Intelligent control method and system for high-frequency vibration food material purification equipment

Abstract

The invention relates to the technical field of equipment control, and discloses an intelligent control method and system for high-frequency vibration food material purification equipment. The method comprises the following steps: optically scanning a food material pollution area to generate a pollution distribution diagram; pollution area parameters are quantitatively calculated, and a vibration purification parameter table is obtained; a vibration element is driven to form a directional cleaning waveform; collecting turbidity data to analyze the pollutant separation rate; vibration and flushing parameters are adjusted, and a linkage mechanism is established; the cleanliness is evaluated through outlet scanning, linkage parameters are optimized according to differences, and closed-loop control is achieved. Based on the real-time state information of food material pollution, accurate configuration of the vibration parameters, cooperative control of multiple purification technologies and closed-loop feedback optimization of the purification effect are achieved, and therefore the food material purification efficiency and effect and the energy utilization rate are improved.

Description

Technical Field

[0001] This application relates to the technical field of equipment control, and particularly to an intelligent control method and system for a high-frequency vibration food purification device. Background Art

[0002] Food purification is an important link in food processing and home cooking. Traditional food purification methods mainly rely on manual handwashing or simple mechanical rinsing, and these methods often have limited effects in removing pesticide residues, microbial contamination, and other harmful substances. With the development of technology, methods such as ultrasonic cleaning, ozone water cleaning, and ultraviolet sterilization have been introduced into the field of food purification. Among them, the high-frequency vibration technology has gradually become an important means of food purification due to its non-contact and low-damage characteristics. Existing high-frequency vibration purification devices usually adopt fixed frequency and intensity parameters, and the purification time is preset manually or the purification process is completed through simple timing control. Some advanced devices have introduced a preset program selection function based on food types, which can automatically adjust the vibration parameters for different foods, improving the pertinence of purification.

[0003] However, there are many deficiencies in existing high-frequency vibration food purification devices: First, there is a lack of accurate recognition ability for the pollution status on the surface of foods, and it is impossible to perform differential treatment for different degrees of pollution in different areas; second, the setting of vibration parameters lacks a scientific basis and is mostly set empirically, making it difficult to achieve the best purification effect; third, the purification process lacks a real-time monitoring and feedback adjustment mechanism, and it is impossible to dynamically adjust the working parameters according to the actual purification situation; fourth, the single vibration purification method has limited effect on removing stubborn pollution, and the combined application of multiple purification technologies lacks an effective collaborative control strategy; finally, the device cannot quantitatively evaluate and continuously optimize the purification effect, resulting in energy waste and food quality loss. These problems seriously restrict the application effect and promotion value of high-frequency vibration purification technology in the field of food safety. Summary of the Invention

[0004] This application provides an intelligent control method and system for a high-frequency vibration food purification device, which is used to realize the precise configuration of vibration parameters, the collaborative control of multiple purification technologies, and the closed-loop feedback optimization of the purification effect based on the real-time status information of food pollution, so as to improve the efficiency, effect, and energy utilization rate of food purification.

[0005] In a first aspect, the present application provides an intelligent control method for a high-frequency vibration food purification device. The intelligent control method for the high-frequency vibration food purification device includes: scanning the surface of the food for contamination degree through a scanning device disposed at the entrance of the purification device, digitally processing the obtained food surface image to obtain a food contamination area distribution map; quantitatively calculating the area and density of each contamination area according to the food contamination area distribution map to obtain a vibration purification intensity parameter table; based on the vibration purification intensity parameter table, driving and controlling the vibration elements in the purification cavity through a pulse signal generator to form a targeted vibration cleaning waveform; using a water turbidity detector installed on the wall of the purification cavity to collect the water turbidity value during the vibration process in real time, performing a time series analysis on the continuously collected turbidity data to obtain a pollutant detachment rate curve; adjusting the vibration frequency and the water flow flushing intensity proportionally according to the pollutant detachment rate curve to establish a vibration-flushing linkage mechanism; re-scanning the surface of the processed food through an imaging comparison system at the outlet, comparing and analyzing the obtained image with the food contamination area distribution map, and automatically optimizing the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference to form a closed-loop control system for the purification process.

[0006] In a second aspect, the present application provides an intelligent control system for a high-frequency vibration food purification device. The intelligent control system for the high-frequency vibration food purification device includes:

[0007] A scanning module, configured to scan the surface of the food for contamination degree through a scanning device disposed at the entrance of the purification device, digitally process the obtained food surface image to obtain a food contamination area distribution map;

[0008] A calculation module, configured to quantitatively calculate the area and density of each contamination area according to the food contamination area distribution map to obtain a vibration purification intensity parameter table;

[0009] A control module, configured to drive and control the vibration elements in the purification cavity through a pulse signal generator based on the vibration purification intensity parameter table to form a targeted vibration cleaning waveform;

[0010] An acquisition module, configured to use a water turbidity detector installed on the wall of the purification cavity to collect the water turbidity value during the vibration process in real time, perform a time series analysis on the continuously collected turbidity data to obtain a pollutant detachment rate curve;

[0011] An adjustment module, configured to adjust the vibration frequency and the water flow flushing intensity proportionally according to the pollutant detachment rate curve to establish a vibration-flushing linkage mechanism;

[0012] A scanning module, configured to re-scan the surface of the processed food materials through the imaging and comparison system at the outlet, compare and analyze the acquired images with the food material contamination area distribution map, and automatically optimize the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference, thereby forming a closed-loop control system for the purification process.

[0013] In a third aspect, a computer device is provided, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor invokes the instructions in the memory to cause the computer device to execute the intelligent control method for the high-frequency vibration food material purification device described above.

[0014] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the instructions are run on a computer, the computer is caused to execute the intelligent control method for the high-frequency vibration food material purification device described above.

[0015] In the technical solution provided by this application, by setting a scanning device at the inlet of the purification equipment, accurate scanning and digital processing of the surface contamination degree of food materials are realized, and a distribution map of the contaminated areas of food materials is generated, providing an accurate data basis for subsequent targeted purification; by quantitatively calculating the area and density of each contaminated area, a vibration purification intensity parameter table is obtained, realizing the refined configuration of purification parameters and avoiding the blindness and experience dependence of traditional equipment parameter settings; based on the vibration purification intensity parameter table, a pulse signal generator is used to drive and control the vibration elements in the purification chamber to form a targeted vibration cleaning waveform, significantly improving the utilization efficiency of purification energy and reducing unnecessary damage to food materials; a water quality turbidity detector installed on the wall of the purification chamber is used to collect the turbidity value of the water body in real time during the vibration process, and a pollutant detachment rate curve is obtained through time series analysis, establishing a real-time monitoring mechanism for the purification process and solving the technical defect that traditional equipment cannot perceive the purification progress; according to the pollutant detachment rate curve, the vibration frequency and the water flow flushing intensity are adjusted proportionally to establish a vibration-flushing linkage mechanism, realizing the synergistic effect of multiple purification technologies and greatly improving the ability to remove stubborn contaminants; the surface of the processed food materials is scanned again through the imaging comparison system at the outlet, compared and analyzed with the distribution map of the contaminated areas of food materials, and the operating parameters of the vibration-flushing linkage mechanism are automatically optimized according to the cleanliness difference, forming a closed-loop control system for the purification process and solving the technical problems that traditional equipment cannot perform effect evaluation and continuous optimization. Especially when applying the lightweight convolutional neural network algorithm to process the images of the food material surface, this algorithm fully considers the characteristics of complex surface texture and uneven illumination of food materials, extracts the features of the contaminated areas through multi-layer convolution and pooling operations, and significantly improves the accuracy of pollution recognition; when processing the pollutant detachment rate data, the Kalman filter algorithm used effectively suppresses the noise interference in the water turbidity measurement through the prediction-correction mechanism and enhances the data reliability; in the vibration-flushing linkage control, the fuzzy logic decision tree algorithm applied converts multi-dimensional information such as food material type, pollution characteristics, and purification stage into accurate control instructions, realizing intelligent decision-making under complex conditions; in the purification effect evaluation, the grey relational analysis method adopted adapts to the characteristics of small samples and incomplete information, and accurately quantifies the correlation degree between purification parameters and cleaning effects. The application of these algorithms and models in specific functional links greatly improves the intelligent level and adaptability of the system, enabling the present invention to automatically adjust the optimal purification strategy for different types and different pollution degrees of food materials, significantly improving the purification quality and efficiency, and reducing energy consumption. Brief Description of the Drawings

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic diagram of an embodiment of the intelligent control method for a high-frequency vibration food purification device in an embodiment of the present application;

[0018] Figure 2 It is a schematic diagram of an embodiment of the intelligent control system for a high-frequency vibration food purification device in an embodiment of the present application;

[0019] Figure 3 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. Specific Embodiments

[0020] The embodiments of the present application provide an intelligent control method and system for a high-frequency vibration food purification device. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the following describes the specific process of the embodiments of the present application. Please refer to Figure 1 An embodiment of the intelligent control method for a high-frequency vibration food purification device in an embodiment of the present application includes:

[0022] Step S101: Use a scanning device set at the entrance of the purification device to scan the contamination degree of the food surface, and digitally process the obtained food surface image to obtain a food contamination area distribution map;

[0023] Step S102: According to the food contamination area distribution map, quantitatively calculate the area and density of each contamination area to obtain a vibration purification intensity parameter table;

[0024] Step S103: Based on the vibration purification intensity parameter table, drive and control the vibration elements in the purification chamber through a pulse signal generator to form a targeted vibration cleaning waveform;

[0025] Step S104: Use the water quality turbidity detector installed on the purification chamber wall to collect the turbidity value of the water body in real time during the vibration process, perform time series analysis on the continuously collected turbidity data, and obtain the pollutant detachment rate curve;

[0026] Step S105: According to the pollutant detachment rate curve, adjust the vibration frequency and water flow flushing intensity proportionally to establish a vibration-flushing linkage mechanism;

[0027] Step S106: Use the imaging comparison system at the outlet to scan the surface of the processed food again, compare the obtained image with the food contamination area distribution map, automatically optimize the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference, and form a closed-loop control system for the purification process.

