatic temperature adjustment of the control center, precise temperature control during the grinding process is ensured. This mechanism not only prevents overheating or overcooling and protects the nutritional components of the raw materials, but also improves the grinding efficiency and product quality, ensuring the stability and consistency of the production process.

[0221] The extraction module, after completing the uniform mixture grinding, separates the original pulp through centrifugal force extraction, and optimizes it through Kalman filtering, filling and sealing, and completes automatic packaging.

[0222] Furthermore, the puree and solid particles in the homogeneous mixture are extracted and separated by centrifugal force after high-speed rotation;

[0223] The PSO algorithm is used to define the speed and time of the centrifugal force as a combined particle;

[0224] Initialize the combined particles and randomly set the position (centrifugal force speed and time) of each particle in the combined particles;

[0225] Calculate the fitness value of each particle in the combined particles (based on the efficiency and purity of the extracted pulp), find the optimal position of each particle (individual optimal position) and the optimal position of the combined particle group (global optimal position); the expression is:

[0226] Φ = η + α - τ;

[0227] Among them, Φ is the fitness value, which represents the comprehensive evaluation result of each particle. In the PSO algorithm, the higher the fitness value, the better the solution of the current particle. η is the extraction efficiency, which represents the proportion of puree extracted from the raw material in the centrifugal extraction unit. The higher the extraction efficiency, the more effective the centrifugal extraction process. α is the puree purity, which represents the amount of impurities contained in the extracted puree. The higher the puree purity, the purer the extracted puree. τ is the separation time, which represents the time required to complete centrifugal separation. The shorter the separation time, the more efficient the centrifugal extraction process.

[0228] A reasonable threshold is set according to industry standards. When the fitness value Φ is greater than or equal to the set threshold, it is optimal, indicating that the extraction efficiency and the purity of the original pulp are the highest, and the separation time is the shortest;

[0229] For example:

[0230] The lowest fitness value threshold is set to 30. When the fitness value Φ is greater than or equal to this threshold, it represents the best;

[0231] According to the position of each particle found and the position of the combined particle group, the position of each particle in the initialized combined particle is updated;

[0232] The final (i.e., the most adaptable) centrifugal force speed and time are found through the PSO algorithm iteration and input into the centrifugal force extraction unit;

[0233] The quality of the extracted puree is evaluated by the weighted average method, and the extraction parameters (including initialization particles, fitness value, individual optimal position, global optimal position, updating particle position and iterative optimization) are optimized according to the feedback (for example, if the extraction efficiency is low, the centrifugal force speed or time can be adjusted);

[0234] The extracted puree is filtered through a fine filter membrane to remove tiny particles and impurities, and the filtration is optimized through Kalman filtering to further remove residual impurities to obtain purified puree;

[0235] The purified puree is transported to the automatic packaging unit for filling, sealing and labeling to complete the packaging;

[0236] It should be noted that the centrifugal force after high-speed rotation is used to separate the raw pulp and solid particles, and the centrifugal force parameters are optimized with the PSO algorithm to ensure extraction efficiency and purity. The weighted average method and Kalman filtering are used to further optimize quality assessment and filtration to purify the raw pulp. Finally, the automatic packaging unit is used to complete the canning, sealing and labeling, which improves production efficiency and consistency of product quality.

[0237] In summary, the present invention realizes real-time optimization and efficient control of the crushing process by using the PID control algorithm. The PID control algorithm monitors and adjusts the speed and force of the crushing tool in real time by adjusting the proportional, integral and differential parameters to ensure the consistency and stability of the crushing effect. By collecting pressure and vibration data, the crushing power can be accurately calculated and adjusted according to the calculation results. This real-time adjustment mechanism not only improves the crushing efficiency, but also significantly reduces energy consumption and reduces resource waste. At the same time, the PID control algorithm can respond to changes in raw material characteristics and automatically adjust the crushing parameters to ensure the uniformity of the mutton powder after crushing, thereby improving product quality and enhancing overall adaptability and flexibility.

[0238] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of an integrated system for intelligent mutton grinding and puree extraction are provided.

