A method and system for detecting juice purity

Abstract

The present invention provides a juice purity detection method and system, which relates to the technical field of liquid detection. The method includes: constructing a modulated juice with different juice concentrations; obtaining magnetic induction tomography reconstruction data of various modulated juices; establishing a data set with magnetic induction tomography reconstruction data as a feature and the juice concentration of the modulated juice corresponding to the magnetic induction tomography reconstruction data as a label; constructing a concentration regression prediction model; obtaining magnetic induction tomography reconstruction data of the juice to be detected; inputting the magnetic induction tomography reconstruction data of the juice to be detected into the concentration regression prediction model to obtain the concentration of the juice to be detected; performing conductivity visualization reconstruction on the magnetic induction tomography reconstruction data of the juice to be detected to obtain foreign matter information in the juice to be detected; outputting the concentration of the juice to be detected and the foreign matter information, and completing the purity detection of the juice to be detected, i.e., foreign matter detection of the juice to be detected and the concentration detection of the juice to be detected. The detection efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid detection, and in particular to a method and system for detecting the purity of juice. Background Art

[0002] With the development of food processing technology, more and more fruits are processed into juice and juice drinks. This method can not only enrich the flavor of fruits, but also prolong the storage period for easy market circulation. Juice refers to unfermented juice made from fruits through various physical methods such as squeezing, centrifugation, extraction, etc., or adding an equal amount of water lost during concentration to the prepared concentrated juice to restore it to a 100% juice product with basically the same color, soluble solids content and flavor as the original fruit pulp; juice drinks refer to juice products obtained by adding water, sugar solution and acidulant to juice or concentrated juice, and the juice content is generally not less than 10%.

[0003] At present, the concentration of juice contained in juice products has always been the focus of attention of manufacturers and consumers. Especially with the intensification of market competition, ensuring the quality and safety of products has become the main issue in their production and sales. In order to better ensure quality and safety and improve product competitiveness, continuous concentration monitoring and foreign matter detection in the juice production process have become a highly anticipated technology. This not only puts forward higher requirements for quality and safety at the social level, but also promotes the continued healthy and sound development of the entire juice products industry.

[0004] However, traditional juice purity testing often combines spectral analysis, refractive index measurement, density measurement and chemical titration with juice concentration detection. These methods are inefficient and prone to missed detection, which leads to the inability to conduct continuous detection and low practicality. Summary of the invention

[0005] In order to solve the technical problems that traditional juice purity detection methods commonly used in the prior art, such as spectral analysis, refractive index measurement, density measurement and chemical titration, are combined with juice concentration detection, and these methods all have the problems of low efficiency and easy missed detection, which in turn lead to the inability to continuously detect and low practicality, the present invention provides a juice purity detection method and system.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect

[0008] A juice purity detection method provided by an embodiment of the present invention is applied to a juice purity detection system, wherein the juice purity detection system comprises a host computer, a digital circuit module, a multiplexing module, a signal detection module and a coil sensor array connected in sequence, and a signal excitation module is respectively connected to the digital circuit and the multiplexing module; the method comprises:

[0009] S1: Constructing blended juices with different juice concentrations;

[0010] S2: obtaining magnetic induction tomography reconstruction data of various prepared juices through the juice purity detection system;

[0011] S3: establishing a data set with the magnetic induction tomography reconstruction data as a feature and the juice concentration of the modulated juice corresponding to the magnetic induction tomography reconstruction data as a label;

[0012] S4: constructing a concentration regression prediction model based on the data set;

[0013] S5: obtaining magnetic induction tomography reconstruction data of the juice to be tested;

[0014] S6: inputting the magnetic induction tomography reconstruction data of the juice to be tested into the concentration regression prediction model to obtain the concentration of the juice to be tested;

[0015] S7: Combining the sensitivity matrix and the conjugate gradient algorithm, the magnetic induction tomography reconstruction data of the juice to be detected is reconstructed by conductivity visualization to obtain the foreign matter information in the juice to be detected, wherein the foreign matter information includes the size and position of the foreign matter;

[0016] S8: Outputting the concentration of the juice to be detected and the foreign matter information, completing the purity detection of the juice to be detected, namely the foreign matter detection of the juice to be detected and the concentration detection of the juice to be detected.

[0017] Second aspect

[0018] An embodiment of the present invention provides a juice purity detection system, comprising:

[0019] processor;

[0020] A memory, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, the juice purity detection method as described in the first aspect is implemented.

[0021] The third aspect

[0022] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the juice purity detection method as described in the first aspect is implemented.

