A dielectric spectrum analysis system based on artificial intelligence

Abstract

The invention discloses a dielectric spectrum analysis system based on artificial intelligence, relates to the technical field of dielectric spectrum detection, and comprises a food dielectric spectrum detection module, a food dielectric spectrum analysis module, a food dielectric heating detection module, an execution terminal and a data warehouse. The invention determines a reference dielectric constant and a reference dielectric loss factor of a food at each processing stage based on environmental detection data of the food, and then evaluates the overall quality appearance value of the food, thereby improving the accuracy of dielectric spectrum detection and analysis of the food. The invention determines the heating condition of the food according to the overall quality appearance value of the food, ensures the rationality of dielectric heating of the food, improves the dielectric heating effect of the food, reduces food safety risks and energy waste rate, performs real-time detection on the dielectric heating process of the food, and corrects the dielectric heating power of the food, thereby ensuring the rationality of the dielectric heating process of the food, thereby ensuring the effect of dielectric heating of the food.

Description

Technical Field

[0001] The present invention relates to the technical field of dielectric spectrum detection, and in particular to a dielectric spectrum analysis system based on artificial intelligence. Background Art

[0002] In the modern food processing process, there is a problem of inaccurate quality control. The application of dielectric spectrum detection technology is becoming more and more extensive. Dielectric spectrum detection is a method of non-destructive detection using the dielectric properties of a substance in an electric field. It plays a vital role in ensuring food quality, improving processing efficiency, and saving energy and reducing emissions. Therefore, it is extremely necessary to conduct dielectric spectrum detection on food.

[0003] The existing technology lacks attention to the impact of the environment on the dielectric spectrum test results. The environment affects the deviation of dielectric spectrum test to a certain extent. The neglect of this aspect in the existing technology reduces the accuracy of dielectric spectrum test and analysis of food, affects the quality of food, and makes it difficult to provide reliable data support for the determination of subsequent food heating conditions. At the same time, the analysis of the heating conditions of food determined according to the dielectric spectrum test results of food and the quality of the finished product is not in-depth enough, which makes it difficult to ensure the rationality of dielectric heating of food, reduces the dielectric heating effect of food, and increases food safety risks and energy waste rate. Summary of the invention

[0004] The purpose of the present invention is to provide a dielectric spectrum analysis system based on artificial intelligence, which solves the problems existing in the background technology.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a dielectric spectrum analysis system based on artificial intelligence.

[0006] The food dielectric spectrum detection module is used to perform dielectric spectrum detection and environmental detection during the food processing process to obtain the food's dielectric spectrum detection data and environmental detection data, and after the food processing is completed, perform image detection on the food to obtain a three-dimensional image of the food.

[0007] The food dielectric spectrum analysis module is used to evaluate the overall quality appearance value of the food based on the dielectric spectrum detection data, environmental detection data and three-dimensional images of the food, and adjust the dielectric heating conditions of the food.

[0008] The food dielectric heating detection module is used to monitor the dielectric heating process of the food when the food is dielectrically heated, obtain the dielectric heating process data of the food, and correct the dielectric heating power of the food.

[0009] The execution terminal is used to control the dielectric heating device according to the adjustment of the dielectric heating conditions of the food and the correction of the dielectric heating power.

[0010] The beneficial effects of the present invention are: (1) The present invention detects the food processing process in the food dielectric spectrum detection module and detects the finished food processing product, providing data support for subsequent food dielectric spectrum analysis.

[0011] (2) In the food dielectric spectrum analysis module of the present invention, the reference dielectric constant and reference dielectric loss factor of the food at each processing stage are first determined based on the environmental detection data of the food, and then the overall quality appearance value of the food is comprehensively evaluated in combination with the finished product detection of the food. This overcomes the shortcomings of the prior art, improves the accuracy of the dielectric spectrum detection and analysis of the food, ensures the quality of the food, and provides reliable data support for the determination of the subsequent heating conditions of the food.