[0028] It can be understood that the execution subject of this application can be the intelligent control system of the high-frequency vibration food purification equipment, or it can also be a terminal or a server. Specifically, it is not limited here. In this embodiment of the application, the server is used as the execution subject for illustration.

[0029] Specifically, a scanning device is configured at the entrance of the purification equipment. This device includes a multi-spectral illumination unit and a high-resolution imaging device. When the food enters the equipment, the multi-spectral illumination unit irradiates the surface of the food comprehensively, collects the reflection data under different lighting conditions, and forms an original light intensity data set. This data set is processed through threshold segmentation technology to separate the suspected contamination areas where the surface brightness value is higher than the background value. At the same time, the high-resolution imaging device captures the color and texture features of the food surface, establishes a reference feature library of the food surface, and determines the actual contamination areas through the difference comparison with the suspected contamination area data. Finally, a food contamination area distribution map is generated. After obtaining the food contamination area distribution map, the system performs grid processing on it, divides the food surface into several micro-region units with equal areas, and constructs a contamination area grid coordinate system. On this basis, the number of contaminated pixel points in each micro-region is counted, and the contamination density value of each region is calculated. These values are normalized and then converted into a standardized contamination intensity index of 0-100. The system combines the regions with similar contamination intensity into the same contamination block according to the contamination intensity difference between adjacent micro-regions, calculates the total area and contamination load value of each block. For different contamination load values, the system queries and matches the parameters from the preset vibration frequency library and vibration intensity library, combines them to form a vibration parameter combination for each contamination block, and associates the position information of the contamination block with the vibration parameters to generate a vibration purification intensity parameter table.

[0030] According to the vibration purification intensity parameter table, the zoning control of vibration elements is carried out. First, the data in the parameter table is subjected to spatial mapping to generate a zoning control mapping table for the vibration elements in the purification chamber. Multiple vibration elements are divided into several independent control groups to form a vibration area control unit. Each control unit is assigned a corresponding vibration frequency and intensity coefficient, which are converted into a sub-region vibration control instruction sequence. These instruction sequences are input into the pulse waveform generation module, converted into digital pulse sequences, and after power amplification, form vibration element drive signals. These signals are transmitted to the corresponding vibration area control unit through a multi-channel transmission system to form a vibration field with multi-region independent control. The system also performs phase synchronization adjustment on the vibration field to eliminate the interference of vibration waves in adjacent regions, establish a vibration wave array for collaborative work, and dynamically adjust the propagation direction of vibration waves in each region according to the position change of the ingredients in the purification chamber to form a targeted vibration cleaning waveform.

[0031] During the purification process, the water turbidity detectors installed on the purification chamber wall collect the water turbidity values in real time. The detectors with multi-point distribution sample at fixed time intervals to obtain the original water turbidity data points, which are processed by digital filtering to form a smoothed turbidity data sequence. The system divides these data into multiple consecutive time windows along the time axis, calculates the average turbidity value of each time window to obtain the set of turbidity values for each time period. By calculating the turbidity difference between adjacent time windows, the turbidity change amount data for each time period is obtained, and then divided by the corresponding time window length to calculate the turbidity change rate per unit time, forming the initial detachment rate data points. The system correlates these data points with the current vibration parameters, establishes a correspondence table between turbidity change and vibration parameters, fits the functional relationship between the turbidity change rate and the purification time, and generates a turbidity time-varying model. Combining the parameters of the purified water volume and the surface area of the ingredients, the system calculates and obtains the pollutant detachment rate curve.

[0032] According to the pollutant detachment rate curve, the system dynamically adjusts the vibration-flushing linkage mechanism. First, calculate the curve slope, determine the change inflection point, divide the purification process into an initial acceleration stage, a stable detachment stage, and a decelerating end stage, and establish a pollutant detachment feature library for each stage. Match the real-time data points with the feature library to determine the current stage type, extract the frequency adjustment coefficient accordingly, and calculate the optimal vibration frequency value. At the same time, combined with the change trend of the pollutant detachment rate, determine the starting time and basic intensity parameters of the water flow flushing, and calculate the dynamic adjustment coefficient of the water flow flushing intensity through the ratio of the current pollutant detachment rate to the historical peak value. Combine the vibration frequency value and the water flow intensity coefficient into a parameter pair, generate a vibration-flushing parameter matching table, establish a signal triggering relationship between the vibration control unit and the water flow control unit, and form a vibration-flushing linkage mechanism. The imaging comparison system at the outlet scans the surface of the processed food ingredients again to obtain a cleanliness state image. Through light correction and geometric transformation, make it have the same coordinate system and scale as the initial pollution area distribution map, and form a standardized post-treatment image. The two images are superposed pixel by pixel, calculate the pollutant removal ratio of each area, and generate a cleanliness distribution matrix. The system conducts statistical analysis on this matrix to determine the areas with uneven cleanliness and insufficient cleanliness, trace back the corresponding vibration parameters and flushing parameters, and establish an association database between the cleaning effect and the process parameters. For the areas with cleanliness lower than the threshold, the system generates vibration frequency correction values and water flow intensity correction values, updates the vibration-flushing parameter matching table, and transmits them to the control center through the parameter feedback channel to adjust the processing parameters of the next batch of food ingredients, forming a closed-loop control system for the purification process.

[0033] For example, when a batch of leafy vegetables enters the purification system, the scanning device collects reflection spectrum data and identifies the surface organic pollutant distribution area. The system divides its surface into 100 micro-region units, calculates that the pollution density of the 37th region is 78, the corresponding vibration frequency is 32 kHz, and the vibration intensity coefficient is 0.85. During the purification process, the turbidity detector data corresponding to the 37th region shows that the turbidity value rises rapidly within the first 30 seconds and then levels off. The system determines that it has entered the stable detachment stage, adjusts the vibration frequency to 36 kHz, and simultaneously starts the water flow flushing with the intensity set at 65% of the medium level. After 120 seconds of processing, the outlet imaging system detects that the cleanliness of the 37th region reaches 94%, while that of the adjacent 36th region is only 87%. Based on this, the system optimizes the vibration-flushing parameters of the 36th region and increases the water flow intensity of this region to 75% in the next batch of processing to continuously optimize the cleaning effect.

[0034] In the embodiments of the present application, by setting a scanning device at the inlet of the purification device, accurate scanning and digital processing of the surface contamination degree of food materials are realized, a distribution map of the contaminated areas of the food materials is generated, providing an accurate data basis for subsequent targeted purification; by quantitatively calculating the area and density of each contaminated area, a vibration purification intensity parameter table is obtained, realizing the refined configuration of purification parameters and avoiding the blindness and experience dependence of traditional equipment parameter settings; based on the vibration purification intensity parameter table, a pulse signal generator is used to drive and control the vibration elements in the purification chamber to form a targeted vibration cleaning waveform, significantly improving the utilization efficiency of purification energy and reducing unnecessary damage to the food materials; a water turbidity detector installed on the wall of the purification chamber is used to collect the water turbidity value in real time during the vibration process, and a pollutant detachment rate curve is obtained through time series analysis, establishing a real-time monitoring mechanism for the purification process and solving the technical defect that traditional equipment cannot sense the purification progress; according to the pollutant detachment rate curve, the vibration frequency and the water flow flushing intensity are adjusted proportionally to establish a vibration-flushing linkage mechanism, realizing the synergistic effect of multiple purification technologies and greatly improving the ability to remove stubborn contaminants; the surface of the processed food materials is scanned again through the imaging comparison system at the outlet, compared and analyzed with the distribution map of the contaminated areas of the food materials, and the operation parameters of the vibration-flushing linkage mechanism are automatically optimized according to the cleanliness difference, forming a closed-loop control system for the purification process and solving the technical problems that traditional equipment cannot evaluate the effect and continuously optimize. Especially when applying the lightweight convolutional neural network algorithm to process the images of the food material surface, this algorithm fully considers the characteristics of complex surface texture and uneven illumination of the food materials, and extracts the features of the contaminated areas through multi-layer convolution and pooling operations, significantly improving the accuracy of contamination recognition; when processing the pollutant detachment rate data, the Kalman filter algorithm used effectively suppresses the noise interference in the water turbidity measurement through the prediction-correction mechanism, enhancing the data reliability; in the vibration-flushing linkage control, the fuzzy logic decision tree algorithm applied converts multi-dimensional information such as food material type, contamination characteristics, and purification stage into accurate control instructions, realizing intelligent decision-making under complex conditions; in the purification effect evaluation, the grey relational analysis method adopted adapts to the characteristics of small samples and incomplete information, accurately quantifying the correlation degree between the purification parameters and the cleaning effect. The application of these algorithms and models in specific functional links greatly improves the intelligent level and adaptability of the system, enabling the present invention to automatically adjust the optimal purification strategy for different types and different contamination degrees of food materials, significantly improving the purification quality and efficiency, and reducing energy consumption.

[0035] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0036] (1) The multi - spectral illumination unit irradiates the surface of the food material in all directions, collects the surface reflection data of the food material under different lighting conditions, and forms an original light intensity data set;

[0037] (2) According to the original light intensity data set, threshold segmentation is performed on the surface brightness value to screen out the data points of the suspected pollution area that are higher than the background value;

[0038] (3) The high - resolution imaging device is used to take multi - angle photos of the food material surface, obtain the surface color and texture information, and establish a reference feature library of the food material surface;

[0039] (4) The data points of the suspected pollution area are compared with the reference feature library of the food material surface for differences, and the areas with large deviations from the reference features are extracted to obtain a candidate set of pollution area points;

[0040] (5) Density clustering is performed on the candidate set of pollution area points, and the points with adjacent distances less than the preset threshold are grouped into the same pollution area to generate the boundary coordinates of the pollution area;

[0041] (6) Based on the boundary coordinates of the pollution area, combined with the color depth value within the pollution area, a distribution map of the food material pollution area is generated through coordinate mapping.