[0023] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0024] In the present invention, since the fruit itself contains a large number of charged particles, it is an electrolyte that can form a bioelectric field. Therefore, the conductivity parameter is one of the important physical properties of juice products. Based on this physical property, a juice purity detection system with a coil sensor array is constructed. By detecting the electromagnetic characteristics inside the field of the object to be tested, non-invasive detection is achieved. The structure is simple and efficient. The magnetic induction tomography reconstruction data of the juice to be tested can be quickly collected, and the tomography reconstruction data is used as a feature. The corresponding juice concentration is used as a label to create a data set. The data set is used to train the constructed concentration regression prediction model to output the juice concentration. In addition, This method can be used to avoid the problem that the liquid transportation process and foreign matter in the pipeline cannot be visualized due to the tightness and opacity of the juice transmission pipeline during the production process. This technology is used to collect magnetic induction tomography reconstruction data, and reconstruct the electrical characteristic parameter distribution of the pipeline fault based on the magnetic induction tomography reconstruction data to identify foreign matter in the juice. Rapid scanning can be used to obtain a large amount of data to improve detection efficiency. Real-time conductivity reconstruction provides a basis for continuous monitoring of juice concentration and foreign matter detection. It can continuously and completely and accurately detect the juice to be tested, avoid the problem of missed detection, and promote the improvement of juice product quality and production safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0026] Figure 1 A schematic diagram of a flow chart of a method for detecting juice purity provided by an embodiment of the present invention;

[0027] Figure 2 A schematic diagram of the structure of a juice purity detection system provided by an embodiment of the present invention;

[0028] Figure 3 A schematic diagram of a structure in which a coil sensor array is installed in a pipeline according to an embodiment of the present invention;

[0029] Figure 4 Another structural schematic diagram of a coil sensor array installed on a pipeline provided by an embodiment of the present invention;

[0030] Figure 5 A schematic structural diagram of a coil sensor array support provided by an embodiment of the present invention;

[0031] Figure 6 A schematic structural diagram of a coil sensor array skeleton provided by an embodiment of the present invention;

[0032] Figure 7 A schematic diagram of the structure of a concentration regression prediction model provided by an embodiment of the present invention;

[0033] Figure 8 A schematic diagram of the structure of another juice purity detection system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0035] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0036] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0037] Reference Manual Attached Figure 1 , showing a flow chart of a juice purity detection method provided by an embodiment of the present invention.

[0038] Reference Manual Attached Figure 2 , showing a schematic structural diagram of a juice purity detection system provided by an embodiment of the present invention.

[0039] An embodiment of the present invention provides a juice purity detection method, which is applied to a juice purity detection system, wherein the juice purity detection system includes a host computer, a digital circuit module, a multiplexing module, a signal detection module and a coil sensor array connected in sequence, and the signal excitation module is respectively connected to the digital circuit and the multiplexing module.

[0040] Among them, the multiplexing module is composed of a digitally controlled analog switch, which is combined with the binary address line output by the FPGA to select the output of the signal, so that the excitation signal is only applied to the current excitation coil, and the reference signal is only provided to the channel being demodulated. The host computer part is used for data processing, image reconstruction and concentration prediction, and processes the obtained magnetic induction tomography reconstruction data. By extracting magnetic induction tomography reconstruction data, the juice purity detection system can non-invasively obtain the distribution information of the internal conductivity of the juice, accurately detect the concentration and foreign matter of the juice, and avoid the limitations of traditional detection methods when facing sealing and opacity problems. At the same time, it realizes efficient and real-time detection, greatly improving the quality control and safety of the production process.

[0041] In a possible implementation, the digital circuit module includes a DAC circuit, an FPGA main control circuit, and an ADC circuit.

[0042] The signal excitation module includes a power amplifier circuit.

[0043] The signal excitation module is mainly composed of a power amplifier circuit, which performs power amplification on the small sinusoidal signal output by the DAC circuit to drive the coil to form a quasi-steady field in the detection area.

[0044] The coil sensor array consists of a plurality of hollow coils arranged around the juice transmission channel in a centrally equidistant manner.

[0045] Reference Manual Attached Figure 3 , shows a schematic structural diagram of a coil sensor array installed on a pipeline provided by an embodiment of the present invention.

[0046] Reference Manual Attached Figure 4 , shows another structural schematic diagram of a coil sensor array installed on a pipeline provided by an embodiment of the present invention.

[0047] Reference Manual Attached Figure 5 , showing a schematic structural diagram of a coil sensor array bracket provided in an embodiment of the present invention.

[0048] Reference Manual Attached Figure 6 , shows a schematic structural diagram of a coil sensor array skeleton provided in an embodiment of the present invention.

[0049] Figure 4 The schematic diagram of the structure of a coil sensor array installed on a pipeline is given along the direction of the pipeline.

[0050] exist Figure 5 middle, Figure 5 In the figure, a is a schematic structural diagram of one half of a coil sensor array bracket, and b is a side view, wherein the bracket is used to install the coil sensor array.

[0051] exist Figure 6 In the figure, view c is a "BB" section view. This figure shows the cross section of the component along the BB section. View d is an "AA" section view, which shows the structure after cutting from another angle, mainly showing the central cylindrical hole of the component and the groove structure on the periphery. This view better shows the internal space layout of the entire structure and helps understand how the central part cooperates with other parts. View e is a view from the bottom, showing the bottom shape of the component, including the two mounting holes of the component (these holes are consistent with the holes in view "c"). This view helps understand the lower layout of the entire component and, together with other views, fully presents the structural details.