[0012] (3) The present invention determines the heating conditions of food according to the overall quality appearance value of food in the food dielectric spectrum analysis module, thereby ensuring the rationality of dielectric heating of food, improving the dielectric heating effect of food, and reducing food safety risks and energy waste rate.

[0013] (4) The present invention performs real-time detection of the dielectric heating process of food in the food dielectric heating detection module and corrects the dielectric heating power of food to ensure the rationality of the food dielectric heating process and avoid local overheating or insufficient temperature, thereby ensuring the effect of food dielectric heating. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Reference Figure 1 As shown, the present invention provides a dielectric spectrum analysis system based on artificial intelligence, including: a food dielectric spectrum detection module, a food dielectric spectrum analysis module, a food dielectric heating detection module, and an execution terminal.

[0018] It should be noted that the present invention also includes a data warehouse for storing dielectric compliance ratio factors at each processing stage, historical processing data and test data of food, safety feature parameter ranges of each environmental data at each processing stage, environmental feature thresholds, standard three-dimensional images of food, allowable defect volumes of each defect type, reduction parameters corresponding to unit increase in overall mass appearance value of each dielectric heating condition, increase parameters corresponding to unit decrease in overall mass appearance value, overall mass convergence value, layout density corresponding to unit thickness of each heating electrode shape, reasonable convergence range of food dielectric heating, and influence weight factors corresponding to the first mass appearance value and the second mass appearance value of each food.

[0019] It should also be noted that the food dielectric spectrum detection module is connected to the food dielectric spectrum analysis module, the food dielectric spectrum analysis module is connected to the food dielectric heating detection module, the food dielectric spectrum analysis module and the food dielectric heating detection module are both connected to the execution terminal, and the data warehouse is connected to the food dielectric spectrum analysis module.

[0020] The food dielectric spectrum detection module is used to perform dielectric spectrum detection and environmental detection during the food processing process to obtain dielectric spectrum detection data and environmental detection data of the food, and to perform image detection on the food after the food processing is completed to obtain a three-dimensional image of the food.

[0021] As a preferred solution, the dielectric spectrum detection data includes the dielectric constant and dielectric loss factor of each processing stage.

[0022] The environmental detection data include characteristic parameters of each environmental data in each processing stage.

[0023] It should be noted that the characteristic parameters of the environmental data include average temperature, average humidity, magnetic field strength, vibration frequency and other environmental parameters that affect the accuracy of dielectric spectrum detection.

[0024] It should also be noted that the various processing stages are specifically the various processing stages of food, for example, the processing stages of vegetables include washing, peeling, seeding, cutting, etc., the processing stages of fruits include picking, sorting, and washing, etc., and the processing stages of meat include slaughtering, cutting, and pickling, etc.

[0025] It should be noted that the food includes but is not limited to vegetables, fruits, meat, etc.

[0026] It should also be noted that the dielectric spectrum detection of the food processing process is carried out by using a dielectric spectrum detector, and the environmental detection of the food processing process is carried out by using temperature sensors, humidity sensors, magnetic field strength testers and speed sensors.

[0027] The present invention detects the food processing process in the food dielectric spectrum detection module and detects the finished food processing product, providing data support for subsequent food dielectric spectrum analysis.

[0028] The food dielectric spectrum analysis module is used to evaluate the overall quality appearance value of the food based on the dielectric spectrum detection data, environmental detection data and three-dimensional images of the food, and adjust the dielectric heating conditions of the food.

[0029] As a preferred solution, the overall quality appearance value of the food is evaluated by a specific evaluation method: based on the dielectric spectrum detection data and environmental detection data of the food, the first quality appearance value of the food is evaluated. .

[0030] Evaluate the second quality appearance value of food based on the three-dimensional image of food .

[0031] Import the first quality manifestation value and the second quality manifestation value of food into the overall quality manifestation value evaluation model Among them , They are respectively represented as the influencing weight factors corresponding to the first quality appearance value and the second quality appearance value of the food matched in the data warehouse, and the overall quality appearance value of the food is output. .