[0239] First, select high-quality mutton, remove unnecessary parts such as bones and fascia, and cut the mutton into pieces of uniform size (about 50 grams) to facilitate the subsequent crushing process. Install multimodal sensors in the detection unit, including infrared spectroscopy analysis sensors, high-resolution cameras, pressure sensors, and vibration sensors to detect the internal fat content, moisture content, protein content, and surface fat distribution, color, shape, size, and texture changes of the mutton.

[0240] Secondly, configure the control center, analyze the multimodal data through the decision tree model, screen out the crushing raw materials that meet the requirements, configure the adaptive crushing unit, use the PID control algorithm to dynamically adjust the crushing tool parameters to achieve efficient crushing, and adjust the temperature sensor of the temperature-controlled grinding unit to ensure that the system can monitor and adjust the temperature in real time.

[0241] Next, the processed mutton is sent to the multimodal sensor detection unit through a conveyor. The sensor data is used to construct a comprehensive data set, and data preprocessing (noise removal, standardization, etc.) is performed to build a decision tree model. The comprehensive data set is divided into multiple subsets through recursive splitting, and the splitting points are selected according to the principle of maximizing information gain to form a classification result. The accuracy of the decision tree model is evaluated through cross-validation, and the preprocessed data is input into the trained model to screen out qualified mutton.

[0242] Finally, the qualified mutton is transmitted to the adaptive crushing unit, and the speed and force target values ​​of the crushing tool are set (the initial speed target value is set to 2000RPM, and the force target value is set to 50N). The PID control algorithm is used to monitor and adjust the tool parameters in real time to ensure that the crushing effect matches the characteristics of the raw materials, generate high-quality crushed mutton powder, and transport it to the intelligent batching unit.

[0243] The details are shown in Table 1 below:

[0244] Table 1 Comparison of intelligent mutton powder grinding and puree extraction performance

[0245]

[0246] By analyzing the contents of the table, it can be seen that the present invention has significant advantages in various parameters. In terms of the accuracy of infrared spectral content detection and high-resolution camera detection, the present invention shows higher precision, which means that the detection of internal components and surface characteristics of mutton is more accurate. In terms of crushing efficiency, the present invention is significantly better than the prior art, showing higher production efficiency and consistency. At the same time, the present invention shows extremely low error in the error rate of ingredients, greatly improving the uniformity and consistency of the mixture. Finally, in terms of extraction purity, the present invention significantly improves the purity of the raw pulp by optimizing centrifugal force extraction parameters and using advanced filtration technology, thereby ensuring the high quality of the final product. These results, taken together, illustrate the obvious advantages of the present invention in processing efficiency, detection accuracy and product quality.

[0247] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent integrated system for mutton grinding and puree extraction, characterized in that: include, Screening module: the control center collects multimodal data through multimodal sensors, selects the crushed raw materials through the decision tree model and transmits them to the adaptive crushing unit; The adjustment module, the crushed raw materials enter the adaptive crushing unit, use the PID control algorithm to dynamically adjust the crushing tool parameters, calculate the crushing power, and obtain the crushed mutton powder; In the mixing module, the intelligent batching unit receives the crushed mutton powder and adjusts the proportion of ingredients according to dynamic fuzzy logic control to generate a uniform mixture; The regulating module transfers the uniform mixture to the temperature-controlled grinding unit, adjusts the temperature in real time using Newton's method, and completes the grinding of the uniform mixture; The extraction module, after completing the grinding of the uniform mixture, extracts and separates the original puree through centrifugal force, optimizes it using Kalman filtering, and automatically packages the optimized original puree.

2. The intelligent integrated system for mutton grinding and puree extraction according to claim 1, characterized in that: The multimodal sensor includes an infrared spectrum analysis sensor, a high-resolution camera, a pressure sensor and a vibration sensor.

3. The intelligent integrated system for mutton grinding and pulp extraction as claimed in claim 2, characterized in that: The collecting of multimodal data comprises the following steps: Collect mutton fat content data, mutton moisture content data and mutton protein content data through infrared spectrum analysis sensor; The high-resolution camera is used to collect the fat distribution data on the surface of mutton, the color data of mutton, the color data of mutton, the shape and size data of mutton, and the texture feature data of mutton; Collect pressure data through pressure sensors; Vibration data is collected through vibration sensors.