[0052] Specifically, the number of hollow coils is 8, and the structural parameters of each hollow coil are consistent. The center interval angle is 45°. Any coil can be used for both excitation and detection in conjunction with a multiplexing module and a signal relay circuit. When one of the coils is used as an excitation coil, the other seven coils are used as detection coils to receive data. After all seven channels of data are collected, the excitation coil is replaced, and so on until each coil is used as an excitation coil once, completing a detection cycle to obtain the required magnetic induction tomography reconstruction data.

[0053] Specifically, the coil sensor arrays are made of resin material, and are sleeved on the measured pipe, and the axis of the coil sensor is perpendicular to the surface of the measured pipe. The coil sensor array is installed on the coil sensor array bracket. The coil sensor array bracket is two semi-octagonal structures. The semi-octagons are connected by plastic bolts. There are 8 slots evenly spaced and arranged on the inside, which are used to embed the coil sensor skeleton to fix the position and distribution of the coil sensor. The coil is wound on the coil skeleton. The copper wire with a fixed number of turns wound on the coil sensor skeleton constitutes a coil sensor. The backs of the 8 coil sensors are provided with slots corresponding to the bracket, which are embedded in the coil sensor bracket in sequence to form a coil sensor array.

[0054] The signal detection module includes a signal relay, a signal amplifier, a filter and an amplitude and phase detection circuit.

[0055] Among them, the various components in the signal detection module, namely the relay, signal amplifier, filter and amplitude and phase detection circuit, are connected to each coil sensor of the coil sensor through the SMA standard interface. The signal relay is used to switch the excitation / detection state of the coil. When in the detection state, the detection signal will be demodulated and processed before transmission.

[0056] The host computer is connected to the coil sensor array through the first circuit, the second circuit, the third circuit and the fourth circuit respectively.

[0057] The first circuit specifically includes an FPGA main control circuit, a DAC circuit, a power amplifier circuit, a multiplexing module and a signal detection module which are connected in sequence.

[0058] The second circuit specifically includes an FPGA main control circuit, a DAC circuit, a multiplexing module and a signal detection module which are connected in sequence.

[0059] The third line specifically includes an FPGA main control circuit, a multiplexing module and a signal detection module which are connected in sequence.

[0060] The fourth circuit specifically includes an FPGA main control circuit, an ADC circuit, a multiplexing module and a signal detection module which are connected in sequence.

[0061] Specifically, the digital circuit module is mainly composed of FPGA main control circuit, DAC circuit and ADC circuit, among which FPGA acts as the main control chip to coordinate the work of the entire system. The DAC circuit converts the digital signal output by the DDS module inside the FPGA into two sine waves with the same frequency, one as a reference signal and the other as an excitation signal transmitted to the signal excitation module, and the frequency and phase of the sine wave signal can be adjusted. The ADC circuit converts the analog signal of the signal detection module into a digital signal.

[0062] In actual use, the juice purity detection system of this structure realizes efficient and flexible magnetic induction tomography data acquisition through precisely controlled DAC and ADC circuits, adjustable sinusoidal excitation signals, and multiplexed coil sensor arrays. Its advantages are that it can accurately detect the conductivity distribution inside the juice, improve the accuracy and stability of data acquisition, and use the switching of multiple coils to ensure that the tested area can be fully covered during the detection process, thereby improving the detection accuracy and efficiency of the system.

[0063] Methods include:

[0064] S1: Constructing blended juices with different juice concentrations.

[0065] It is understandable that by preparing juice samples of different concentrations and simulating various juice concentrations in actual production, a variety of standard references can be provided for data collection and model training of subsequent detection systems to ensure that the detection system can accurately identify and predict juices of different concentrations.

[0066] In a possible implementation manner, S1 specifically includes:

[0067] The original juice is obtained and distilled water is added to the original juice to obtain a variety of prepared juices of different concentrations.

[0068] S2: Obtain magnetic induction tomography reconstruction data of various prepared juices through the juice purity detection system.

[0069] Among them, magnetic induction tomography reconstruction data is the distribution information of electromagnetic properties obtained through magnetic induction tomography technology. These data can reflect the internal structure and composition of an object. The magnetic induction tomography reconstruction data of juices of different concentrations are extracted through the juice purity detection system, which provides accurate conductivity distribution information for subsequent analysis, helps the model to accurately identify and distinguish juices of different concentrations, and ensures the reliability and accuracy of the test results. At the same time, this non-invasive data acquisition method improves detection efficiency and avoids physical damage to the juice.

[0070] In a possible implementation, S2 specifically includes:

[0071] S201: Inputting the prepared juices of different concentrations into the juice transmission channel.

[0072] S202: using any one of the hollow coils as an excitation coil and the remaining hollow coils as detection coils to collect amplitude and phase data of the modulated juice.

[0073] S203: Replace the hollow coil representing the excitation coil, and return to step S202 until all the hollow coils are used as primary excitation coils to obtain multiple groups of amplitude and phase data.