[0032] It should be noted that the influencing weight factors corresponding to the first quality manifestation value and the second quality manifestation value are specifically set by food experts for the first quality manifestation value and the second quality manifestation value of each food and stored in the data warehouse, so as to match the influencing weight factors corresponding to the first quality manifestation value and the second quality manifestation value of the food.

[0033] As a preferred solution, the first quality manifestation value of the food is evaluated by a specific method of extracting characteristic parameters of each environmental data at each processing stage from the environmental data detection data of the food, and then constructing a reference dielectric constant and a reference dielectric loss factor at each processing stage of the food.

[0034] The dielectric constant of each processing stage of food , dielectric loss factor , reference dielectric constant and reference dielectric loss factor Import into the first quality appearance value evaluation model Among them The dielectric compliance ratio factor of the i-th processing stage preset in the data warehouse is used to output the first quality appearance value of the food. is the number of each processing stage, , is any integer greater than 2, is the number of processing stages.

[0035] It should be noted that the dielectric compliance ratio factor of each processing stage specifically refers to the demand for dielectric detection quality in each processing stage. If the demand for dielectric detection quality in a certain processing stage is higher, the dielectric compliance ratio factor will be larger, indicating that the importance of dielectric detection quality in this processing stage is higher.

[0036] As a preferred scheme, the reference dielectric constant and reference dielectric loss factor of each processing stage of the food are constructed by the following specific construction method: historical processing data of the food is obtained from a data warehouse, wherein the historical processing data includes the dielectric constant, dielectric loss factor and characteristic parameters of each environmental data of each historical processing at each processing stage, and test data of the food is obtained from the data warehouse, wherein the test data includes the test dielectric constant and test dielectric loss factor of each processing stage.

[0037] It should be noted that the test data of the food specifically refers to the data obtained by testing under safe environmental conditions.

[0038] Characteristic parameters of each environmental data at each processing stage based on each historical processing of food , evaluate the environmental characteristics of each historical processing of food at each processing stage , is the safety feature parameter interval of the bth environmental data of the i-th processing stage stored in the data warehouse, is the number of each historical processing. , is any integer greater than 2, is the number of each environmental data, , is any integer greater than 2, is the total amount of environmental data, and e is a natural constant.

[0039] The environmental characteristic values ​​of each historical processing of food at each processing stage are subtracted from the environmental characteristic thresholds preset in the data warehouse to obtain the environmental characteristic deviation values ​​of each historical processing of food at each processing stage.

[0040] The dielectric constant and dielectric loss factor of each historical processing of the food at each processing stage are subtracted from the test dielectric constant and test dielectric loss factor of each processing stage in the test data of the food, respectively, to obtain the dielectric constant deviation value and dielectric loss factor deviation value of each historical processing of the food at each processing stage.

[0041] The environmental characteristic deviation values ​​of each historical processing of food at each processing stage are classified according to the preset classification standards to obtain the environmental characteristic deviation intervals of the food at each processing stage, and the dielectric constant deviation value and dielectric loss factor deviation value of each historical processing of the food within each environmental characteristic deviation interval at each processing stage are obtained, and the average is processed to obtain the dielectric constant deviation value and dielectric loss factor deviation value of each environmental characteristic deviation interval of the food at each processing stage.

[0042] It should be noted that the classification standard specifically refers to the numerical increment step. For example, if the numerical increment step is 0.1, then the environmental characteristic deviation intervals are 0-0.1, 0.1-0.2 and 0.2-0.3.

[0043] Based on the characteristic parameters of each environmental data at each processing stage of food, the environmental characteristic values ​​of each processing stage of food are evaluated in the same way, combined with the dielectric constant deviation value and dielectric loss factor deviation value of each environmental characteristic deviation interval of food at each processing stage. If the environmental characteristic value of food at a certain processing stage is within a certain environmental characteristic deviation interval, the dielectric constant deviation value and dielectric loss factor deviation value of the environmental characteristic deviation interval are used as the dielectric constant deviation value and dielectric loss factor deviation value of the food at this processing stage. The dielectric constant deviation value and dielectric loss factor deviation value of each processing stage of food are screened and obtained, and they are added to the tested dielectric constant deviation value and dielectric loss factor deviation value of the food at each processing stage, respectively, to obtain the reference dielectric constant and reference dielectric loss factor of the food at each processing stage.