[0042] Specifically, the multi - spectral illumination unit irradiates the surface of the food material in all directions. Here, the multi - spectral illumination unit refers to an illumination device equipped with light sources of different wavelengths, including visible light, near - infrared light, and ultraviolet light sources. When the food material passes through the conveyor belt at the entrance, these light sources of different wavelengths irradiate the surface of the food material in a preset sequence. Each wavelength of light has specific reflection characteristics for different types of pollutants. The reflected light signals are received by the photoelectric sensor array, and the reflection intensity values of the food material surface under each spectrum are recorded. These values are organized into a three - dimensional data structure according to the coordinate positions of the food material surface to form an original light intensity data set. This data set contains the reflection intensity values of each tiny area on the food material surface at each wavelength, providing a multi - dimensional information basis for subsequent pollution identification. Threshold segmentation processing is performed on the original light intensity data set, which is a basic technique for dividing an image into foreground and background. First, the average reflection intensity value of the pollution - free area on the food material surface is calculated as the background value, and then the difference threshold is set according to the spectral characteristics of different pollutant types. By traversing each point in the original light intensity data set, the reflection intensity value is compared with the background value. If the reflection intensity deviation at a specific wavelength exceeds the preset threshold, the point is marked as a suspected pollution point. The preset threshold is a discriminant value obtained based on a large amount of experimental data statistics. For example, for organic residues, the reflection deviation threshold in the near - infrared band is set to ±15% of the background value; for pesticide residues, the reflection deviation threshold in the ultraviolet band is set to ±20% of the background value. The output of this step is a set of data points of the suspected pollution area, including the coordinate positions of each suspected pollution point and the corresponding reflection intensity deviation value.

[0043] A high-resolution imaging device takes multi-angle pictures of the surface of food ingredients. The high-resolution imaging device refers to an industrial camera with a resolution of no less than 12 million pixels, equipped with an angle-adjustable mounting mechanism, and capable of capturing images of the surface of food ingredients from different perspectives. The device takes pictures of the surface of food ingredients from at least three different angles (usually 0°, 45°, and 90°) to obtain high-precision color images. After these images are subjected to color correction processing, the color information (RGB values) and texture features (such as roughness, glossiness, etc.) of the surface of the food ingredients are extracted. For each type of food ingredient, a standard color range and texture feature descriptor in a pollution-free state of its surface are established and stored in the reference feature library of the surface of food ingredients. This feature library is a structured data set containing standard appearance feature parameters of different types of food ingredients, serving as a reference basis for judging pollution. The data points in the suspected pollution area are compared with the reference feature library of the surface of food ingredients for differences. For each suspected pollution data point, the high-resolution image data at its corresponding position is taken, and the color value and texture feature value of this point are extracted and numerically compared with the standard features of this type of food ingredient in the reference feature library. The Euclidean distance or Mahalanobis distance of the feature values is calculated to quantify the degree of feature deviation. A feature deviation threshold is set. When the feature deviation of a certain point exceeds the threshold, it is identified as a candidate point for the pollution area. The feature deviation threshold is a discrimination criterion set according to the type of food ingredient. For example, for food ingredients with a smooth surface (such as apples), the color deviation threshold is relatively low, set within 15% of the total deviation of the RGB three channels; while for food ingredients with an irregular surface (such as cauliflower), the deviation threshold is relatively high, set within 25%. Through this step, a set of candidate points for the real pollution area is screened out, reducing false positive judgments caused by factors such as light changes and surface unevenness.

[0044] Density clustering processing is performed on the set of candidate points for the pollution area. Density clustering is a clustering algorithm based on the density distribution of data points and is suitable for identifying clusters of any shape. In this method, first, an influence radius is set for each candidate point for the pollution area. This radius value is set according to the type of food ingredient and the expected size of the pollutant, usually between 0.5 mm and 2 mm. The number of neighbor points within the influence radius of each point is calculated to determine the density value of the point. Points with high density values are selected as clustering centers, and through the connection relationship between points, points with an adjacent distance less than a preset threshold are grouped into the same pollution area. Here, the preset threshold refers to the Euclidean distance threshold between two points, usually set to 1.5 times the influence radius. Finally, several independent groups of pollution areas are obtained, and the set of peripheral boundary points of each area is calculated to form the boundary coordinates of the pollution area.

[0045] Based on the boundary coordinates of the contaminated area and combined with the color depth values within the contaminated area, a distribution map of the food ingredient contaminated area is generated through coordinate mapping. The surface of the food ingredient is mapped onto a two-dimensional plane coordinate system, and a surface grid is established at an appropriate resolution (usually 0.1 mm / pixel). For each contaminated area, the area contour is determined according to its boundary coordinates, and the internal grid points of the area are filled. Different contamination degree values are assigned according to the color depth values of each point (converted from RGB values), forming an area coloring map with contamination degree information. The finally generated distribution map of the food ingredient contaminated area is a two-dimensional image containing dual information of position and intensity, intuitively showing the spatial distribution and severity of the pollutants.

[0046] For example, when a batch of leafy vegetables pass through the inlet of the purification device, the multi-spectral lighting unit irradiates the surface of the leafy vegetables with visible light, 365 nm ultraviolet light, and 850 nm near-infrared light in sequence, and the photoelectric sensor records the reflection data at each wavelength. Under ultraviolet light, the areas with pesticide residues show obvious fluorescence reactions, and the reflection intensity is 30% higher than the background value. These points are marked as suspected contamination points. The high-resolution camera takes pictures of the surface of the leafy vegetables from three angles to establish the standard green value range and surface texture characteristics of the healthy leafy vegetable surface. By comparing the suspected contamination points with the standard characteristics, it is found that some areas not only have abnormal reactions under ultraviolet light, but their RGB values also deviate from the standard range of normal leafy vegetables by up to 22%. These points are determined as candidate contamination areas. Through the density clustering algorithm, with an influence radius of 1.2 mm and a clustering distance threshold of 1.8 mm, the adjacent candidate contamination points are classified, and finally three independent contaminated areas are identified, located at the leaf edge, near the leaf vein, and in the center of the leaf surface respectively, forming their respective boundary coordinates. According to the degree of RGB deviation within the contaminated area, a contamination degree distribution map is drawn. The darker areas indicate a higher degree of contamination and require a stronger vibration cleaning intensity. This precise contamination distribution information directly guides the configuration of subsequent vibration purification intensity parameters to ensure the pertinence and efficiency of the purification process.

[0047] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0048] (1) Perform grid division on the distribution map of the food ingredient contaminated area, divide the surface of the food ingredient into several micro-region units with equal areas, and form a contaminated area grid coordinate system;

[0049] (2) According to the contaminated area grid coordinate system, count the number of contaminated pixel points in each micro-region unit to obtain the contamination density values of each micro-region, and perform normalization processing on the contamination density values of each micro-region to convert the contamination density into a standardized contamination intensity index of 0 - 100;

[0050] (3) Based on the standardized pollution intensity index, cluster and group adjacent micro-regions, merge micro-regions with pollution intensity differences less than the threshold into the same pollution block, and perform pixel accumulation calculation on the total area of each pollution block. Combine with the standardized pollution intensity index to generate the pollution load value of each pollution block;

[0051] (4) According to the pollution load value, query the matching optimal vibration frequency range from the vibration frequency library and query the matching vibration intensity coefficient from the vibration intensity library;

[0052] (5) Combine and pair the optimal vibration frequency range with the vibration intensity coefficient to form a vibration parameter combination for each pollution block;

[0053] (6) Map the coordinate position information of each pollution block to the corresponding vibration parameter combination to generate a vibration purification intensity parameter table.

[0054] Specifically, perform grid division processing, divide the surface of the food material equally according to the preset grid size, and create a regular two-dimensional grid structure. Grid division refers to the process of discretizing a continuous two-dimensional space into a finite number of grid cells. The surface of the food material is divided into square micro-region units with equal side lengths, and the area of each unit is usually 1mm 2 to 4mm 2 , and the specific size is determined according to the type of food material and the characteristics of the pollutant. For example, for smooth-surfaced fruits, smaller grid cells can be used; for root vegetables with uneven surfaces, larger grid cells are used. The pollution area grid coordinate system formed after grid division is a two-dimensional rectangular coordinate system, and each micro-region unit has a unique coordinate identifier.

[0055] Based on the established pollution area grid coordinate system, quantitatively analyze the pollution situation in each micro-region unit. The specific operation is to overlay the food material pollution area distribution map on the grid coordinate system, count the number of pixel points marked as polluted in each micro-region unit, and calculate the original pollution density value. The original pollution density value directly reflects the pollution degree in the micro-region. However, due to the large differences in the surface area and total pollution amount of different food materials, normalization processing is required to make the data comparable. The normalization processing uses a linear mapping method to map the original pollution density value to the standard range of 0 - 100. The calculation process is to first determine the maximum pollution density value and the minimum pollution density value in all micro-regions, and then map the original pollution density value of each micro-region proportionally to obtain the standardized pollution intensity index. This normalization processing ensures the comparability of pollution degrees between different food materials and different batches.

[0056] Based on the standardized pollution intensity index, adjacent micro-regions are clustered and grouped. The purpose of clustering and grouping is to merge micro-regions with similar pollution characteristics into larger pollution blocks, reducing the computational complexity and facilitating subsequent parameter configuration. The clustering process uses the region growing method, starting from the micro-region with the highest pollution intensity and gradually examining its adjacent micro-regions around it. If the difference between the pollution intensity index of the adjacent micro-region and the current region is less than a preset threshold (usually set to 10 - 15), it is merged into the current pollution block. The selection of the preset threshold is based on the continuity characteristics of pollutant distribution. A smaller threshold will result in more fragmented pollution blocks, while a larger threshold will form fewer but larger blocks. For each merged pollution block, calculate its total area (the number of micro-region units multiplied by the area of a single micro-region) and accumulate the standardized pollution intensity index of all micro-regions within the block to obtain the pollution load value. The pollution load value comprehensively considers two factors: the block area and the pollution degree.

