Parameter optimization method and system based on pomegranate function factor extraction

Abstract

The invention relates to the technical field of parameter optimization, in particular to a parameter optimization method and system based on pomegranate function factor extraction. The method comprises the following steps: firstly, acquiring reflected wave spectrum intensity of to-be-detected pomegranates through a surface wave detection technology, comparing the reflected wave spectrum intensity with standard spectrum intensity, and further classifying the to-be-detected pomegranates as cracked pomegranates or complete pomegranates; then, a weighted PID controller is provided, dissociation rate feedback in the dissociation process is received in real time through the weighted PID controller, the water flow speed and the water pressure are dynamically adjusted according to the dissociation rate, and then complete dissociation of the pomegranate seeds and the pomegranate rinds is achieved; finally, the invention provides a self-adaptive weight particle swarm optimization algorithm, and accurate adjustment of extraction parameters is realized by updating and optimizing key parameters in the pomegranate seed functional factor extraction process through dynamic weights, so that the efficiency and purity of functional factor extraction are improved.

Description

Technical Field

[0001] The invention relates to the technical field of parameter optimization, and in particular to a parameter optimization method and system based on pomegranate functional factor extraction. Background Art

[0002] Pomegranate is a nutritious fruit that contains many ingredients that are beneficial to human health. However, during the growth process of pomegranates, many pomegranates are discarded or wasted due to poor appearance due to the occurrence of fruit cracking. Cracked pomegranates usually lose their market value due to cracking of the peel or damage to the appearance. In fact, the intrinsic quality and nutritional components of cracked pomegranates are not damaged and they still have extremely high health value. By effectively processing and utilizing cracked pomegranates, not only can food waste be reduced, but also more cost-effective and healthy products can be provided to consumers. Improving the added value of cracked pomegranates, especially in the field of functional factor extraction, has important economic and social significance.

[0003] At present, there are some devices for cracked pomegranate detection and dissociation on the market, but most of them have not been able to effectively achieve efficient and accurate detection and dissociation effects in practical applications. In addition, in the extraction process of pomegranate functional factors, due to the differences in extraction parameters (such as ultrasonic frequency, power, temperature and solvent concentration, etc.), the extraction effects vary greatly, resulting in unstable quality of the final extract, which affects the comprehensive utilization of pomegranate functional factors. Therefore, there is an urgent need for more accurate detection and efficient dissociation of cracked pomegranates, as well as more optimized extraction parameter control schemes.

[0004] Therefore, a parameter optimization method and system based on pomegranate functional factor extraction was proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a parameter optimization method and system based on the extraction of pomegranate functional factors. Firstly, cracked pomegranates and intact pomegranates are efficiently classified by surface wave detection technology; then, the peel and seeds of cracked pomegranates are efficiently separated by a weighted PID control method to ensure that the dissociation rate reaches a set threshold; finally, an adaptive weighted particle swarm optimization algorithm is used to dynamically optimize the functional factor extraction parameters to obtain the optimal extraction parameters to improve the extraction efficiency and the purity of the functional factors, and finally achieve efficient utilization of cracked pomegranates.

[0006] To achieve the above object, the present invention provides the following technical solutions: A parameter optimization method based on pomegranate functional factor extraction, comprising: Washing and drying the pomegranates to be tested, and conveying the pomegranates to be tested to a testing area; The first piezoelectric transducer in the detection area emits a surface wave, which is converted into a reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; pre-processing and analyzing the reflected wave, calculating the test spectrum intensity of the reflected wave according to a spectrum intensity calculation formula; calculating the intensity difference between the test spectrum intensity and the standard spectrum intensity, and if the intensity difference is greater than a first threshold, classifying the pomegranate to be detected as a cracked pomegranate; otherwise, classifying it as a complete pomegranate; Splitting the cracked pomegranate into pomegranate pieces and dissociating the pomegranate pieces; receiving the dissociation rate in the dissociation process through a weighted PID controller, and adjusting the water flow parameters according to the dissociation rate; when the dissociation rate reaches a dissociation threshold, collecting and screening the pomegranate seeds and pomegranate peels; the water flow parameters include water flow velocity and water pressure; The functional factors of the pomegranate seeds are extracted, and the extraction parameters in the extraction process are updated and optimized by an adaptive weighted particle swarm optimization algorithm; after the update is completed, a functional factor extract and pomegranate seed residue are obtained, and the extraction parameters are recorded; the extraction parameters include ultrasonic frequency, ultrasonic power, extraction time, solvent concentration and extraction temperature.