[0074] S204: Input the amplitude and phase data to the host computer, and output the magnetic induction tomography reconstruction data of the modulated juice:

[0075]

[0076] in, A Indicates including w The magnetic induction tomography reconstruction data matrix of juices modulated in different concentration categories, Indicates i Magnetic induction tomography reconstruction array of juices modulated by concentration categories, , Indicates p The first collection q The values ​​are the magnetic induction tomography reconstruction data, , , m Indicates the total number of collections. n Indicates the total number of collected values ​​for each concentration category of blended juice, U represents the magnetic induction tomography reconstruction data, V 1 indicates the voltage data when the detection area of ​​the juice transmission channel is the measured medium. V 0 represents the voltage data when the detection area of ​​the juice transmission channel is air. It means taking the imaginary part of a complex number. V represents a complex number representing voltage data, C represents the real part of a complex number,B represents the imaginary part of a complex number, j Represents the complex imaginary unit.

[0077] Specifically, first, the host computer communicates with the FPGA to send a digital signal to the DAC circuit to generate two sine waves. One is directly transmitted to the multiplexing module as a reference signal, and the other is first transmitted to the signal excitation module for power amplification and then transmitted to the multiplexing module. Then, the FPGA selects the coils in the coil sensor array as excitation coils in a certain order, and the multiplexing module outputs an excitation signal to the current excitation coil to drive it. After that, the remaining coils in the coil sensor array are used as detection coils in turn, and the multiplexing module will transmit the reference signal to the signal detection module corresponding to the current detection coil. The detection signal is amplified and filtered, and then demodulated with the reference signal, and finally transmitted to the multiplexing module for data collection by the ADC circuit. Repeat the above steps until all coils are used as excitation coils. For 8 coil sensors, each round of detection can collect a total of 8×7=56 data. Finally, 56 data are used as the final magnetic induction tomography reconstruction data.

[0078] It should be noted that by gradually switching each coil as an excitation and detection coil, the amplitude and phase data at multiple angles and positions are systematically collected to ensure comprehensive coverage of the conductivity changes in the juice transmission channel. This strategy of gradual and multiple acquisitions improves the richness and accuracy of the data and can more accurately reconstruct the conductivity distribution map of the juice. In addition, through the precise control of the FPGA and the efficient switching of the multiplexing module, the entire process achieves efficient data acquisition and processing, ensuring the high accuracy and consistency of the magnetic induction tomography reconstruction data, thereby improving the overall effect of juice purity detection.

[0079] S3: Establish a data set using magnetic induction tomography reconstruction data as features and using the juice concentration of the modulated juice corresponding to the magnetic induction tomography reconstruction data as a label.

[0080] It should be noted that by using magnetic induction tomography reconstruction data as features and juice concentration as labels to establish a data set, a clear input-output relationship is provided for model training. This method helps the model to accurately learn the relationship between juice concentration and its conductivity properties, improves the model's predictive ability and generalization performance, and ultimately enables the system to more accurately predict the concentration of unknown juice samples, thereby improving the reliability and practicality of detection.

[0081] In a possible implementation, the data set includes a training set and a test set divided according to a preset ratio.

[0082] It should be noted that those skilled in the art can set the size of the preset ratio according to actual needs, and the present invention is not limited here. Specifically, the preset ratio can be 8:2. Among them, the test set can be used to test the performance of the trained model to verify and provide a basis for improving the model training effect.

[0083] Reference Manual Attached Figure 7 , showing a schematic diagram of the structure of a concentration regression prediction model provided by an embodiment of the present invention.

[0084] S4: Construct a concentration regression prediction model based on the dataset.

[0085] It should be noted that by constructing a concentration regression prediction model based on the established data set, the system can automatically learn the relationship between juice concentration and magnetic induction tomography reconstruction data. This method not only improves the prediction accuracy of the model, but also makes the system adaptive, able to maintain efficient and accurate detection capabilities when facing different types or concentrations of juice, thereby greatly enhancing the practicality and scalability of the system.

[0086] In one possible embodiment, the concentration regression prediction model includes an input module, a convolution module, an attention module and an output module connected in sequence, wherein the convolution module includes a first convolution layer, a first batch processing unit, a RELU activation function, a second convolution layer and a second batch processing unit connected in sequence, the attention module includes an adaptive average pooling unit, a first fully connected layer, a RELU activation function, a second fully connected layer and a Sigmoid activation function connected in sequence, and the output module includes a third fully connected layer, a fourth fully connected layer and an output layer connected in sequence, wherein the first convolution layer is a 1x1 convolution kernel, a channel is 32, and a stride is 1, the second convolution layer is a 3x3 convolution kernel, an output channel is 64, and a stride is 1, the output layer has a single neuron, the input module and the attention module are connected by a residual connection, and the convolution module and the output module are connected by a channel weighted manner.