[0057] According to the pollution load value of the pollution block, corresponding vibration parameters need to be determined. In this solution, a vibration frequency library and a vibration intensity library are established. These two libraries store the optimal vibration frequency ranges and vibration intensity coefficients corresponding to different types of pollutants respectively. The vibration frequency library is established based on a large amount of experimental data, recording the vibration frequency ranges at which different pollutants are most easily detached under different adhesion conditions. For example, for oil-based pollutants, the optimal vibration frequency range is 25 - 30 kHz; for pesticide residues, the optimal vibration frequency range is 32 - 38 kHz; for microbial adhesion, the optimal vibration frequency range is 40 - 45 kHz. Similarly, the vibration intensity library records the vibration intensity coefficients corresponding to different pollution load values. The vibration intensity coefficient is a value between 0 and 1, used to adjust the output power of the vibration element. According to the pollution load value of each pollution block, the optimal vibration frequency range and vibration intensity coefficient are obtained from these two parameter libraries through table lookup or interpolation calculation.

[0058] The obtained optimal vibration frequency range and vibration intensity coefficient are combined and paired to form a vibration parameter combination for each pollution block. The vibration parameter combination includes multiple parameters such as the upper limit of vibration frequency, the lower limit of vibration frequency, the vibration intensity coefficient, and the vibration duration. These parameters jointly determine the vibration cleaning effect for a specific pollution block. The parameter pairing process considers the mutual influence between different parameters. For example, a higher frequency usually needs to be combined with a lower intensity to avoid damaging the food ingredients. Finally, the coordinate position information of each pollution block (including the coordinates of the block center point and the boundary coordinates) is associated and mapped with the corresponding vibration parameter combination to generate a vibration purification intensity parameter table. This parameter table is a structured data set, containing multiple data items. Each data item corresponds to a pollution block, recording the block location, area, pollution load, and the corresponding vibration parameter combination.

[0059] For example: When an apple passes through the inlet of the purification device, a distribution map of the contaminated areas on its surface is obtained through optical scanning. The surface of the apple is divided into a 10×10 grid, with a total of 100 micro-region units, and the area of each unit is 2 mm 2 . It is statistically found that 78 pixel points are marked as contaminated in the micro-region unit at the coordinate (3,4), while the maximum contamination density in the entire grid is 120 pixel points and the minimum is 0 pixel points. Through normalization processing, the standardized contamination intensity index of this micro-region is calculated as (78 - 0) / (120 - 0)×100 = 65. Then, clustering is performed on adjacent micro-regions, and it is found that the contamination intensity indices of the four micro-regions at coordinates (3,4), (3,5), (4,4), and (4,5) are 65, 62, 59, and 63 respectively. The maximum difference between them is 6, which is less than the preset clustering threshold of 12. Therefore, these four micro-regions are merged into one contaminated block. The total area of this block is 4×2 = 8 mm 2 , and the average contamination intensity index is (65 + 62 + 59 + 63) / 4 = 62.25. From this, the contamination load value is calculated as 8×

[0060] 62.25 = 498. By querying the vibration parameter library, the optimal vibration frequency range corresponding to this contamination load value is 33 - 37 kHz, and the vibration intensity coefficient is 0.75. The central point coordinates (3.5,4.5) of this block are associated with the vibration parameter combination and recorded in the vibration purification intensity parameter table. This parameter table ultimately guides the vibration elements to apply precisely controlled vibrations to different areas of the apple surface, achieving efficient removal of contaminants while minimizing damage to the food ingredients themselves.

[0061] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0062] (1) Perform spatial mapping on the vibration parameters of each contaminated block in the vibration purification intensity parameter table to generate a partition control mapping table for the vibration elements in the purification chamber. According to the partition control mapping table, divide the multiple vibration elements in the purification chamber into several independent control groups to form a vibration area control unit;

[0063] (2) Assign corresponding vibration frequency values and vibration intensity coefficients to the vibration area control unit to generate a sub-region vibration control instruction sequence;

[0064] (3) Input the sub-region vibration control instruction sequence into the pulse waveform generation module to convert it into a digital pulse sequence of voltage-time relationship;

[0065] (4) Perform power amplification processing on the digital pulse sequence, adjust the voltage amplitude and current intensity of the output signal to generate a vibration element drive signal;

[0066] (5) Transmit the vibration element drive signal through the sub-channel transmission system to the corresponding vibration area control unit, form a vibration field with multi-area independent control, perform phase synchronization adjustment on the vibration field with multi-area independent control, eliminate the interference phenomenon of vibration waves in adjacent areas, and establish a vibration wave array for collaborative work;

[0067] (6) Based on the vibration wave array for collaborative work, dynamically adjust the propagation direction of vibration waves in each area according to the position change of the food ingredients in the purification chamber to form a targeted vibration cleaning waveform.

[0068] Specifically, perform spatial mapping processing on the vibration parameters of each pollution block in the vibration purification intensity parameter table. This process is to convert the pollution block information on the two-dimensional plane into the control information of the vibration elements in the three-dimensional space. Spatial mapping uses a coordinate transformation algorithm to map the position coordinates of the pollution blocks on the surface of the food ingredients to the physical space coordinates in the purification chamber. A plurality of vibration elements are pre-installed in the purification chamber, and each vibration element has a unique spatial coordinate identifier. By calculating the spatial distance between the center point of the pollution block and each vibration element, determine which vibration elements affect each pollution block, and generate a partition control mapping table of the vibration elements in the purification chamber. The partition control mapping table is a data matrix that records the pollution block information corresponding to the control of each vibration element and the spatial relationship between the elements. Based on the partition control mapping table, divide the multiple vibration elements in the purification chamber into several independent control groups, and each control group is responsible for the cleaning work of a specific pollution area to form a vibration area control unit. The vibration area control unit refers to a set of vibration elements that are physically adjacent and logically act together on a specific pollution area. The division of the control unit follows the principle of the nearest influence, that is, the control unit is preferably composed of the vibration elements closest to the pollution block. For each vibration area control unit, extract the vibration frequency range and vibration intensity coefficient of the corresponding pollution block from the vibration purification intensity parameter table to determine the actual parameter values to be executed. The vibration frequency value usually takes the median of the vibration frequency range, and the vibration intensity coefficient directly uses the value in the parameter table. Organize these parameter values in a time series to form an instruction sequence. Each instruction contains information such as start time, end time, frequency value, intensity coefficient, etc., to form a sub-area vibration control instruction sequence.

[0069] The sub-area vibration control instruction sequence needs to be converted into specific electronic signals to drive the vibration elements to work. Input the instruction sequence into the pulse waveform generation module, which is a digital signal processor that can generate electronic signals with a specific waveform according to the input parameters. In the conversion process, first convert the frequency value into a period value, calculate the number of sampling points in each period, and then calculate the amplitude of each sampling point according to the specified waveform type (usually a sine wave), and finally generate a series of data points representing the change of voltage with time, that is, a digital pulse sequence. The digital pulse sequence is a low-voltage signal that needs to be power-amplified to drive the vibration elements.

[0070] Perform power amplification processing on the digital pulse sequence, adjust the voltage amplitude and current intensity of the output signal, so that it can drive the vibration element to generate mechanical vibration with sufficient intensity. The power amplification process includes two links: voltage amplification and current gain. Voltage amplification raises the low-level control signal to the working voltage range required by the vibration element (usually 30 - 150V), and current gain ensures the provision of sufficient driving current (usually 0.5 - 2A). During the amplification process, the gain level is dynamically adjusted according to the vibration intensity coefficient to output driving energy matching the instruction requirements. The amplified signal is the driving signal for the vibration element, which is a high-power electronic signal that can directly drive piezoelectric ceramics or other types of vibration elements to work.

[0071] Transmit the driving signal of the vibration element to the corresponding vibration area control unit through a sub-channel transmission system. The sub-channel transmission system is a multiplexed signal transmission network that can ensure that different driving signals are accurately sent to the target vibration element, avoiding signal aliasing and interference. Each vibration area control unit receives an exclusive driving signal and independently controls the vibration state of its respective responsible area, forming a vibration field with multi-area independent control. The vibration field with multi-area independent control means that each area in the purification chamber can vibrate at different frequencies, different intensities, and different phases. Since there may be spatial overlap or proximity between the vibration areas, interference will occur between adjacent vibration waves, affecting the purification effect.

[0072] In order to eliminate the interference of adjacent area vibration waves, it is necessary to perform phase synchronization adjustment on the multi-area vibration field. The phase synchronization adjustment calculates the spatial relationship and vibration frequency difference between the vibration areas, and assigns a specific phase offset value to each vibration element, so that the vibration waves in adjacent areas form a synergistic enhancement rather than mutual cancellation effect in space. The adjusted vibration waves form a vibration wave array that works in coordination, which is an ordered vibration field structure formed by precisely controlling the phase relationship of multiple vibration sources, and can form the focusing or uniform distribution of vibration energy in a specific area.

[0073] Vibration wave array based on collaborative work. During the purification process of food ingredients, they will continuously move and rotate, and it is necessary to dynamically adjust the propagation directions of vibration waves in each area. The monitoring of the position change of food ingredients is achieved through an optoelectronic sensor array installed in the purification chamber. The sensors capture the real-time position and attitude information of the food ingredients, and calculate the new position relationship between the contaminated areas on the surface of the food ingredients and the vibration elements. According to the new position relationship, recalculate the optimal propagation direction and phase configuration of the vibration waves, and adjust the timing and phase parameters of the drive signal to ensure that the vibration energy always acts precisely on the target contaminated area, forming a targeted vibration cleaning waveform. The targeted vibration cleaning waveform refers to a vibration waveform that can be adaptively adjusted according to the contamination distribution characteristics and position changes of food ingredients, and has the characteristics of directivity, focusing, and time-variation.