[0007] Furthermore, the test spectrum intensity is calculated including: The first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected; the surface wave is converted into the reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; the reflected wave is amplified by an amplifier, and the noise of the reflected wave is removed by a high-pass filter to obtain a denoised reflected wave; the denoised reflected wave is subjected to a fast Fourier transform to obtain a test frequency domain signal; the test frequency domain signal is calculated according to a spectrum intensity calculation formula to obtain the test spectrum intensity.

[0008] Furthermore, the calculation formula of the weighted PID controller is: ; Wherein, t represents the time when the weighted PID controller receives the dissociation rate, represents the control quantity of the weighted PID controller, represents the water velocity weighting coefficient, represents the water pressure weighting coefficient, Indicates the PID control value of water flow rate, Indicates the water pressure PID control value.

[0009] Further, updating and optimizing the extraction parameters in the extraction process by the adaptive weighted particle swarm optimization algorithm includes: S401: Determine the number of particles in the particle swarm, randomly initialize the number of particles, initial position, initial velocity, initial inertia weight, individual optimal position and global optimal position of the particles in the particle swarm, and set the maximum number of updates and update threshold; S402: Calculating the current comfort level of each particle, and updating the individual optimal position and the global optimal position of each particle according to the current comfort level; S403: updating the particle speed and particle position of each particle according to the updated individual optimal position and the global optimal position of each particle, and adjusting the initial inertia weight according to the number of updates; S404: Determine whether the update satisfies the stop condition. If so, output the extraction parameters; if not, return to S402. The stop condition includes: the number of updates is greater than the maximum number of updates or the update change of the current comfort level is less than the update threshold.

[0010] Furthermore, the method also includes: collecting the pomegranate peel and the pomegranate seed residue, extracting functional components in the pomegranate peel and pomegranate seed oil in the pomegranate seed residue, wherein the functional components include polyphenols, tannins, alkaloids and organic acids.

[0011] The present invention also provides a parameter optimization system based on pomegranate functional factor extraction, comprising: A pomegranate pretreatment module, which cleans and dries the pomegranates to be tested, and transfers the pomegranates to be tested to a testing area; The pomegranate sorting module comprises: a first piezoelectric transducer in the detection zone transmits a surface wave, the surface wave is converted into a reflected wave after propagating along the pomegranate to be detected, and a second piezoelectric transducer in the detection zone receives the reflected wave; the reflected wave is pre-processed and analyzed, and a test spectrum intensity of the reflected wave is calculated according to a spectrum intensity calculation formula; an intensity difference between the test spectrum intensity and the standard spectrum intensity is calculated, and if the intensity difference is greater than a first threshold value, the pomegranate to be detected is classified as a cracked pomegranate; otherwise, it is classified as a complete pomegranate; A cracked pomegranate dissociation module is used to split the cracked pomegranate into pomegranate pieces and dissociate the pomegranate pieces; a dissociation rate in the dissociation process is received by a weighted PID controller, and a water flow parameter is adjusted according to the dissociation rate; when the dissociation rate reaches a dissociation threshold, pomegranate seeds and pomegranate peels are collected and sieved; the water flow parameters include water flow velocity and water pressure; The functional factor extraction module extracts the functional factors of the pomegranate seeds, and updates and optimizes the extraction parameters in the extraction process through an adaptive weighted particle swarm optimization algorithm; after the update is completed, the functional factor extract and pomegranate seed residue are obtained, and the extraction parameters are recorded; the extraction parameters include ultrasonic frequency, ultrasonic power, extraction time, solvent concentration and extraction temperature.