[0087] Among them, the output module of the model contains the third fully connected layer, the fourth fully connected layer and the output layer. Among them, the third and fourth fully connected layers are used to perform linear transformations. The number of input channels of the third connection layer is 32*(variable size / / 2) (the variable size in this technology is the number of magnetic induction tomography reconstruction data, that is, 56, and the " / / " symbol represents the integer division and rounding down symbol in Python, so the number of input channels of the third connection layer here should be 32*(56 / / 2)=896), the number of output channels is 32, the number of input channels of the fourth connection layer is 32, the number of output channels is 64, and the final output layer has a single neuron. This structure gradually reduces the number of channels, can effectively extract and compress data features, reduce computational complexity, and avoid overfitting. Through multiple layers of fully connected layers, the model has strong feature extraction and nonlinear mapping capabilities. The final output single neuron is suitable for regression or classification tasks, ensuring the accuracy of the results and the simplicity of the model.

[0088] It should be noted that the model combines the convolution module and the attention module. By stacking 1x1 and 3x3 convolution layers, it captures the features of the magnetic induction tomography reconstruction data corresponding to the juice, and uses batch processing and activation functions to improve the nonlinear expression ability of the model. The attention module focuses on important features through adaptive pooling and fully connected layers, enhancing the feature extraction ability of the model. Information transmission and feature fusion are optimized through residual connections and channel weighting, and finally a more accurate concentration prediction result is output. This architecture effectively improves the generalization ability and prediction accuracy of the model under complex data, ensuring the accuracy of juice concentration prediction.

[0089] In a possible implementation, S4 specifically includes:

[0090] S401: Initialize hyper parameters, where the hyper parameters include learning rate and batch size.

[0091] S402: Input the features in the data set into the concentration regression prediction model, and output the prediction label, i.e., the prediction value.

[0092] S403: Calculate the coefficient of determination and root mean square error between the actual value and the predicted value corresponding to the actual label corresponding to the feature.

[0093] S404: Adjust the hyperparameters through the Adam optimizer, increase the number of iterations by 1, and return to step S402 until the first condition or the second condition is met to obtain the trained concentration regression prediction model, wherein the first condition is that the determination coefficient is greater than the preset determination coefficient and the root mean square error is less than the preset root mean square error, and the second condition is that the number of iterations is greater than the preset number of iterations.

[0094] Optionally, the Adam optimizer can be used for model training, with a learning rate of 0.001, a batch size of 32, and a maximum number of training generations of 100. On this basis, the detection accuracy of the model is evaluated by calculating the coefficient of determination R2 and RMSE between the predicted value and the actual value.

[0095] It should be noted that this model training method introduces hyperparameters such as learning rate and batch size, and uses the Adam optimizer to dynamically adjust these hyperparameters, so that the model can be gradually optimized during the training process. By iteratively calculating the coefficient of determination and root mean square error of the predicted value, the model parameters are continuously adjusted until the preset accuracy requirements are met or the maximum number of iterations is reached. This training method has the advantages of fast convergence speed and high accuracy, which can ensure that the model reaches the best performance in a short time, thereby improving the accuracy and stability of juice concentration prediction.

[0096] S5: Obtain magnetic induction tomography reconstruction data to be detected.

[0097] S6: Inputting the magnetic induction tomography reconstruction data to be detected into the concentration regression prediction model to obtain the concentration of the juice to be detected.

[0098] S7: Conduct conductivity visualization reconstruction of the magnetic induction tomography reconstruction data of the juice to be detected is performed by combining the sensitivity matrix and the conjugate gradient algorithm to obtain foreign matter information in the juice to be detected, wherein the foreign matter information includes the size and position of the foreign matter.

[0099] Among them, the sensitivity matrix is ​​a matrix that describes the sensitivity of the system output to input changes. In magnetic induction tomography, it represents the response of each measurement point to changes in internal conductivity, helping to determine how the conductivity of different areas affects the measurement signal. The conjugate gradient algorithm is an iterative method for solving large-scale linear systems or optimization problems, and is particularly suitable for processing sparse matrices. It approximates the solution of the problem through step-by-step optimization and is an effective method for solving large-scale problems with limited computing resources. Combining the sensitivity matrix and the conjugate gradient algorithm to generate a conductivity distribution map can efficiently reconstruct the conductivity image inside the juice transmission pipeline, thereby accurately reflecting the concentration and internal structure of the juice. This method not only improves the accuracy of the reconstructed image, but is also more efficient in computation, suitable for processing complex juice samples, and improves the overall performance of the detection system.

[0100] In a possible implementation, S7 specifically includes:

[0101] S701: The detection area of ​​the juice transmission channel is segmented using the multi-physics software COMSOL.

[0102] S702: Perform finite element simulation calculation on each subdivided unit to obtain the induced voltage value and sensitivity matrix of the magnetic induction tomography reconstruction data of the juice to be detected.

[0103] S703: Using MATLAB, solve the linear equation representing image reconstruction based on the sensitivity matrix and the voltage value:

[0104] U=Sg

[0105] in, g Represents the grayscale value of the image and has a dimension of Conductivity distribution matrix, U The dimension is Voltage variation matrix, S The dimension is The sensitivity matrix, Indicates the number of combinations of collected data related to the number of air-core coils, Indicates the number of mesh elements.