[0044] As a preferred solution, the second quality appearance value of the food is evaluated by a specific evaluation method: based on the three-dimensional image of the food, the defect parameters of the food are identified, wherein the defect parameters include each defect type and its corresponding defect volume. , and identify the color uniformity of food , is the number of each defect type, , is any integer greater than 2.

[0045] It should be noted that identifying food defect parameters through image recognition technology is an existing technology and will not be elaborated here. Importing the three-dimensional image of the food into the image analysis software and outputting the color uniformity of the food is also an existing technology and will not be elaborated here.

[0046] It should also be noted that the defect types include dents, protrusions and scratches.

[0047] Compare the 3D image of the food with the standard 3D image of the food preset in the data warehouse to evaluate the shape compliance index of the food. .

[0048] It should be noted that the specific evaluation method of the shape compliance index of the food is: obtain the volume of the three-dimensional image of the food, obtain the overlapping volume of the three-dimensional image of the food and the standard three-dimensional image of the food preset in the data bin, and divide it by the volume in the three-dimensional image of the food to obtain the shape compliance index of the food.

[0049] Import the food defect parameters, color uniformity and shape conformity index into the second quality appearance value evaluation model and output the second quality appearance value of the food ,in is the allowable defect volume of the fth defect type preset in the data bin, is the number of defect types.

[0050] In the food dielectric spectrum analysis module of the present invention, the reference dielectric constant and reference dielectric loss factor of the food at each processing stage are first determined based on the environmental detection data of the food, and then the overall quality appearance value of the food is comprehensively evaluated in combination with the finished product detection of the food. This overcomes the shortcomings of the prior art, improves the accuracy of the dielectric spectrum detection and analysis of the food, ensures the quality of the food, and provides reliable data support for the determination of the subsequent heating conditions of the food.

[0051] As a preferred solution, the dielectric heating conditions of the food are adjusted, and the specific adjustment method is: based on the three-dimensional image of the food, determine the shape of the heating electrode of the food, and determine the appropriate distribution map of the heating electrode of the food.

[0052] The shape of the heating electrode of the food is determined by a specific method: based on the three-dimensional image of the food, the three-dimensional contour and volume of the food are obtained, and the three-dimensional contour of the food is compared with the three-dimensional contour set corresponding to each heating electrode shape stored in the data warehouse to obtain the overlapping volume of the food with each element in the three-dimensional contour set corresponding to each heating electrode shape, and the overlapping volume is divided by the volume of the food to obtain the overlap degree of the food with each element in the three-dimensional contour set corresponding to each heating electrode shape. If the overlap degree between the food and an element in the three-dimensional contour set corresponding to a certain heating electrode shape is the largest, the heating electrode shape is used as the shape of the heating electrode of the food, thereby determining the shape of the heating electrode of the food.

[0053] Obtain the food's physical property data set and heating target from the data warehouse, combined with the food's overall quality display value , and define the adjustment values ​​of each dielectric heating condition of the food ,in is the initial value of the mth dielectric heating condition of the food, , They are respectively represented as the reduction parameter corresponding to the unit increase in the overall mass appearance value and the increase parameter corresponding to the unit decrease in the overall mass appearance value of the mth dielectric heating condition preset in the data bin, is the overall quality convergence value preset in the data bin, is the number of each dielectric heating condition, , is any integer greater than 2, and the initial value of each dielectric heating condition of the food is determined by the physical property data of the food and the heating target.

[0054] It should be noted that the dielectric heating conditions include power, duration and frequency, etc.