[0012] Furthermore, it is characterized in that calculating the test spectrum intensity includes: The first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected; the surface wave is converted into the reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; the reflected wave is amplified by an amplifier, and the noise of the reflected wave is removed by a high-pass filter to obtain a denoised reflected wave; the denoised reflected wave is subjected to a fast Fourier transform to obtain a test frequency domain signal; the test frequency domain signal is calculated according to a spectrum intensity calculation formula to obtain the test spectrum intensity.

[0013] Furthermore, the calculation formula of the weighted PID controller is: ; Wherein, t represents the time when the weighted PID controller receives the dissociation rate, represents the control quantity of the weighted PID controller, represents the water velocity weighting coefficient, represents the water pressure weighting coefficient, Indicates the PID control value of water flow rate, Indicates the water pressure PID control value.

[0014] Further, updating and optimizing the extraction parameters in the extraction process by the adaptive weighted particle swarm optimization algorithm includes: S401: Determine the number of particles in the particle swarm, randomly initialize the number of particles, initial position, initial velocity, initial inertia weight, individual optimal position and global optimal position of the particles in the particle swarm, and set the maximum number of updates and update threshold; S402: Calculating the current comfort level of each particle, and updating the individual optimal position and the global optimal position of each particle according to the current comfort level; S403: updating the particle speed and particle position of each particle according to the updated individual optimal position and the global optimal position of each particle, and adjusting the initial inertia weight according to the number of updates; S404: Determine whether the update satisfies the stop condition. If so, output the extraction parameters; if not, return to S402. The stop condition includes: the number of updates is greater than the maximum number of updates or the update change of the current comfort level is less than the update threshold.

[0015] Furthermore, the system further comprises: The residual material utilization module is used to collect the pomegranate peel and the pomegranate seed residue, and extract the functional components in the pomegranate peel and the pomegranate seed oil in the pomegranate seed residue, wherein the functional components include polyphenols, tannins, alkaloids and organic acids.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention transmits surface waves and receives reflected waves through a piezoelectric transducer, and combines the calculation and analysis of spectrum intensity to efficiently and accurately identify surface defects of the pomegranate to be tested; cracked pomegranates and intact pomegranates are automatically classified by the difference between the test spectrum intensity and the standard spectrum intensity, thereby achieving non-destructive and rapid quality detection, which not only improves the detection efficiency, but also provides a reliable classification basis for subsequent cracked pomegranate processing, and indirectly improves the efficiency and accuracy of parameter optimization in the extraction of pomegranate functional factors.

[0017] 2. The present invention proposes a weighted PID controller, which receives dissociation rate feedback in the dissociation process in real time and dynamically adjusts the water flow velocity and water pressure according to the dissociation rate, thereby realizing precise control of water flow parameters. The controller effectively improves the dissociation rate, avoids incomplete dissociation due to inappropriate parameters, ensures that the separation of pomegranate seeds and peels is more efficient and stable, and reduces the error of parameter optimization in the extraction of pomegranate functional factors due to incomplete dissociation.

[0018] 3. The present invention proposes an adaptive weighted particle swarm optimization algorithm, which dynamically updates and optimizes key parameters in the extraction process of pomegranate seed functional factors, and realizes precise adjustment of factors such as ultrasonic frequency, power, extraction time, solvent concentration and extraction temperature, thereby improving the efficiency and purity of functional factor extraction; this method not only optimizes the extraction process and reduces energy consumption, but also ensures the quality and stability of the extract, providing scientific support for the efficient utilization of pomegranate functional factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a parameter optimization method based on pomegranate functional factor extraction provided in Example 1 of the present invention; Figure 2 A flow chart of an adaptive weighted particle swarm optimization algorithm provided in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the structure of a parameter optimization system based on pomegranate functional factor extraction according to Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0020] 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. Embodiment 1