4. The intelligent integrated system for mutton grinding and pulp extraction as claimed in claim 3 is characterized in that: The crushed raw materials are screened out through the decision tree model and transmitted to the adaptive crushing unit, and the specific steps are: The mutton fat content data, mutton moisture content data and mutton protein content data collected by the infrared spectrum analysis sensor and the mutton surface fat distribution data, mutton color data, mutton color data, mutton shape and size data and mutton texture feature data collected by the high-resolution camera are aggregated to generate a comprehensive data set; Preprocess the data in the comprehensive dataset; The decision tree model is trained by recursive splitting to divide the comprehensive data set into multiple subsets to form classification results; Use cross-validation to evaluate the decision tree model, and input the preprocessed comprehensive data set into the trained decision tree model to perform actual classification judgment and generate classification results; The decision tree model judges the quality of mutton based on the classification results; For qualified mutton, the qualified mutton is conveyed to the adaptive crushing unit through the conveying equipment, and for unqualified mutton, it is directly discarded; The qualified mutton is transported to the adaptive crushing unit as the crushing raw material.

5. The intelligent integrated system for mutton grinding and pulp extraction as claimed in claim 4 is characterized in that: The crushed raw material enters the adaptive crushing unit, uses the PID control algorithm to dynamically adjust the crushing tool parameters, calculates the crushing power, and obtains the crushed mutton powder. The specific steps are: The hardness of the mutton pulverization is monitored in real time in the adaptive pulverization unit through the real-time pressure, average pressure, pressure fluctuation and pressure intensity in the collected pressure data; The texture of lamb mince was monitored in real time through the vibration frequency, amplitude, acceleration and vibration pattern in the collected vibration data; Using PID control algorithm, the speed and strength of the crushing tool can be adjusted in real time by adjusting the proportional control item, integral control item and differential control item; In the process of adjusting the crushing tool, the crushing power is calculated based on the feedback of pressure data and vibration data, and the expression is: Among them, P is the crushing power, D is the density of mutton, L is the tool moving distance, μ is the friction coefficient, n is the rotation speed, R is the tool radius, v is the feed speed, E is the input energy, f is the friction force, and H is the material hardness; According to the obtained crushing power, the speed and strength of the crushing tool are adjusted to generate crushed mutton powder; The ground mutton powder is conveyed to the intelligent batching unit.

6. The intelligent integrated system for mutton powder grinding and puree extraction as claimed in claim 5, characterized in that: The PID control algorithm is used to adjust the speed and strength of the crushing tool in real time by adjusting the proportional control item, the integral control item and the differential control item. The specific steps are: Define the ideal values ​​of crushing tool speed and force as crushing set points; The pressure data and the vibration data are taken as actual measured values, and the error value is obtained according to the difference between the actual measured value and the crushing set value; According to the error value pairing adjustment ratio as a proportional control item, the speed and strength of the initial crushing tool are adjusted; Accumulate the current error value and the historical error value to obtain the integral control parameter as the integral control item to adjust the speed and strength of the current crushing tool; Calculate the difference between the current error value and the historical error value, obtain the differential control parameter as the differential control item, and predict the future error trend; The proportional control term, the integral control term and the differential control term are aggregated to generate a control signal; According to the control signal, the speed and strength of the crushing tool are adjusted in real time.

7. The intelligent integrated system for mutton grinding and pulp extraction according to claim 6, characterized in that: The intelligent batching unit receives the crushed mutton powder, adjusts the proportion of ingredients according to dynamic fuzzy logic control, and generates a uniform mixture. The specific steps are: Install a weight sensor in the intelligent batching unit to collect weight data and transmit the weight data to the control center; Dynamic fuzzy logic control is used in combination with weight data to dynamically adjust the proportion of each component in the weight data; The control center adjusts the proportion of ingredients through dynamic fuzzy logic control, regulates the amount of each ingredient, and generates a uniform mixture; During the batching process, the weight data is filtered by Kalman filtering to remove noise and outliers; The control center further adjusts the proportion of each component based on the filtered weight data.