[0055] It should be noted that the data set of the physical properties of the food includes density, specific heat capacity and electrical conductivity, etc. After the food processing stage is completed and before dielectric heating, it is usually necessary to test the physical properties of the food. The detection method belongs to the existing technology and will not be described here. It will be uploaded to the data warehouse.

[0056] It should also be noted that the initial values ​​of the dielectric heating conditions of the food are specifically determined by matching the characteristic values ​​of the dielectric heating conditions of the heating targets of the food in each physical property data set with the characteristic values ​​of the dielectric heating conditions of the heating targets of the food in each physical property data set preset in the data warehouse; if the heating target of the food is the same as a certain heating target, and the physical property data set of the food is the same as a certain physical property data set of the heating target, then the characteristic values ​​of the dielectric heating conditions of the heating target in the physical property data set are used as the initial values ​​of the dielectric heating conditions of the food.

[0057] The dielectric heating conditions of the food are adjusted based on the shape of the heating electrode of the food, the appropriate portion map, and the appropriate adjustment value of each dielectric heating condition.

[0058] As a preferred solution, the suitable distribution diagram of the heating electrodes for the food is determined by the following specific method: a three-dimensional image of the placement of the food is obtained through a camera terminal built into the heating container of the food, and the internal space of the heating container of the food is gridded to obtain several internal sub-areas of the heating container of the food, and then the structural thickness of the food in the several internal sub-areas of the heating container is obtained.

[0059] The layout density corresponding to the unit thickness of each heating electrode shape is obtained from the data warehouse, and the layout density corresponding to the unit thickness of the heating electrode shape of the food is screened in combination with the shape of the heating electrode of the food.

[0060] It should be noted that the layout density corresponding to the unit thickness of each heating electrode shape is specifically set by a dielectric heating expert and stored in the data warehouse.

[0061] The heating electrode layout density of the food in each internal sub-region of the heating container is obtained by multiplying the structural thickness of the food in each internal sub-region of the heating container by the heating layout density corresponding to the unit thickness.

[0062] The internal sub-areas of the food in the heating container are grid-divided to obtain a number of layout sub-areas, and the center points of the several layout sub-areas are used as the layout points of the heating electrodes. Based on the layout density of the heating electrodes of the food in the internal sub-areas of the heating container, a layout diagram of the heating electrodes of the food in the internal sub-areas of the heating container is obtained, and the appropriate distribution diagram of the heating electrodes of the food is obtained by summarizing.

[0063] The present invention determines the heating conditions of food according to the overall quality appearance value of food in the food dielectric spectrum analysis module, ensures the rationality of dielectric heating of food, improves the dielectric heating effect of food, and reduces food safety risks and energy waste rate.

[0064] The food dielectric heating detection module is used to monitor the dielectric heating process of the food when the food is dielectrically heated, obtain the dielectric heating process data of the food, and correct the dielectric heating power of the food.

[0065] As a preferred solution, the dielectric heating power of the food is corrected, and the specific correction method is: based on the dielectric heating process data of the food, where the dielectric heating process data includes the temperature distribution diagram of each internal sub-area in the current monitoring cycle, the dielectric heating characteristic value of the food in each internal sub-area in the current monitoring cycle is evaluated.

[0066] The food dielectric heating characteristic value of each internal sub-region in the current monitoring period is compared with the food dielectric heating reasonable convergence interval preset in the data warehouse. If the food dielectric heating characteristic value of a certain internal sub-region in the current monitoring period is not within the food dielectric heating reasonable convergence interval, and the food dielectric heating characteristic value of the internal sub-region is greater than the upper limit of the food dielectric heating reasonable convergence interval, then the food dielectric heating reasonable convergence value is subtracted from the food dielectric heating characteristic value of the internal sub-region in the current monitoring period to obtain the food dielectric heating reasonable deviation value of the internal sub-region in the current monitoring period, and the deviation value is compared with the power reduction value of the unit food dielectric heating reasonable deviation value corresponding to each heating electrode layout density stored in the data warehouse, and the power reduction value of the unit food dielectric heating reasonable deviation value of the internal sub-region in the current monitoring period is screened and multiplied by the food dielectric heating reasonable deviation value to obtain the power reduction value of the internal sub-region. Otherwise, the power increase value of the internal sub-region is obtained by the same analysis, and the power adjustment value of the internal sub-region is obtained by summarizing. By analogy, the power adjustment value of each internal sub-region is obtained.