[0021] A beverage company A introduced a parameter optimization method based on the extraction of pomegranate functional factors provided by the present invention in the production process of pomegranate beverages to achieve effective utilization of cracked pomegranates, improve the extraction efficiency of cracked pomegranate functional factors, and reduce the loss of functional factors caused by parameter design during the extraction process. The specific process of the method is as follows: Figure 1 As shown, the specific implementation is as follows: S100: washing and drying the pomegranate to be tested, and transferring the pomegranate to be tested to a testing area; S200: The first piezoelectric transducer in the detection area transmits a surface wave, and the surface wave is converted into a reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; pre-processing and analyzing the reflected wave, and calculating the test spectrum intensity of the reflected wave according to a spectrum intensity calculation formula; calculating the intensity difference between the test spectrum intensity and the standard spectrum intensity, and if the intensity difference is greater than a first threshold, classifying the pomegranate to be detected as a cracked pomegranate; otherwise, classifying it as a whole pomegranate; Furthermore, the test spectrum intensity is calculated including: The first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected; the surface wave is converted into a reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer receives the reflected wave; an amplifier is used to amplify the reflected wave, and a high-pass filter is used to remove the noise of the reflected wave to obtain a denoised reflected wave; a fast Fourier transform is performed on the denoised reflected wave to obtain a test frequency domain signal; the test frequency domain signal is calculated according to a spectrum intensity calculation formula to obtain the test spectrum intensity.

[0022] Table 1. Reflection wave data of cracked pomegranate and intact pomegranate

[0023] Specifically, the first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected. The surface wave generated by the first piezoelectric transducer is selected in the range of 100kHz to 1MHz; the surface wave will attenuate after propagating along the pomegranate to be detected. As shown in Table 1, the attenuation degree between the cracked pomegranate and the intact pomegranate is different, and the reflected wave received by the second piezoelectric transducer will also be different; an amplifier is used to amplify the reflected wave, and a high-pass filter is used to remove the noise of the reflected wave to obtain a denoised reflected wave. The calculation formula of the denoised reflected wave is: ; in, represents the denoised reflected wave, G represents the amplification factor, represents the reflected wave, H() represents the filtering function of the high-pass filter; the denoised reflected wave is subjected to fast Fourier transform to obtain the test frequency domain signal, and the calculation formula of the test frequency domain signal is: ; Where f represents the frequency, represents the test frequency domain signal, and F() represents the fast Fourier transform function; the test frequency domain signal is calculated according to the spectrum intensity calculation formula to obtain the test spectrum intensity, where the calculation formula for the test spectrum intensity is: ; in, Indicates the test spectrum intensity, Represents the modulus of the test frequency domain signal.

[0024] By starting the first piezoelectric transducer in the detection area to generate a pulse signal and apply a surface wave to the pomegranate to be detected, the second piezoelectric transducer is used to receive the reflected wave, and the amplifier and high-pass filter are used to amplify the signal and denoise it respectively, so as to effectively obtain the pure time domain signal of the reflected wave; after being converted into a frequency domain signal by fast Fourier transform, the test spectrum intensity is calculated based on the spectrum intensity calculation formula, thereby realizing efficient and accurate detection of the surface characteristics of the pomegranate, and providing a precise basis for identifying cracked pomegranates.

[0025] Further, the intensity difference between the test spectrum intensity and the standard spectrum intensity is calculated. If the intensity difference is greater than the first threshold, the pomegranate to be tested is classified as a cracked pomegranate, and the cracked pomegranate is sorted. The intensity difference is calculated as follows: ; in, represents the intensity difference, In this embodiment, the value of the standard spectrum intensity is 800, which is obtained by calculating the spectrum intensity of multiple groups of complete pomegranates, the value of the first threshold is 100, which is obtained by Gaussian distribution, and the comparison of the spectrum intensity of cracked pomegranates and complete pomegranates is shown in Table 2.