[0106] S704: Inversely solve the linear equation using the conjugate gradient algorithm to obtain the conductivity distribution map:

[0107]

[0108] in, represents the initial value obtained by the Tikhonov regularization algorithm, S represents the sensitivity matrix, represents the transpose of the sensitivity matrix, represents the Tikhonov regularization factor, U represents the voltage variation matrix, I represents the identity matrix, represents the norm square operator, k represents the number of iterations, Indicates k The step size of the iteration, Indicates k The search direction of the iteration, Indicates k The residual vector of the iteration, Representation and number of iterations k+ 1-time related conductivity distribution diagram, Representation and number of iterations k The conductivity distribution diagram of the sub-correlation, represents the intermediate parameter value related to the number of iterations and the residual vector, Representation and number of iterations k -1 related search direction, Representation and number of iterations k -1-order correlation residual vector, Representation and number of iterations k -1 related step size.

[0109] S705: Obtain foreign matter information according to the conductivity distribution map.

[0110] It should be noted that by using the multi-physics software COMSOL to perform detection area segmentation and simulation calculations, high-precision induced voltage values ​​and sensitivity matrices were obtained, ensuring accurate modeling of the conductivity distribution. On this basis, the linear equations were solved by MATLAB and reconstructed using the conjugate gradient algorithm, and the stability and accuracy of the inversion were improved in combination with Tikhonov regularization. The overall process ensures that the reconstruction of the conductivity distribution map is highly accurate and robust, and can accurately reflect the physical properties inside the juice, thereby improving the detection effect of the juice purity detection system.

[0111] S8: Output the concentration of the juice to be tested and the foreign matter information, and complete the purity detection of the juice to be tested, that is, the foreign matter detection of the juice to be tested and the concentration detection of the juice to be tested.

[0112] In actual application, by constructing juice samples of different concentrations and obtaining their magnetic induction tomography reconstruction data, the system can accurately generate a conductivity distribution map. Subsequently, these magnetic induction tomography reconstruction data are used to establish a data set and train a concentration regression prediction model, so that the system can accurately predict the concentration of the juice to be tested and identify foreign matter. This process not only realizes non-invasive, high-precision juice purity detection, but also ensures the system's adaptability, real-time and continuity, effectively improving the efficiency and reliability of juice quality control.

[0113] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0114] In the present invention, since the fruit itself contains a large number of charged particles, it is an electrolyte that can form a bioelectric field. Therefore, the conductivity parameter is one of the important physical properties of juice products. Based on this physical property, a juice purity detection system with a coil sensor array is constructed. By detecting the electromagnetic characteristics inside the field of the object to be tested, non-invasive detection is achieved. The structure is simple and efficient. The magnetic induction tomography reconstruction data of the juice to be tested can be quickly collected, and the tomography reconstruction data is used as a feature. The corresponding juice concentration is used as a label to create a data set. The data set is used to train the constructed concentration regression prediction model to output the juice concentration. In addition, This method can be used to avoid the problem that the liquid transportation process and foreign matter in the pipeline cannot be visualized due to the tightness and opacity of the juice transmission pipeline during the production process. This technology is used to collect magnetic induction tomography reconstruction data, and reconstruct the electrical characteristic parameter distribution of the pipeline fault based on the magnetic induction tomography reconstruction data to identify foreign matter in the juice. Rapid scanning can be used to obtain a large amount of data to improve detection efficiency. Real-time conductivity reconstruction provides a basis for continuous monitoring of juice concentration and foreign matter detection. It can continuously and completely and accurately detect the juice to be tested, avoid the problem of missed detection, and promote the improvement of juice product quality and production safety.

[0115] Reference Manual Attached Figure 8 , showing a schematic structural diagram of another juice purity detection system provided by the present invention.

[0116] The present invention further provides a juice purity detection system 20, which is applied to the above-mentioned juice purity detection method, comprising:

[0117] Processor 201.

[0118] The memory 202 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 201, the juice purity detection method of the method embodiment is implemented.

[0119] The juice purity detection system 20 provided by the present invention can execute the above-mentioned juice purity detection method and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate on it.

[0120] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0121] In the present invention, since the fruit itself contains a large number of charged particles, it is an electrolyte that can form a bioelectric field. Therefore, the conductivity parameter is one of the important physical properties of juice products. Based on this physical property, a juice purity detection system with a coil sensor array is constructed. By detecting the electromagnetic characteristics inside the field of the object to be tested, non-invasive detection is achieved. The structure is simple and efficient. The magnetic induction tomography reconstruction data of the juice to be tested can be quickly collected, and the tomography reconstruction data is used as a feature. The corresponding juice concentration is used as a label to create a data set. The data set is used to train the constructed concentration regression prediction model to output the juice concentration. In addition, This method can be used to avoid the problem that the liquid transportation process and foreign matter in the pipeline cannot be visualized due to the tightness and opacity of the juice transmission pipeline during the production process. This technology is used to collect magnetic induction tomography reconstruction data, and reconstruct the electrical characteristic parameter distribution of the pipeline fault based on the magnetic induction tomography reconstruction data to identify foreign matter in the juice. Rapid scanning can be used to obtain a large amount of data to improve detection efficiency. Real-time conductivity reconstruction provides a basis for continuous monitoring of juice concentration and foreign matter detection. It can continuously and completely and accurately detect the juice to be tested, avoid the problem of missed detection, and promote the improvement of juice product quality and production safety.