[0067] The dielectric heating power of the food is corrected based on the power adjustment value of each internal sub-area in the current monitoring cycle.

[0068] As a preferred solution, the dielectric heating characteristic value of the food in each internal sub-region of the current monitoring cycle is evaluated by a specific evaluation method: based on the temperature distribution diagram of each internal sub-region of the food in the current monitoring cycle, the average temperature value of each internal sub-region of the food in the current monitoring cycle is obtained, and the average temperature value is imported into the food dielectric heating characteristic value evaluation model. The food dielectric heating characteristic values ​​of each internal sub-area in the current monitoring period are output, where is the reference temperature of the current monitoring cycle, is the number of each internal sub-area, , is any integer greater than 2.

[0069] It should be noted that the reference temperature of each internal sub-region in the current monitoring period is specifically obtained by averaging the average temperature values ​​of each internal sub-region in the current monitoring period to obtain the reference temperature of the current monitoring period.

[0070] The present invention performs real-time detection of the dielectric heating process of food in a food dielectric heating detection module, and corrects the dielectric heating power of food, thereby ensuring the rationality of the dielectric heating process of food, avoiding local overheating or insufficient temperature, and thus ensuring the effect of dielectric heating of food.

[0071] The execution terminal is used to control the dielectric heating device according to the adjustment of the dielectric heating conditions of the food and the correction of the dielectric heating power.

[0072] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A dielectric spectrum analysis system based on artificial intelligence, characterized in that: include: The food dielectric spectrum detection module is used to perform dielectric spectrum detection and environmental detection during the food processing process to obtain the food's dielectric spectrum detection data and environmental detection data, and after the food processing is completed, perform image detection on the food to obtain a three-dimensional image of the food; The food dielectric spectrum analysis module is used to evaluate the overall quality appearance value of the food based on the dielectric spectrum detection data, environmental detection data and three-dimensional images of the food, and adjust the dielectric heating conditions of the food; The dielectric heating conditions of the food are adjusted, and the specific adjustment method is as follows: Based on the three-dimensional image of the food, determine the shape of the heating electrodes of the food, and determine a suitable distribution map of the heating electrodes of the food; Obtain the physical property data set and heating target of the food, combine it with the overall quality manifestation value δ of the food, and define the adjustment value of each dielectric heating condition of the food Where R m is the initial value of the mth dielectric heating condition of the food, R _0m , R _1m They are respectively represented as the reduction parameter corresponding to the unit increase of the overall mass appearance value of the mth dielectric heating condition preset in the data bin, and the increase parameter corresponding to the unit decrease of the overall mass appearance value, δ′ is the overall mass convergence value preset in the data bin, m is the number of each dielectric heating condition, m=1,2,...,l, l is any integer greater than 2, and the initial value of each dielectric heating condition of the food is determined by the physical property data of the food and the heating target; adjusting the dielectric heating conditions of the food based on the shape of the heating electrode of the food, the appropriate portion map, and the appropriate adjustment values ​​of each dielectric heating condition; The specific method for determining the appropriate distribution diagram of the heating electrodes of the food is as follows: Acquire a three-dimensional image of the food placement, and divide the internal space of the food heating container into grids to obtain a number of internal sub-regions of the food heating container, and then obtain the structural thickness of the food in the several internal sub-regions of the heating container; Obtaining the layout density corresponding to the unit thickness of each heating electrode shape from the data warehouse, and screening the layout density corresponding to the unit thickness of the heating electrode shape of the food in combination with the shape of the heating electrode of the food; Multiplying the structural thickness of each internal sub-region of the food in the heating container by the heating layout density corresponding to the unit thickness to obtain the heating electrode layout density of the food in each internal sub-region of the heating container; Divide each internal sub-region of the food in the heating container into a grid to obtain a number of layout sub-regions, and use the center points of the several layout sub-regions as the layout points of the heating electrodes. Based on the layout density of the heating electrodes of the food in each internal sub-region of the heating container, obtain a layout diagram of the heating electrodes of the food in each internal sub-region of the heating container, and summarize to obtain a suitable distribution diagram of the heating electrodes of the food; The food dielectric heating detection module is used to monitor the dielectric heating process of the food when the food is dielectrically heated, obtain the dielectric heating process data of the food, and correct the dielectric heating power of the food; The execution terminal is used to control the dielectric heating device according to the adjustment of the dielectric heating conditions of the food and the correction of the dielectric heating power.