8. The intelligent integrated system for mutton grinding and pulp extraction according to claim 7, characterized in that: The control center adjusts the proportion of ingredients through dynamic fuzzy logic control, adjusts the amount of each ingredient, and generates a uniform mixture. The specific steps are: Define the dynamic fuzzy logic control input as weight data and the output as the ingredient ratio adjustment value; The input weight data is fuzzified and divided into three fuzzy sets: low, medium and high; A fuzzy rule base is established based on historical multimodal data and weight data to define the proportion of each ingredient; If the ingredient weight is low, the amount of ingredients added is increased; If the weight of the ingredient is medium, the amount of ingredients put in is reduced; If the ingredient weight is high, the ingredient dosage remains unchanged; According to the input weight data, the proportion of ingredients is dynamically adjusted. The expression is: Among them, T x (t+1) is the dynamic adjustment value of the batching ratio x at time t+1, t is time, x is the batching ratio, T x (t) is the current value of the batching ratio x at time t, γ is the adjustment coefficient, e x (t) is the error value of the batching ratio x at time t; Match the corresponding ingredient ratio according to the dynamically adjusted ingredient ratio; The matched ingredient ratio is converted into a specific ingredient dosage adjustment value by weighted average method; The control center adjusts the value of the specific ingredient dosage and issues dosage instructions to generate a uniform mixture.

9. The intelligent integrated system for mutton grinding and pulp extraction as claimed in claim 8, characterized in that: The uniform mixture is transferred to the temperature-controlled grinding unit, and the temperature is adjusted in real time by Newton's method to complete the grinding of the uniform mixture. The specific steps are: A temperature sensor is installed in the temperature-controlled grinding unit to monitor the temperature of the grinding mill, generate temperature data, and transmit the temperature data to the control center; The control center receives the temperature data, sets the upper and lower temperature limits, and automatically adjusts the temperature of the temperature-controlled grinding unit; If the temperature exceeds the set upper limit, the heating power is reduced; If the temperature is lower than the set lower limit, the heating power is increased; The control center sets the target temperature, observes the difference between the current temperature and the target temperature, and generates a temperature error; Through the iterative calculation of Newton's method, the temperature optimization adjustment strategy is generated to obtain the heat transfer amount, which is expressed as: Where O is the amount of heat transferred, m is the mass of the material, h is the specific heat capacity of the material, ΔW is the temperature change, β is the correction factor, Y is the fluid velocity, A is the radius of the fluid path, g is the gravitational acceleration, ∈ is the thermal conductivity, and ΔS is the height difference; The control center adjusts the temperature-controlled grinding unit equipment according to the heat transfer amount obtained by Newton's method iteration; The uniform mixture is ground into powder and conveyed to the adjusted temperature-controlled grinding unit to complete the uniform mixture grinding.

10. The intelligent integrated system for mutton powder grinding and puree extraction according to claim 9, characterized in that: The method extracts and separates the puree by centrifugal force, optimizes it by using Kalman filtering, and automatically packages the optimized puree. The specific steps are: Extract and separate the puree and solid particles in the homogeneous mixture through the centrifugal force after high-speed rotation; The PSO algorithm is used to define the speed and time of the centrifugal force as a combined particle; Initialize the combined particles and randomly set the position of each particle in the combined particles; Calculate the fitness value of each particle in the combined particle, find the position of each particle and the position of the combined particle group; According to the position of each particle found and the position of the combined particle group, the position of each particle in the initialized combined particle is updated; The final centrifugal force speed and time are found through PSO algorithm iteration and input into the centrifugal force extraction unit; The quality of the extracted puree was evaluated by weighted average method, and the extraction parameters were optimized based on the feedback; The extracted puree is filtered through a fine filter membrane, and the filtration is optimized using a Kalman filter to obtain a purified puree; The purified puree is conveyed to the automatic packaging unit for filling, sealing and labeling.

Images