[0074] For example: When a batch of carrots passes through a high-frequency vibration purification device, three main contaminated areas on the surface of the carrots are identified by pre-scanning, which are located at the top, middle, and root respectively, and a corresponding vibration purification intensity parameter table is generated. Through spatial mapping, the corresponding relationship between 12 vibration elements in the purification chamber and these three contaminated areas is determined. According to the spatial position, vibration elements No. 1-4 are divided into control group A, responsible for the top area; vibration elements No. 5-8 are divided into control group B, responsible for the middle area; vibration elements No. 9-12 are divided into control group C, responsible for the root area. Assign a vibration frequency of 35 kHz and an intensity coefficient of 0.8 to control group A, assign a vibration frequency of 28 kHz and an intensity coefficient of 0.7 to control group B, and assign a vibration frequency of 42 kHz and an intensity coefficient of 0.9 to control group C, generating three different control instruction sequences. These instruction sequences are converted into digital pulse signals by a waveform generator, and after power amplification, they are sent to the vibration elements of the three control groups respectively. Since groups B and C are relatively close in space, there is a risk of interference between their vibration waves. Therefore, a 45-degree phase shift is applied to the vibration signal of group C to eliminate harmful interference. When the carrots slowly rotate during the purification process, the optoelectronic sensors detect that the contaminated area at the root has turned to a position away from the vibration elements of group C. The control system quickly reallocates the vibration element combination, adjusts vibration element No. 8 from group B to group C, and adjusts its vibration frequency and phase to ensure that the vibration energy continuously acts on the contaminated area at the root, achieving efficient cleaning in all directions.

[0075] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0076] (1) Sample the water body in the purification chamber at fixed time intervals through a water turbidity detector with multi-point distribution, collect the original data points of the water turbidity during the purification process, and perform digital filtering processing on the original data points of the water turbidity to filter out random fluctuations and outliers, forming a smoothed turbidity data sequence;

[0077] (2) Divide the smoothed turbidity data sequence into multiple consecutive time windows along the time axis, calculate the average turbidity value within each time window, and obtain a set of turbidity values for each time period.

[0078] (3) Based on the set of turbidity values for each time period, calculate the turbidity difference between adjacent time windows to obtain the turbidity change data for each time period.

[0079] (4) Divide the turbidity change data by the corresponding time window length to calculate the turbidity change rate per unit time, and form the initial detachment rate data points.

[0080] (5) Associate and mark the initial detachment rate data points with the current vibration parameters to establish a correspondence table between turbidity changes and vibration parameters.

[0081] (6) According to the data distribution characteristics in the correspondence table, fit the functional relationship between the turbidity change rate and the purification time to generate a turbidity time-varying model. Based on the turbidity time-varying model, combined with the parameters of the purified water volume and the surface area of the food ingredients, convert to obtain the pollutant detachment rate curve of the pollutant detachment mass per unit area changing with time.

[0082] Specifically, install a water quality turbidity detector with multi-point distribution on the purification chamber wall. This is an optoelectronic sensing device for measuring the concentration of suspended particles in water. Usually, it adopts the principle of scattered light and judges the turbidity value by detecting the scattering intensity of the incident light by the water body. The water quality turbidity detectors are evenly distributed in a ring in the purification chamber to ensure that the water body state can be monitored in all directions. After the purification process is started, the turbidity detectors continuously sample the water body in the purification chamber at a fixed time interval (usually 0.1 - 1 second), record the time series of the water body turbidity values, and form a set of original data points of the water body turbidity. Due to the inevitable influence of random factors such as water flow disturbance and bubble interference in the actual measurement process, the original turbidity data often contains noise and outliers, and digital filtering processing is required. The digital filtering adopts a method combining median filtering and moving average filtering. First, use the median filter to process the outliers, take the median of a total of 5 points before and after each data point to replace the original value, and effectively remove the interference of outlier points; then apply 5-point moving average filtering to calculate the weighted average of the two data points before and after each moment to smooth the random fluctuations. The smoothed turbidity data sequence obtained after such processing can more accurately reflect the actual change trend of the water body turbidity.

[0083] The smoothed turbidity data sequence is divided into multiple consecutive time windows along the time axis. A time window is a time period of a fixed length, usually set to 5 - 10 seconds, which can be adjusted according to the type of food ingredient and the expected pollutant detachment rate. The time window division adopts the sliding window method, and adjacent windows have a 50% overlapping area. This overlapping design can capture the trend of turbidity changes more meticulously. Calculate the arithmetic mean of the turbidity data within each time window to obtain the representative turbidity value for that period. The turbidity values of all periods form a set of period turbidity values. Compared with the original data sequence, the set of period turbidity values further reduces the data volume while retaining the key change characteristics. Based on the set of period turbidity values, calculate the turbidity difference between adjacent time windows. The specific operation is to subtract the average turbidity value of the previous time window from the average turbidity value of the current time window to obtain the turbidity increment. This difference calculation is performed for all adjacent time windows to form a data sequence of turbidity change amounts. The turbidity change amount data directly reflects the situation of pollutants detaching into the water per unit time. A positive value indicates an increase in turbidity (continuous detachment of pollutants), and a negative value indicates a decrease in turbidity (sedimentation or filtration of pollutants).

[0084] To obtain a more practically meaningful pollutant detachment rate, it is necessary to divide the turbidity change amount data by the corresponding time window length. This calculation converts the turbidity change amount into the turbidity change rate per unit time, with the unit of NTU / second (nephelometric turbidity unit / second). These rate data form an initial detachment rate data point sequence, which directly reflects the speed of pollutants detaching from the food ingredient surface into the water at different moments.

[0085] Associating and marking the initial detachment rate data points with the current vibration parameters is a key step in understanding the vibration effect. The vibration parameters include information such as the frequency value, intensity coefficient, and phase configuration of the vibration element at the current moment. Through timestamp matching, each initial detachment rate data point is bound to the vibration parameters at the corresponding moment to establish a correspondence table between turbidity changes and vibration parameters. This correspondence table is a multi-dimensional data structure that contains information in multiple dimensions such as time, detachment rate, vibration frequency, and vibration intensity, providing a data basis for subsequent analysis of the relationship between vibration parameters and purification effects.

[0086] According to the data distribution characteristics in the correspondence table, through mathematical model fitting techniques, establish a functional relationship between the turbidity change rate and the purification time. The fitting process uses polynomial regression or exponential decay models, and selects the appropriate model function according to the actual data distribution characteristics. Taking the exponential decay model as an example, the relationship between the turbidity change rate and time can be expressed as:

[0087] R T (t) = C0·e -λt +K

[0088] where, R T(t) represents the turbidity change rate at time t, C0 is the initial turbidity change rate, λ is the attenuation coefficient, which is related to the pollutant properties and vibration parameters, and K is the residual rate constant. The model parameter values are determined by the least squares method to minimize the sum of the squared errors between the fitted curve and the actual data points, obtaining the time-varying turbidity model.

[0089] The time-varying turbidity model is a mathematical expression that describes the relationship between turbidity change and time. However, to obtain the true pollutant detachment rate curve, it is necessary to convert it in combination with the purified water volume and the food material surface area parameters. The conversion formula is:

[0090]

[0091] Where, R C (t) is the pollutant detachment mass rate per unit area (mg / cm 2 ·s), R T (t) is the turbidity change rate (NTU / s), V w is the volume of water in the purification chamber (L), α is the turbidity-mass conversion coefficient (mg / L·NTU), A f is the surface area of the food material (cm 2 ). The turbidity-mass conversion coefficient α is a parameter related to the pollutant type and is determined through pre-calibration experiments. For example, for common sediment pollutants, α is approximately 0.8 - 1.2 mg / L·NTU. Through this conversion, the curve of the pollutant detachment mass per unit area changing with time is finally obtained, that is, the pollutant detachment rate curve.

[0092] For example, when a batch of potatoes is purified by high-frequency vibration, the eight turbidity detectors installed around the purification chamber collect data every 0.5 seconds. Within 30 seconds after the start of purification, the raw turbidity data shows a rapid rise from the initial 2NTU to 15NTU, but the data contains more fluctuations and several obvious outliers. After digital filtering, the smoothed turbidity curve shows an obvious rise-platform-slow decline trend. The data is divided into a time window of 5 seconds, and the average turbidity values ​​of 12 time periods are calculated, which are 2.1, 6.3, 10.8, 13.5, 14.7, 15.0, 14.9, 14.6, 13.9, 13.2, 12.5, and 11.8NTU respectively. The turbidity difference between adjacent time periods is calculated, and 11 turbidity changes are obtained: 4.2, 4.5, 2.7, 1.2, 0.3, -0.1, -0.3, -0.7, -0.7, -0.7, -0.7NTU. Divided by the time window length of 5 seconds, the turbidity change rate is obtained: 0.84, 0.90, 0.54, 0.24, 0.06, -0.02, -0.06, -0.14, -0.14, -0.14, -0.14 NTU / s. These data points are associated with the vibration parameters at the corresponding time (such as frequency 32kHz, intensity coefficient 0.75) to establish a corresponding relationship table. Through fitting analysis, it is determined that the relationship between the turbidity change rate and time conforms to the exponential decay model, and C0 = 1.2, λ = 0.095, K = -0.15 is calculated. Combined with the water volume of the purification chamber of 30L, the surface area of ​​the potato is about 1200cm 2 , turbidity-mass conversion factor 1.0mg / L·NTU, and the pollutant removal rate curve was calculated. The curve shows that at the beginning of purification, about 0.025mg of pollutants were removed per square centimeter of potato surface per second, and after 30 seconds, it dropped to 0.005mg / cm 2 ·s, basically stabilized at 0.001mg / cm after 60 seconds 2 ·s or less, indicating that most of the pollutants have been separated in the first 60 seconds.

[0093] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0094] (1) Calculate the slope of the pollutant separation rate curve, determine the inflection point of the curve, and divide the purification process into the initial acceleration stage, the stable separation stage, and the slowdown tail stage;

[0095] (2) Establish a pollutant separation feature library for each stage based on the characteristics of the initial acceleration stage, stable separation stage, and mitigation tail stage;

[0096] (3) Match the real-time data points of the pollutant detachment rate curve with the pollutant detachment feature library, determine the type of the current purification stage, extract the corresponding frequency adjustment coefficient from the vibration frequency optimization rule set based on the stage type, and calculate the current optimal vibration frequency value;

[0097] (4) According to the optimal vibration frequency value, combined with the change trend of the pollutant detachment rate, determine the start timing and the basic intensity parameter of the water flow flushing;

[0098] (5) Multiply the basic intensity parameter by the ratio of the current pollutant detachment rate to the historical peak value, calculate the real-time dynamic adjustment coefficient of the water flow flushing intensity, combine the optimal vibration frequency value and the real-time dynamic adjustment coefficient of the water flow flushing intensity into a parameter pair, and generate a vibration-flushing parameter matching table;

[0099] (6) Based on the vibration-flushing parameter matching table, establish the signal triggering relationship between the vibration control unit and the water flow control unit to form a vibration-flushing linkage mechanism.