2. The dielectric spectrum analysis system based on artificial intelligence according to claim 1, characterized in that: The dielectric spectrum detection data includes the dielectric constant and dielectric loss factor at each processing stage; The environmental detection data include characteristic parameters of each environmental data in each processing stage.

3. The dielectric spectrum analysis system based on artificial intelligence according to claim 2, characterized in that: The specific evaluation method for evaluating the overall quality appearance value of the food is as follows: Based on the dielectric spectrum detection data and environmental detection data of food, the first quality appearance value α of food is evaluated _0 ; Based on the three-dimensional image of the food, the second quality appearance value α of the food is evaluated _1 ; The first quality manifestation value and the second quality manifestation value of the food are introduced into the overall quality manifestation value evaluation model δ=ln(1+α _0 *λ _0 +α _1 *λ _1 ), where λ _0 , _1 They are respectively represented as the influence weight factors corresponding to the first quality appearance value and the second quality appearance value of the food matched in the data warehouse, and the overall quality appearance value δ of the food is output.

4. The dielectric spectrum analysis system based on artificial intelligence according to claim 3, characterized in that: The specific evaluation method of evaluating the first quality manifestation value of the food is as follows: Extracting characteristic parameters of each environmental data at each processing stage from the environmental data detection data of the food, and then constructing a reference dielectric constant and a reference dielectric loss factor at each processing stage of the food; The dielectric constant ε of each processing stage of food i , dielectric loss factor η i , reference dielectric constant ε i ′ and reference dielectric loss factor η i 'Imported into the first quality appearance value evaluation model In which χ i The dielectric compliance ratio factor of the i-th processing stage preset in the data warehouse is used to output the first quality appearance value of the food, i is the number of each processing stage, i=1,2,...,n, n is any integer greater than 2, and n is the number of processing stages.