[0026] Table 2. Comparison of spectrum intensity between cracked pomegranate and intact pomegranate

[0027] S300: splitting the cracked pomegranate into pomegranate pieces and dissociating the pomegranate pieces; receiving a dissociation rate in the dissociation process through a weighted PID controller, and adjusting a water flow parameter according to the dissociation rate; when the dissociation rate reaches a dissociation threshold, collecting and screening pomegranate seeds and pomegranate peels; the water flow parameters include water flow velocity and water pressure; Furthermore, the calculation formula of the weighted PID controller is: ; ; ; ; Wherein, t represents the time when the weighted PID controller receives the feedback signal, represents the control quantity of the weighted PID controller, represents the water velocity weighting coefficient, represents the water pressure weighting coefficient, Indicates the PID control value of water flow rate, Indicates the water pressure PID control amount, , and represents the gain coefficient of the water flow velocity PID controller, , and represents the gain coefficient of the water pressure PID controller, represents the current dissociation error, represents the dissociation threshold, Represents the current dissociation rate.

[0028] Specifically, a splitting device is used to split the cracked pomegranate into pomegranate pieces, and the pomegranate pieces are placed in a dissociation device for further dissociation, wherein the dissociation device uses a hydraulic method to dissociate the pomegranate pieces. During dissociation, the control host needs to set a dissociation threshold, an initial water flow rate, and an initial water pressure. In this embodiment, the dissociation threshold is set to 100%, the initial water flow rate is set to 2.5 meters per second, and the initial water pressure is set to 3 bar, wherein bar is a commonly used pressure measurement unit in the industrial field, and 1 bar=100 kPa; during the dissociation process, the sensor detects the current dissociation rate in real time and compares it with the dissociation threshold, and then controls the water flow rate and water pressure in real time through a weighted PID controller; when the dissociation rate reaches the dissociation threshold, the dissociation device is stopped, and the pomegranate seeds and pomegranate peels are collected and sieved.

[0029] The weighted PID controller receives the dissociation rate feedback signal in real time during the dissociation process, and dynamically adjusts the water flow parameters based on the dissociation rate, thereby achieving precise control of the water flow rate and water pressure. This method effectively optimizes the dissociation rate, avoids energy waste or insufficient dissociation caused by excessively high or low water flow parameters, and ensures that the separation process of pomegranate seeds and peels is more efficient.

[0030] S400: extracting the functional factors of the pomegranate seeds, and updating and optimizing the extraction parameters in the extraction process by using an adaptive weighted particle swarm optimization algorithm; obtaining a functional factor extract and pomegranate seed residue after the update, and recording the extraction parameters; the extraction parameters include ultrasonic frequency, ultrasonic power, extraction time, solvent concentration and extraction temperature.

[0031] Furthermore, if Figure 2 As shown, updating and optimizing the extraction parameters in the extraction process by the adaptive weighted particle swarm optimization algorithm includes: S401: Determine the number of particles in the particle swarm, randomly initialize the number of particles, initial position, initial velocity, initial inertia weight, individual optimal position and global optimal position of the particles in the particle swarm, and set the maximum number of updates and update threshold; S402: Calculating the current comfort level of each particle, and updating the individual optimal position and the global optimal position of each particle according to the current comfort level; S403: updating the particle speed and particle position of each particle according to the updated individual optimal position and the global optimal position of each particle, and adjusting the initial inertia weight according to the number of updates; S404: Determine whether the update satisfies the stop condition, and if so, output the extraction parameters; if not, return to S402; the stop condition includes: the number of updates is greater than the maximum number of updates or the update change of the current comfort level is less than the update threshold.