[0122] It should be understood that the processor in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0123] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0124] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0125] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0126] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0127] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0128] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0129] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0130] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0131] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0132] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0133] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the 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, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0134] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the juice purity detection method as described in the method embodiment is implemented.

[0135] A computer-readable storage medium provided by the present invention can implement the steps and effects of the juice purity detection method of the above method embodiment, and to avoid repetition, the present invention will not go into details.

[0136] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0137] In the present invention, since the fruit itself contains a large number of charged particles, it is an electrolyte that can form a bioelectric field. Therefore, the conductivity parameter is one of the important physical properties of juice products. Based on this physical property, a juice purity detection system with a coil sensor array is constructed. By detecting the electromagnetic characteristics inside the field of the object to be tested, non-invasive detection is achieved. The structure is simple and efficient. The magnetic induction tomography reconstruction data of the juice to be tested can be quickly collected, and the tomography reconstruction data is used as a feature. The corresponding juice concentration is used as a label to create a data set. The data set is used to train the constructed concentration regression prediction model to output the juice concentration. In addition, This method can be used to avoid the problem that the liquid transportation process and foreign matter in the pipeline cannot be visualized due to the tightness and opacity of the juice transmission pipeline during the production process. This technology is used to collect magnetic induction tomography reconstruction data, and reconstruct the electrical characteristic parameter distribution of the pipeline fault based on the magnetic induction tomography reconstruction data to identify foreign matter in the juice. Rapid scanning can be used to obtain a large amount of data to improve detection efficiency. Real-time conductivity reconstruction provides a basis for continuous monitoring of juice concentration and foreign matter detection. It can continuously and completely and accurately detect the juice to be tested, avoid the problem of missed detection, and promote the improvement of juice product quality and production safety.

[0138] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0139] There are a few points to note:

[0140] (1) The drawings of the embodiments of the present invention only involve structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0141] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0142] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0143] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for detecting juice purity, characterized in that: Applied to a juice purity detection system, wherein the juice purity detection system comprises a host computer, a digital circuit module, a multiplexing module, a signal detection module and a coil sensor array connected in sequence, and a signal excitation module is respectively connected to the digital circuit and the multiplexing module; the method comprises: S1: Constructing blended juices with different juice concentrations; S2: obtaining magnetic induction tomography reconstruction data of various prepared juices through the juice purity detection system; S3: establishing a data set with the magnetic induction tomography reconstruction data as a feature and the juice concentration of the modulated juice corresponding to the magnetic induction tomography reconstruction data as a label; S4: constructing a concentration regression prediction model based on the data set; S5: obtaining magnetic induction tomography reconstruction data of the juice to be tested; S6: inputting the magnetic induction tomography reconstruction data of the juice to be tested into the concentration regression prediction model to obtain the concentration of the juice to be tested; S7: Combining the sensitivity matrix and the conjugate gradient algorithm, the magnetic induction tomography reconstruction data of the juice to be detected is reconstructed by conductivity visualization to obtain the foreign matter information in the juice to be detected, wherein the foreign matter information includes the size and position of the foreign matter; S8: Outputting the concentration of the juice to be detected and the foreign matter information, completing the purity detection of the juice to be detected, that is, the foreign matter detection of the juice to be detected and the concentration detection of the juice to be detected.

2. The method for detecting juice purity according to claim 1, characterized in that: The digital circuit module includes a DAC circuit, an FPGA main control circuit and an ADC circuit; The signal excitation module includes a power amplification circuit; The coil sensor array is composed of a plurality of hollow coils arranged around the juice transmission channel in a centrally equidistant manner; The signal detection module includes a signal relay, a signal amplifier, a filter and an amplitude and phase detection circuit; The host computer is connected to the coil sensor array via a first circuit, a second circuit, a third circuit and a fourth circuit respectively; The first circuit specifically includes the FPGA main control circuit, the DAC circuit, the power amplifier circuit, the multiplexing module and the signal detection module connected in sequence; The second circuit specifically includes the FPGA main control circuit, the DAC circuit, the multiplexing module and the signal detection module connected in sequence; The third circuit specifically includes the FPGA main control circuit, the multiplexing module and the signal detection module connected in sequence; The fourth circuit specifically includes the FPGA main control circuit, the ADC circuit, the multiplexing module and the signal detection module which are connected in sequence.

3. The method for detecting juice purity according to claim 1, characterized in that: The S1 is specifically: Original juice is obtained, and distilled water is added to the original juice to obtain a plurality of modulated juices of different concentrations.