5. The dielectric spectrum analysis system based on artificial intelligence according to claim 4, characterized in that: The specific construction method of constructing the reference dielectric constant and reference dielectric loss factor of each processing stage of the food is as follows: Acquire historical processing data of food from the data warehouse, wherein the historical processing data includes dielectric constant, dielectric loss factor and characteristic parameters of each environmental data at each processing stage of each historical processing, and acquire test data of food from the data warehouse, wherein the test data includes test dielectric constant and test dielectric loss factor at each processing stage; Characteristic parameters B based on the environmental data of each historical processing of food at each processing stage pib , evaluate the environmental characteristics of each historical processing of food at each processing stage is the safety characteristic parameter interval of the bth environmental data in the i-th processing stage stored in the data warehouse, p is the number of each historical processing, p=1,2,...,q, q is any integer greater than 2, b is the number of each environmental data, b=1,2,...,d, d is any integer greater than 2, d is the total number of environmental data, and e is a natural constant; Subtract the environmental characteristic value of each historical processing of the food at each processing stage from the environmental characteristic threshold preset in the data warehouse to obtain the environmental characteristic deviation value of each historical processing of the food at each processing stage; Subtract the dielectric constant and dielectric loss factor of each historical processing of the food at each processing stage from the test dielectric constant and test dielectric loss factor of each processing stage in the test data of the food, respectively, to obtain the dielectric constant deviation value and dielectric loss factor deviation value of each historical processing of the food at each processing stage; Classify the environmental characteristic deviation values ​​of each historical processing of the food at each processing stage according to the preset classification standard to obtain each environmental characteristic deviation interval of the food at each processing stage, and obtain the dielectric constant deviation value and dielectric loss factor deviation value of each historical processing of the food within each environmental characteristic deviation interval at each processing stage, and perform mean processing on them to obtain the dielectric constant deviation value and dielectric loss factor deviation value of each environmental characteristic deviation interval of the food at each processing stage; Based on the characteristic parameters of each environmental data at each processing stage of food, the environmental characteristic values ​​of each processing stage of food are evaluated in the same way. Combined with the dielectric constant deviation values ​​and dielectric loss factor deviation values ​​of each environmental characteristic deviation interval of food at each processing stage, the dielectric constant deviation values ​​and dielectric loss factor deviation values ​​of food at each processing stage are screened out, and added with the tested dielectric constant deviation values ​​and dielectric loss factor deviation values ​​of food at each processing stage, respectively, to obtain the reference dielectric constant and reference dielectric loss factor of food at each processing stage.

6. The dielectric spectrum analysis system based on artificial intelligence according to claim 3, characterized in that: The specific evaluation method of evaluating the second quality manifestation value of the food is as follows: Based on the 3D image of the food, the defect parameters of the food are identified, where the defect parameters include each defect type and its corresponding defect volume V f , and identify the color uniformity J of the food, f is the number of each defect type, f = 1, 2, ..., t, t is any integer greater than 2; Compare the three-dimensional image of the food with the standard three-dimensional image of the food preset in the data warehouse, and evaluate the shape compliance index Z of the food; Import the food defect parameters, color uniformity and shape conformity index into the second quality appearance value evaluation model and output the second quality appearance value of the food Where V f ′ is the allowable defect volume of the fth defect type preset in the data bin, and t is the number of defect types.

7. The dielectric spectrum analysis system based on artificial intelligence according to claim 1, characterized in that: The dielectric heating power of the food is corrected, and the specific correction method is as follows: Based on the dielectric heating process data of the food, wherein the dielectric heating process data includes a temperature distribution diagram of each internal sub-region in the current monitoring cycle, evaluating the dielectric heating characteristic value of the food in each internal sub-region in the current monitoring cycle; Compare the food dielectric heating characteristic value of each internal sub-region in the current monitoring period with the food dielectric heating reasonable convergence interval preset in the data warehouse. If the food dielectric heating characteristic value of a certain internal sub-region in the current monitoring period is not within the food dielectric heating reasonable convergence interval, and the food dielectric heating characteristic value of the internal sub-region is greater than the upper limit of the food dielectric heating reasonable convergence interval, analyze and obtain the power down-regulation value of the internal sub-region. Otherwise, analyze and obtain the power up-regulation value of the internal sub-region in the same way, and summarize and obtain the power adjustment value of the internal sub-region. By analogy, obtain the power adjustment value of each internal sub-region. The dielectric heating power of the food is corrected based on the power adjustment value of each internal sub-area in the current monitoring cycle.

8. The dielectric spectrum analysis system based on artificial intelligence according to claim 7, characterized in that: The specific evaluation method for evaluating the food dielectric heating characteristic value of each internal sub-area in the current monitoring cycle is as follows: Based on the temperature distribution diagram of each internal sub-region of the current monitoring period of the food, the average temperature value of each internal sub-region of the current monitoring period of the food is obtained, and it is imported into the food dielectric heating characteristic value evaluation model The food dielectric heating characteristic value of each internal sub-area of ​​the current monitoring cycle is output, where T′ is the reference temperature of the current monitoring cycle, x is the number of each internal sub-area, x=1, 2, ..., y, and y is any integer greater than 2.

Images