[0032] Specifically, the size of the particle swarm, that is, the number of particles N, is determined, and the position of each particle is randomly initialized. The position vector of each particle corresponds to a set of extraction parameters, including ultrasonic frequency, ultrasonic power, extraction time, solvent concentration, and extraction temperature. The particle position information can be expressed as: ; in, represents the position of the ith particle, m represents the mth extraction parameter of the particle, and the speed of each particle is randomly initialized to adjust the position of the particle in the search space, where the speed information of the particle can be expressed as: ; in, represents the speed of the ith particle, m represents the mth extraction parameter of the particle, and sets the initial inertia weight For each particle, its individual optimal position is initially set is the current position of the particle, the global optimal position Set to the position of the particle with the best fitness among all particles; for each particle, calculate the comfort of its current position. The comfort indicates the quality of the extraction effect, which can be expressed as: ; in, represents the comfort level of the i-th particle at the current position, Represents the comfort calculation function; for each particle, if the current objective function value is better than the individual optimal position of the particle's history The comfort level of the particle is updated as follows: ; If the objective function value of a particle is better than the global optimal position The objective function value of , then update the global optimal position: ; According to the particle swarm optimization formula, update the particle speed: ; in, represents the particle velocity at time t+1, represents the inertia weight at the current time t, represents the particle velocity at the current time t, and represents the learning factor, usually a constant, and Represents a random number in the range [0,1], Indicates the particle position at the current time t. Update the particle position according to the updated velocity: ; in, Represents the particle position at time t+1, and adjusts the initial inertia weight according to the number of updates: ; in, represents the maximum value of the inertia weight, represents the minimum value of the inertia weight. In this embodiment, =2, =1; determine whether the update meets the stop condition, if so, output the extraction parameters; if not, calculate the updated comfort level, and update the individual optimal position and the global optimal position again; the stop condition includes: the number of updates is greater than the maximum number of updates Or the updated change in comfort level is less than the update threshold In this embodiment, the maximum number of updates is =1000, update threshold =0.01.

[0033] By dynamically updating and optimizing the extraction parameters during the extraction process through the adaptive weighted particle swarm optimization algorithm, the search strategy can be intelligently adjusted according to the changes in the objective function, thereby improving the global search capability and convergence speed of the algorithm, thereby accurately finding the best combination of optimization parameters and effectively improving the efficiency and purity of the extraction of pomegranate functional factors.

[0034] Furthermore, the method also includes: collecting the pomegranate peel and the pomegranate seed residue, extracting functional components in the pomegranate peel and pomegranate seed oil in the pomegranate seed residue, wherein the functional components include polyphenols, tannins, alkaloids, and organic acids.

[0035] By collecting pomegranate peel and pomegranate seed residue and extracting the functional components and pomegranate seed oil respectively, the efficient and comprehensive utilization of pomegranate by-products is achieved.

[0036] In this embodiment, a piezoelectric transducer is used to emit surface waves and receive reflected waves, and combined with spectrum intensity calculation and difference analysis, cracked pomegranates are accurately identified; the cracked pomegranates are split and a weighted PID controller is used to adjust the water flow rate and water pressure to optimize the dissociation rate, and the pomegranate seeds and pomegranate peels are efficiently separated; then, the functional factors are extracted from the pomegranate seeds, and key parameters such as ultrasonic frequency, power, extraction time, solvent concentration and temperature are dynamically adjusted through an adaptive weighted particle swarm optimization algorithm to obtain high-purity functional factor extract and pomegranate seed residue, and the optimized extraction parameters are fully recorded, thereby providing a scientific parameter optimization method for the extraction of pomegranate functional factors. Embodiment 2

[0037] In order to support local agriculture and reduce the waste caused by cracked pomegranates, a pomegranate beverage processing company B established a production line that integrates cracked pomegranate screening, dissociation and functional factor extraction to achieve the reuse of cracked pomegranates. The production line adopts a parameter optimization system based on the extraction of pomegranate functional factors, including: A pomegranate pretreatment module, which cleans and dries the pomegranates to be tested, and transfers the pomegranates to be tested to a testing area; The pomegranate sorting module comprises: a first piezoelectric transducer in the detection zone transmits a surface wave, the surface wave is converted into a reflected wave after propagating along the pomegranate to be detected, and a second piezoelectric transducer in the detection zone receives the reflected wave; the reflected wave is pre-processed and analyzed, and a test spectrum intensity of the reflected wave is calculated according to a spectrum intensity calculation formula; an intensity difference between the test spectrum intensity and the standard spectrum intensity is calculated, and if the intensity difference is greater than a first threshold value, the pomegranate to be detected is classified as a cracked pomegranate; otherwise, it is classified as a complete pomegranate; A cracked pomegranate dissociation module is used to split the cracked pomegranate into pomegranate pieces and dissociate the pomegranate pieces; a dissociation rate in the dissociation process is received by a weighted PID controller, and a water flow parameter is adjusted according to the dissociation rate; when the dissociation rate reaches a dissociation threshold, pomegranate seeds and pomegranate peels are collected and sieved; the water flow parameters include water flow velocity and water pressure; The functional factor extraction module extracts the functional factors of the pomegranate seeds, and updates and optimizes the extraction parameters in the extraction process through an adaptive weighted particle swarm optimization algorithm; after the update is completed, the functional factor extract and pomegranate seed residue are obtained, and the extraction parameters are recorded; the extraction parameters include ultrasonic frequency, ultrasonic power, extraction time, solvent concentration and extraction temperature.