4. The method for detecting juice purity according to claim 2, characterized in that: The S2 specifically includes: S201: Inputting the prepared juices of different concentrations into the juice transmission channel; S202: using any hollow coil as an excitation coil and the remaining hollow coils as detection coils to collect amplitude and phase data of the modulated juice; S203: replacing the hollow coil representing the excitation coil, and returning to step S202, until all the hollow coils are used as primary excitation coils, and multiple sets of amplitude and phase data are obtained; S204: Input the amplitude and phase data to the host computer, and output the magnetic induction tomography reconstruction data of the prepared juice: in, A Indicates including w The magnetic induction tomography reconstruction data matrix of juices modulated in different concentration categories, Indicates i Magnetic induction tomography reconstruction array of juices modulated by concentration categories, , Indicates p The first collection q The values ​​are the magnetic induction tomography reconstruction data, , , m Indicates the total number of collections. n Indicates the total number of collected values ​​for each concentration category of blended juice, U represents the magnetic induction tomography reconstruction data, V 1 indicates the voltage data when the detection area of ​​the juice transmission channel is the measured medium. V 0 represents the voltage data when the detection area of ​​the juice transmission channel is air. represents taking the imaginary part of a complex number, V represents a complex number representing voltage data, C represents the real part of a complex number, B represents the imaginary part of a complex number, j Represents the complex imaginary unit.

5. The method for detecting juice purity according to claim 1, characterized in that: The data set includes a training set and a test set divided according to a preset ratio.

6. The method for detecting juice purity according to claim 1, characterized in that: The concentration regression prediction model includes an input module, a convolution module, an attention module and an output module connected in sequence, wherein the convolution module includes a first convolution layer, a first batch processing unit, a RELU activation function, a second convolution layer and a second batch processing unit connected in sequence, the attention module includes an adaptive average pooling unit, a first fully connected layer, a RELU activation function, a second fully connected layer and a Sigmoid activation function connected in sequence, and the output module includes a third fully connected layer, a fourth fully connected layer and an output layer connected in sequence, wherein the first convolution layer is a 1x1 convolution kernel, a channel is 32, and a stride is 1, the second convolution layer is a 3x3 convolution kernel, an output channel is 64, and a stride is 1, the output layer has a single neuron, the input module is connected to the attention module through a residual connection, and the convolution module is connected to the output module through a channel weighted manner.

7. The method for detecting juice purity according to claim 1, characterized in that: The S4 specifically includes: S401: Initialize hyper parameters, wherein the hyper parameters include learning rate and batch size; S402: Input the features in the data set into the concentration regression prediction model, and output a prediction label, i.e., a prediction value; S403: Calculate the coefficient of determination and the root mean square error between the actual value corresponding to the actual label corresponding to the feature and the predicted value; S404: Adjust the hyperparameters through the Adam optimizer, increase the number of iterations by 1, and return to step S402 until the first condition or the second condition is met to obtain the trained concentration regression prediction model, wherein the first condition is that the determination coefficient is greater than the preset determination coefficient and the root mean square error is less than the preset root mean square error, and the second condition is that the number of iterations is greater than the preset number of iterations.

8. The method for detecting juice purity according to claim 1, characterized in that: The S7 specifically includes: S701: segmenting the detection area of ​​the juice transmission channel by using the multi-physics software COMSOL; S702: performing finite element simulation calculation on each subdivided unit to obtain an induced voltage value and a sensitivity matrix of magnetic induction tomography reconstruction data of the juice to be detected; S703: Using MATLAB, solve the linear equation representing image reconstruction according to the sensitivity matrix and the voltage value: U=Sg in, g Represents the grayscale value of the image and has a dimension of Conductivity distribution matrix, U The dimension is Voltage variation matrix, S The dimension is The sensitivity matrix, Indicates the number of combinations of collected data related to the number of air-core coils, Indicates the number of subdivision units; S704: Inversely solve the linear equation using a conjugate gradient algorithm to obtain a conductivity distribution map: in, represents the initial value obtained by the Tikhonov regularization algorithm, S represents the sensitivity matrix, represents the transpose of the sensitivity matrix, represents the Tikhonov regularization factor, U represents the voltage variation matrix, I represents the identity matrix, represents the norm square operator, k represents the number of iterations, Indicates k The step size of the iteration, Indicates k The search direction of the iteration, Indicates k The residual vector of the iteration, Representation and number of iterations k+ 1-time related conductivity distribution diagram, Representation and number of iterations k The conductivity distribution diagram of the sub-correlation, represents the intermediate parameter value related to the number of iterations and the residual vector, Representation and number of iterations k -1 related search direction, Representation and number of iterations k -1-order correlation residual vector, Representation and number of iterations k -1 related step size; S705: Acquire the foreign matter information according to the conductivity distribution map.

9. A juice purity detection system, characterized in that: include: processor; A memory, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, the juice purity detection method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the juice purity detection method according to any one of claims 1 to 8 is implemented.

Images