[0100] Specifically, calculate the slope of the pollutant detachment rate curve. Calculate the first derivative of the curve at each time point through the numerical differentiation method, that is, the change rate of the detachment rate. The slope calculation uses the central difference method. For the time point t, its slope value is equal to (RC(t + Δt) - RC(t - Δt)) / (2×Δt), where RC represents the pollutant detachment rate and Δt is the sampling time interval. By analyzing the change trend of the slope value, determine the important change inflection points in the curve. An inflection point refers to the position where the slope of the curve changes significantly, usually identified by the zero point or local extreme value of the second derivative. During the purification process, the pollutant detachment rate curve usually shows a trend of rising first and then falling, containing two typical inflection points: the first inflection point corresponds to the turning point where the rate rises to the maximum value, and the second inflection point corresponds to the turning point where the rate changes from a rapid decline to a slow decline. Based on these two inflection points, divide the entire purification process into three consecutive stages: the initial acceleration stage, the stable detachment stage, and the slowdown end stage. The initial acceleration stage refers to the time period from the start of purification to the first inflection point. In this stage, the pollutant detachment rate rises rapidly, indicating that the vibration energy is effectively loosening and separating the surface pollutants. The stable detachment stage refers to the time period from the first inflection point to the second inflection point. In this stage, the pollutant detachment rate remains at a high level and begins to decline, indicating that a large amount of loosened pollutants are continuously detaching. The slowdown end stage refers to the time period from the second inflection point to the end of purification. In this stage, the pollutant detachment rate slowly declines and tends to be stable, indicating that the residual pollutants are gradually detached.

[0101] For the different characteristics of these three stages, a pollutant detachment characteristic library for each stage is established. The characteristic library is a data set that records the typical detachment characteristic parameters of different food ingredients and different pollution types in each stage, including information such as the stage duration range, detachment rate range, and rate change pattern. The characteristic library is established through the statistics of a large amount of experimental data and provides a reference benchmark for the purification process under different conditions. For example, for pesticide residues on leafy vegetables, the initial acceleration stage usually lasts for 5 - 15 seconds, and the detachment rate increases approximately exponentially; the stable detachment stage lasts for 20 - 40 seconds, and the detachment rate first maintains the peak value and then gradually decreases; the slowdown ending stage lasts for 30 - 60 seconds, and the detachment rate decays approximately exponentially. During the real-time purification control process, the real-time data points of the pollutant detachment rate curve are matched and analyzed with the pollutant detachment characteristic library to determine the stage type of the current purification process. The matching method uses pattern recognition technology to calculate the similarity between the real-time detachment curve and the typical curves of each stage in the characteristic library, and selects the type with the highest similarity as the judgment result of the current stage. Based on the determined stage type, the corresponding frequency adjustment coefficient is extracted from the vibration frequency optimization rule set. The vibration frequency optimization rule set is a set of preset rules that specify the adjustment strategy of the vibration frequency for different purification stages. For example, in the initial acceleration stage, using a lower frequency (25 - 30 kHz) is beneficial to loosening pollutants; in the stable detachment stage, a medium frequency (30 - 40 kHz) is beneficial to continuously separating pollutants; in the slowdown ending stage, a higher frequency (40 - 50 kHz) is beneficial to removing residual pollutants. Through look-up table or interpolation calculation, the frequency adjustment coefficient corresponding to the current stage is obtained, and this coefficient is multiplied by the reference vibration frequency to calculate the current optimal vibration frequency value.

[0102] According to the optimal vibration frequency value, combined with the changing trend of the pollutant detachment rate, determine the starting time and basic intensity parameters of the water flow flushing. Water flow flushing is an important means to assist vibration purification. By directing the water flow to scour the surface of the food ingredient, it accelerates the removal of the loosened pollutants. The starting time of the water flow flushing is triggered according to the threshold value of the pollutant detachment rate, usually starting when the detachment rate reaches 50 - 70% of the peak value. At this time, a large amount of pollutants have been loosened by vibration but have not been completely detached. The basic intensity parameters of the water flow flushing refer to the combined values of parameters such as water pressure, flow rate, and spraying angle set under standard conditions, usually expressed as a normalized value between 0 - 100.

[0103] Perform real-time dynamic adjustment on the basic intensity parameters to adapt to the purification requirements of different stages. The calculation formula for the real-time dynamic adjustment coefficient of the water flow flushing intensity is:

[0104]

[0105] where, γ wf is the dynamic adjustment coefficient of the water flow flushing intensity, with a value range of 0 - 1; βbase is the base adjustment coefficient, related to the type of food ingredient, usually taking values from 0.5 to 0.8; R C R(t) is the pollutant detachment rate at the current moment; R C,max is the maximum detachment rate observed historically; α stage is the stage characteristic coefficient, taking 1.2 in the initial acceleration stage, 1.0 in the stable detachment stage, and 0.8 in the slowdown end stage; is the change rate of the current detachment rate; T stage is the characteristic time scale of the current stage. By comprehensively considering the ratio of the current detachment rate to the peak value, the change trend of the detachment rate, and the characteristics of the stage, the precise dynamic adjustment of the water flow intensity is achieved.

[0106] The calculated optimal vibration frequency value and the dynamic adjustment coefficient of the water flow flushing intensity are combined into a parameter pair. Each moment corresponds to a set of parameters, and they are arranged in chronological order to form a vibration - flushing parameter matching table. This matching table is a two - dimensional data structure, with the abscissa being time and the ordinate containing two parameters: the vibration frequency value and the flushing intensity coefficient. The parameters in the matching table will be dynamically updated as the purification process progresses and the stage changes, realizing the precise control of the purification process.

[0107] Based on the vibration - flushing parameter matching table, a signal triggering relationship is established between the vibration control unit and the water flow control unit to form a vibration - flushing linkage mechanism. The vibration control unit is responsible for driving the vibration element to work at a specified frequency, and the water flow control unit is responsible for adjusting the water pump output and nozzle angle to control the water flow flushing intensity. The two control units work in coordination through the central controller, and according to the timing arrangement of the parameter matching table, the synergy between vibration and flushing is realized. The linkage mechanism includes functions of state monitoring and feedback adjustment. When it is detected that the actual detachment effect does not match the expectation, the values in the parameter matching table can be automatically adjusted to ensure the purification effect.

[0108] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0109] (1) Use the high - resolution image acquisition device at the outlet to perform a full - range scan on the surface of the purified food ingredient to obtain the cleaning state image of the surface of the purified food ingredient;

[0110] (2) Perform illumination correction and geometric transformation on the cleaning state image to make it have the same coordinate system and scale as the food ingredient pollution area distribution map, forming a standardized post - processing image;

[0111] (3) Perform pixel - level corresponding superposition of the standardized post - processing image and the food ingredient pollution area distribution map, calculate the pollutant removal ratio of each area, and generate a cleanliness distribution matrix;

[0112] (4) Conduct statistical analysis on the cleanliness distribution matrix, calculate the average cleanliness and cleanliness variance of each area, and determine the areas with uneven cleanliness and insufficient cleanliness;

[0113] (5) Based on the position information of the areas with uneven cleanliness and insufficient cleanliness, trace back their corresponding vibration parameters and flushing parameters, and establish an association database between the cleaning effect and process parameters;

[0114] (6) Based on the association database, formulate a parameter optimization strategy for the areas with cleanliness below the threshold, and generate corrected values for vibration frequency and water flow intensity;

[0115] (7) Apply the corrected values of vibration frequency and water flow intensity to the control parameters of the vibration-flushing linkage mechanism, and update the vibration-flushing parameter matching table;

[0116] (8) Transmit the updated vibration-flushing parameter matching table to the control center through the parameter feedback channel, adjust the processing parameters of the next batch of ingredients, and form a closed-loop control system for the purification process.

[0117] Specifically, after the ingredients are subjected to vibration-flushing linkage purification treatment, a high-resolution image acquisition device installed at the outlet conducts a full-range scan of the surface of the purified ingredients. The high-resolution image acquisition device refers to an industrial camera array with a resolution of not less than 20 million pixels, equipped with a ring-shaped uniform light source and a multi-angle adjustable bracket, capable of capturing the minute details on the surface of the ingredients. During the image acquisition process, the ingredients are placed on a rotating platform and rotated at a fixed angular velocity, while multiple cameras take pictures from different angles to ensure that all areas of the surface of the ingredients are captured, and finally, an image of the cleanliness state of the surface of the purified ingredients is obtained. The cleanliness state image is a set of high-definition digital images containing the complete information of the surface of the ingredients, directly reflecting the actual state of the purified ingredients. After obtaining the cleanliness state image, it is necessary to perform illumination correction and geometric transformation processing to make it consistent with the distribution map of the contaminated areas of the ingredients collected at the inlet in the spatial reference system. Illumination correction refers to eliminating the brightness differences generated under different illumination conditions through image processing algorithms, mainly using histogram equalization and white balance adjustment techniques. Geometric transformation is to convert the coordinate system of the cleanliness state image into the same standard coordinate system as the contaminated area distribution map through a coordinate mapping function. Geometric transformation includes four basic operations: translation, rotation, scaling, and perspective transformation. By using a feature point matching algorithm, the corresponding feature points in the two sets of images (such as the corners of the ingredient contour, obvious marks on the surface, etc.) are automatically identified, the optimal transformation matrix is calculated, and then this transformation matrix is applied to the entire image to generate a standardized post-processed image that is precisely corresponding to the contaminated area distribution map in terms of spatial position.

[0118] Perform pixel-level corresponding superposition analysis on the standardized post-treatment image and the food ingredient contamination area distribution map, and calculate the pollutant removal ratio for each area. The superposition analysis first precisely aligns the two images according to pixel positions, and then calculates the color and brightness difference values for each corresponding pixel point. For each pixel point marked as a contaminated area in the contamination area distribution map, compare the characteristics of its corresponding pixel point in the standardized post-treatment image. If the difference between the two exceeds a preset threshold, it is determined that the pollutant at this point has been removed; if the difference is less than the threshold, it is determined that there is pollutant residue. According to this determination rule, count the ratio of the number of pixel points with removed pollutants to the total number of contaminated pixel points in each predefined area to obtain the pollutant removal ratio for this area. The removal ratio values for all areas form a cleanliness distribution matrix. The cleanliness distribution matrix is a two-dimensional array, and each element corresponds to the cleanliness level of a specific area on the food ingredient surface, with a numerical range of 0 - 100%.

[0119] Conduct statistical analysis on the cleanliness distribution matrix and calculate various key indicators. First, calculate the global average cleanliness, which is the arithmetic mean of the cleanliness levels of all areas and reflects the overall purification effect; then calculate the cleanliness variance, which reflects the purification uniformity of each area; finally, identify areas with insufficient cleanliness according to a preset cleanliness threshold (usually 85%) and identify areas with uneven cleanliness according to a variance threshold (usually 10%). Areas with insufficient cleanliness refer to areas where the cleanliness level is lower than the preset threshold, indicating that the purification effect in this area does not meet the requirements; areas with uneven cleanliness refer to areas where the cleanliness difference from adjacent areas is too large, indicating that there is an imbalance in the purification process. The identification results of these two types of areas are represented by area identifiers and coordinate information and serve as the key targets for subsequent parameter optimization.

[0120] According to the position information of the identified areas with uneven cleanliness and areas with insufficient cleanliness, trace back and query the vibration parameters and flushing parameters used in the purification process for this area. Trace back and query means extracting parameter data such as frequency values and intensity coefficients applied to this area from the vibration-flushing parameter matching table through the area coordinates. Correlate the cleanliness data of the area with the corresponding process parameter data, analyze the influence rules of different parameter combinations on the purification effect, and establish an association database of cleaning effect and process parameters. The association database is a multi-dimensional data structure that contains three types of data, namely area characteristics (position, pollution type, pollution degree), process parameters (vibration frequency, vibration intensity, flushing start timing, flushing intensity), and purification effect (cleanliness, uniformity), as well as their mutual relationships, providing data support for parameter optimization.

[0121] Based on the data patterns accumulated in the associated database, a parameter optimization strategy is formulated for areas with cleanliness below the threshold. The parameter optimization uses the gradient descent method to analyze the sensitivity relationship between cleanliness and each parameter, and determine the direction and amplitude of parameter adjustment. For areas with insufficient cleaning, the vibration frequency correction value and water flow intensity correction value are calculated according to their characteristics. The vibration frequency correction value is usually adjusted within the range of ±5 kHz, and whether it is higher or lower depends on the pollutant type and cleanliness gap in this area; the water flow intensity correction value is usually adjusted within the range of ±0.2, mainly considering the area location and cleaning uniformity. The calculation of the correction value comprehensively considers multiple factors such as area cleanliness, area location, and pollution type to ensure the pertinence and effectiveness of the adjustment.

[0122] Apply the calculated vibration frequency correction value and water flow intensity correction value to the control parameters of the vibration-flushing linkage mechanism, and update the vibration-flushing parameter matching table. The update process not only adjusts the parameters for the identified problem areas, but also appropriately corrects the parameters of adjacent areas to ensure a smooth transition of the parameter space and avoid causing new non-uniform phenomena. The updated parameter matching table contains optimized parameter combinations for specific food ingredients and specific pollution conditions, providing a more accurate control basis for the purification treatment of subsequent batches. Transmit the updated vibration-flushing parameter matching table to the control center through the parameter feedback channel to complete the last link of the closed-loop control. The parameter feedback channel is a data transmission mechanism that ensures that the optimized parameters can be applied to the treatment of the next batch of food ingredients in a timely and accurate manner. The control center receives the updated parameter matching table, selects the applicable parameter combination according to the characteristics of the newly entered food ingredients, and guides the operation of the vibration element and the water flow system, thus forming a closed-loop control system for the purification process. The closed-loop control system continuously improves the purification effect and efficiency through real-time monitoring, effect evaluation, parameter optimization, and feedback adjustment.

[0123] The intelligent control method of the high-frequency vibration food ingredient purification equipment in the embodiments of the present application has been described above. Next, the intelligent control system of the high-frequency vibration food ingredient purification equipment in the embodiments of the present application will be described. Please refer to Figure 2 , an embodiment of the intelligent control system of the high-frequency vibration food ingredient purification equipment in the embodiments of the present application includes:

[0124] A scanning module, configured to scan the surface of the food ingredient for pollution degree through a scanning device arranged at the entrance of the purification equipment, perform digital processing on the obtained food ingredient surface image, and obtain a food ingredient pollution area distribution map;

[0125] A calculation module, configured to quantitatively calculate the area and density of each pollution area according to the food ingredient pollution area distribution map, and obtain a vibration purification intensity parameter table;

[0126] A control module, configured to drive and control a vibration element in a purification chamber through a pulse signal generator based on the vibration purification intensity parameter table to form a targeted vibration cleaning waveform;

[0127] An acquisition module, configured to use a water quality turbidity detector installed on the purification chamber wall to collect the turbidity value of the water body in real time during the vibration process, perform a timing analysis on the continuously collected turbidity data, and obtain a pollutant detachment rate curve;

[0128] An adjustment module, configured to proportionally adjust the vibration frequency and the water flow flushing intensity according to the pollutant detachment rate curve, and establish a vibration-flushing linkage mechanism;

[0129] A scanning module, configured to re-scan the surface of the processed food material through an imaging comparison system at the outlet, compare and analyze the obtained image with the food material pollution area distribution map, and automatically optimize the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference to form a closed-loop control system for the purification process.

[0130] Through the collaborative cooperation of the above-mentioned various components, by setting up a scanning device at the entrance of the purification equipment, the accurate scanning and digital processing of the surface contamination degree of the food materials are realized, and a distribution map of the contaminated areas of the food materials is generated, providing an accurate data basis for subsequent targeted purification; by quantitatively calculating the area and density of each contaminated area, a vibration purification intensity parameter table is obtained, realizing the refined configuration of the purification parameters and avoiding the blindness and experience dependence of the traditional equipment parameter setting; based on the vibration purification intensity parameter table, the vibration elements in the purification chamber are driven and controlled by a pulse signal generator to form a targeted vibration cleaning waveform, significantly improving the utilization efficiency of the purification energy and reducing unnecessary damage to the food materials; the water turbidity detector installed on the wall of the purification chamber is used to collect the water turbidity value in real time during the vibration process, and the pollutant detachment rate curve is obtained through time series analysis, establishing a real-time monitoring mechanism for the purification process and solving the technical defect that the traditional equipment cannot sense the purification progress; according to the pollutant detachment rate curve, the vibration frequency and the water flow flushing intensity are adjusted proportionally to establish a vibration-flushing linkage mechanism, realizing the synergistic effect of various purification technologies and greatly improving the removal ability of stubborn contaminants; the surface of the processed food materials is scanned again by the imaging comparison system at the outlet, compared and analyzed with the distribution map of the contaminated areas of the food materials, and the operating parameters of the vibration-flushing linkage mechanism are automatically optimized according to the cleanliness difference, forming a closed-loop control system for the purification process and solving the technical problems that the traditional equipment cannot conduct effect evaluation and continuous optimization. Especially when applying the lightweight convolutional neural network algorithm to process the surface image of the food materials, this algorithm fully considers the characteristics of complex surface texture and uneven illumination of the food materials, extracts the features of the contaminated areas through multi-layer convolution and pooling operations, and significantly improves the accuracy of pollution recognition; when processing the pollutant detachment rate data, the Kalman filter algorithm used effectively suppresses the noise interference in the water turbidity measurement through the prediction-correction mechanism and enhances the data reliability; in the vibration-flushing linkage control, the fuzzy logic decision tree algorithm applied converts multi-dimensional information such as the food material type, pollution characteristics, and purification stage into accurate control instructions, realizing intelligent decision-making under complex conditions; in the purification effect evaluation, the grey relational analysis method adopted adapts to the characteristics of small samples and incomplete information and accurately quantifies the correlation degree between the purification parameters and the cleaning effect. The application of these algorithms and models in specific functional links greatly improves the intelligent level and adaptability of the system, enabling the present invention to automatically adjust the optimal purification strategy for different types and different pollution degrees of food materials, significantly improving the purification quality and efficiency and reducing the energy consumption.

[0131] Referring to Figure 3 , in the embodiment of the present invention, a computer device is further provided. This computer device can be a server, and its internal structure can be as Figure 3As shown in the figure. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements the above method.

[0132] Those skilled in the art can understand that Figure 3 the structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0133] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0134] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to memory, storage, database, or other media provided by the present invention and used in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0135] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, systems, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0136] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0137] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. An intelligent control method for a high-frequency vibration food purification device, characterized in that, The intelligent control method for the high-frequency vibration food purification equipment includes: Scanning the surface of the food for contamination degree through a scanning device arranged at the entrance of the purification equipment, digitally processing the obtained food surface image, and obtaining a food contamination area distribution map; According to the food contamination area distribution map, quantitatively calculating the area and density of each contamination area to obtain a vibration purification intensity parameter table; Based on the vibration purification intensity parameter table, driving and controlling the vibration elements in the purification chamber through a pulse signal generator to form a targeted vibration cleaning waveform; Using a water turbidity detector installed on the purification chamber wall to collect the water turbidity value in real time during the vibration process, performing time series analysis on the continuously collected turbidity data, and obtaining a pollutant detachment rate curve; According to the pollutant detachment rate curve, proportionally adjusting the vibration frequency and the water flow flushing intensity to establish a vibration-flushing linkage mechanism; Scanning the surface of the processed food again through an imaging comparison system at the outlet, comparing and analyzing the obtained image with the food contamination area distribution map, and automatically optimizing the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference to form a closed-loop control system for the purification process.

2. The intelligent control method of the high-frequency vibration food purification equipment according to claim 1, characterized in that, The step of scanning the surface of the food for contamination degree through a scanning device arranged at the entrance of the purification equipment, digitally processing the obtained food surface image, and obtaining a food contamination area distribution map includes: Illuminating the surface of the food omnidirectionally through a multi-spectral lighting unit, collecting the surface reflection data of the food under different lighting conditions, and forming an original light intensity data set; According to the original light intensity data set, performing threshold segmentation on the surface brightness value to screen out the suspected contamination area data points higher than the background value; Using a high-resolution imaging device to take multi-angle pictures of the food surface, obtaining the surface color and texture information, and establishing a food surface reference feature library; Comparing the suspected contamination area data points with the food surface reference feature library for difference comparison, extracting the areas with large deviations from the reference features, and obtaining a candidate set of contamination areas; Performing density clustering on the candidate set of contamination areas, classifying the points with adjacent distances less than a preset threshold into the same contamination area, and generating the boundary coordinates of the contamination area; Based on the boundary coordinates of the contamination area, combining the color depth value in the contamination area, generating the food contamination area distribution map through a coordinate mapping method.

3. The intelligent control method of the high-frequency vibration food purification equipment according to claim 1, characterized in that, The step of quantitatively calculating the area and density of each contamination area according to the food contamination area distribution map to obtain a vibration purification intensity parameter table includes: Performing grid division on the food contamination area distribution map, dividing the food surface into several micro-region units with equal areas, and forming a contamination area grid coordinate system; According to the contamination area grid coordinate system, counting the number of contaminated pixel points in each micro-region unit to obtain the contamination density value of each micro-region, and normalizing the contamination density values of each micro-region to convert the contamination density into a standardized contamination intensity index of 0-100; Based on the standardized pollution intensity index, adjacent micro-regions are clustered and grouped. Micro-regions with a pollution intensity difference less than the threshold are merged into the same pollution block, and the total area of each pollution block is calculated by pixel accumulation. Combining the standardized pollution intensity index, the pollution load value of each pollution block is generated; According to the pollution load value, the best vibration frequency range is queried and matched from the vibration frequency library, and the vibration intensity coefficient is queried and matched from the vibration intensity library; The best vibration frequency range and the vibration intensity coefficient are combined and paired to form a vibration parameter combination for each pollution block; The coordinate position information of each pollution block is associated and mapped with the corresponding vibration parameter combination to generate the vibration purification intensity parameter table.

4. The intelligent control method of the high-frequency vibration food purification equipment according to claim 1, characterized in that Based on the vibration purification intensity parameter table, the vibration elements in the purification chamber are driven and controlled by a pulse signal generator to form a targeted vibration cleaning waveform, including: The vibration parameters of each pollution block in the vibration purification intensity parameter table are spatially mapped to generate a partition control mapping table for the vibration elements in the purification chamber. According to the partition control mapping table, multiple vibration elements in the purification chamber are divided into several independent control groups to form a vibration area control unit; The corresponding vibration frequency value and vibration intensity coefficient are assigned to the vibration area control unit to generate a sub-region vibration control instruction sequence; The sub-region vibration control instruction sequence is input into a pulse waveform generation module and converted into a digital pulse sequence of voltage-time relationship; The digital pulse sequence is subjected to power amplification processing to adjust the voltage amplitude and current intensity of the output signal to generate a vibration element drive signal; The vibration element drive signal is transmitted to the corresponding vibration area control unit through a multi-channel transmission system to form a multi-region independent control vibration field. The phase synchronization of the multi-region independent control vibration field is adjusted to eliminate the interference phenomenon of adjacent region vibration waves and establish a vibration wave array for collaborative work; Based on the collaborative work vibration wave array, according to the position change of the food ingredients in the purification chamber, the propagation direction of the vibration waves in each region is dynamically adjusted to form the targeted vibration cleaning waveform.

5. The intelligent control method of the high-frequency vibration food purification equipment according to claim 1, characterized in that, The water turbidity detector installed on the purification chamber wall is used to real-time collect the water turbidity value during the vibration process, and the time series analysis is performed on the continuously collected turbidity data to obtain the pollutant detachment rate curve, including: The water in the purification chamber is sampled at a fixed time interval by a multi-point distributed water turbidity detector to collect the original water turbidity data points during the purification process. The original water turbidity data points are subjected to digital filtering processing to filter out random fluctuations and outliers interference to form a smoothed turbidity data sequence; The smoothed turbidity data sequence is divided into multiple continuous time windows along the time axis, and the average turbidity value in each time window is calculated to obtain a set of time period turbidity values; Based on the set of time period turbidity values, the turbidity difference between adjacent time windows is calculated to obtain the turbidity change amount data for each time period; The turbidity change amount data is divided by the corresponding time window length to calculate the turbidity change rate per unit time to form the initial detachment rate data points; Associate and mark the initial detachment rate data points with the current vibration parameters to establish a correspondence table between turbidity change and vibration parameters; According to the data distribution characteristics in the correspondence table, fit the functional relationship between the turbidity change rate and the purification time to generate a turbidity time-varying model. Based on the turbidity time-varying model, combined with the purified water volume and the food material surface area parameters, convert to obtain the pollutant detachment rate curve of the pollutant detachment mass per unit area changing with time.

6. The intelligent control method for the high-frequency vibration food material purification equipment according to claim 1, characterized in that Based on the pollutant detachment rate curve, perform proportional adjustment on the vibration frequency and the water flow flushing intensity to establish a vibration-flushing linkage mechanism, including: Calculate the slope of the pollutant detachment rate curve to determine the change inflection point in the curve, and divide the purification process into an initial acceleration stage, a stable detachment stage, and a slowdown ending stage; According to the characteristics of the initial acceleration stage, the stable detachment stage, and the slowdown ending stage, establish a pollutant detachment feature library for each stage; Match the real-time data points of the pollutant detachment rate curve with the pollutant detachment feature library, judge the stage type of the current purification process, and based on the stage type, extract the corresponding frequency adjustment coefficient from the vibration frequency optimization rule set to calculate the current optimal vibration frequency value; According to the optimal vibration frequency value, combined with the change trend of the pollutant detachment rate, determine the starting time and the basic intensity parameter of the water flow flushing; Multiply the basic intensity parameter by the ratio of the current pollutant detachment rate to the historical peak value to calculate the real-time water flow flushing intensity dynamic adjustment coefficient, and combine the optimal vibration frequency value and the water flow flushing intensity dynamic adjustment coefficient into a parameter pair to generate a vibration-flushing parameter matching table; Based on the vibration-flushing parameter matching table, establish a signal triggering relationship between the vibration control unit and the water flow control unit to form the vibration-flushing linkage mechanism.

7. The intelligent control method of the high-frequency vibration food purification equipment according to claim 1, characterized in that Through the imaging comparison system at the outlet, re-scan the surface of the processed food material, compare and analyze the obtained image with the food material pollution area distribution map, and automatically optimize the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference to form a closed-loop control system for the purification process, including: Use the high-resolution image acquisition device at the outlet to perform an omnidirectional scan of the surface of the purified food material to obtain the cleanliness state image of the surface of the purified food material; Perform illumination correction and geometric transformation on the cleanliness state image to make it have the same coordinate system and scale as the food material pollution area distribution map to form a standardized post-processed image; Perform pixel-level corresponding superposition of the standardized post-processed image and the food material pollution area distribution map, calculate the pollutant removal ratio of each area, and generate a cleanliness distribution matrix; Perform statistical analysis on the cleanliness distribution matrix, calculate the average cleanliness and cleanliness variance of each area, and determine the areas with uneven cleanliness and areas with insufficient cleanliness; According to the position information of the areas with uneven cleanliness and areas with insufficient cleanliness, trace back their corresponding vibration parameters and flushing parameters to establish an association database between the cleaning effect and the process parameters; Based on the associated database, formulate a parameter optimization strategy for areas with cleanliness lower than the threshold, and generate a vibration frequency correction value and a water flow intensity correction value; Apply the vibration frequency correction value and the water flow intensity correction value to the control parameters of the vibration-flushing linkage mechanism, and update the vibration-flushing parameter matching table; Transmit the updated vibration-flushing parameter matching table to the control center through the parameter feedback channel, adjust the processing parameters of the next batch of ingredients, and form a closed-loop control system for the purification process.

8. An intelligent control system for a high-frequency vibration food purification device, which is used to implement the intelligent control method for the high-frequency vibration food purification device as described in any one of claims 1-7, and is characterized in that, The intelligent control system of the high-frequency vibration food purification equipment includes: A scanning module, which is used to scan the surface pollution degree of the ingredients through a scanning device arranged at the entrance of the purification equipment, digitally process the obtained ingredient surface image, and obtain an ingredient pollution area distribution map; A calculation module, which is used to quantitatively calculate the area and density of each pollution area according to the ingredient pollution area distribution map, and obtain a vibration purification intensity parameter table; A control module, which is used to drive and control the vibration elements in the purification chamber through a pulse signal generator based on the vibration purification intensity parameter table to form a targeted vibration cleaning waveform; An acquisition module, which is used to use a water quality turbidity detector installed on the purification chamber wall to collect the water turbidity value in real time during the vibration process, perform time series analysis on the continuously collected turbidity data, and obtain a pollutant detachment rate curve; An adjustment module, which is used to proportionally adjust the vibration frequency and the water flow flushing intensity according to the pollutant detachment rate curve, and establish a vibration-flushing linkage mechanism; A scanning module, which is used to scan the surface of the processed ingredients again through an imaging comparison system at the outlet, compare and analyze the obtained image with the ingredient pollution area distribution map, and automatically optimize the operating parameters of the vibration-flushing linkage mechanism according to the cleanliness difference, and form a closed-loop control system for the purification process.

9. A computer device, characterized in that, It includes a memory and a processor, and the memory stores a computer program that can run on the processor. The characteristic is that when the processor executes the computer program, it implements the intelligent control method of the high-frequency vibration food purification equipment according to any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the processor is caused to execute the intelligent control method of the high-frequency vibration food purification equipment according to any one of claims 1 to 7.

Images