[0038] Furthermore, it is characterized in that calculating the test spectrum intensity includes: The first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected; the surface wave is converted into the reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; the reflected wave is amplified by an amplifier, and the noise of the reflected wave is removed by a high-pass filter to obtain a denoised reflected wave; the denoised reflected wave is subjected to a fast Fourier transform to obtain a test frequency domain signal; the test frequency domain signal is calculated according to a spectrum intensity calculation formula to obtain the test spectrum intensity.

[0039] Furthermore, the calculation formula of the weighted PID controller is: ; Wherein, t represents the time when the weighted PID controller receives the dissociation rate, represents the control quantity of the weighted PID controller, represents the water velocity weighting coefficient, represents the water pressure weighting coefficient, Indicates the PID control value of water flow rate, Indicates the water pressure PID control value.

[0040] Further, updating and optimizing the extraction parameters in the extraction process by the adaptive weighted particle swarm optimization algorithm includes: S401: Determine the number of particles in the particle swarm, randomly initialize the number of particles, initial position, initial velocity, initial inertia weight, individual optimal position and global optimal position of the particles in the particle swarm, and set the maximum number of updates and update threshold; S402: Calculating the current comfort level of each particle, and updating the individual optimal position and the global optimal position of each particle according to the current comfort level; S403: updating the particle speed and particle position of each particle according to the updated individual optimal position and the global optimal position of each particle, and adjusting the initial inertia weight according to the number of updates; S404: Determine whether the update satisfies the stop condition. If so, output the extraction parameters; if not, return to S402. The stop condition includes: the number of updates is greater than the maximum number of updates or the update change of the current comfort level is less than the update threshold.

[0041] Furthermore, the system further comprises: The residual material utilization module is used to collect the pomegranate peel and the pomegranate seed residue, and extract the functional components in the pomegranate peel and the pomegranate seed oil in the pomegranate seed residue, wherein the functional components include polyphenols, tannins, alkaloids and organic acids.

[0042] In the process of designing the pomegranate sorting module, the detection scheme 1 based on YOLO V5 and the current detection scheme 2 based on surface waves were compared. The comparison results are shown in Table 3.

[0043] It can be seen from Table 3 that the surface wave detection scheme 2 currently used is better than the detection scheme 1 based on YOLO V5. The detection speed of a single pomegranate in detection scheme 2 is 0.11 seconds per piece. The main reason is that the surface wave transmission and calculation process are simple, avoiding complex neural network calculations. The detection rate of cracked pomegranates in detection scheme 2 is 96%, which is higher than 87% of detection scheme 1. This is because the wave transmission will be along the surface of the pomegranate, and the result will not fluctuate greatly due to the different positions of the pomegranates.

[0044] In the process of designing the cracked pomegranate dissociation module, the traditional roller press-based dissociation scheme 1 and the current weighted PID controller-based detection scheme 2 were compared, and the comparison results are shown in Table 4.

[0045] Table 3. Data comparison of different detection schemes

[0046] It can be seen from Table 3 that the detection scheme 2 based on the weighted PID controller currently used is superior to the traditional dissociation scheme 1 based on the crushing box in terms of dissociation rate and loss rate. The main reason is that the dissociation equipment based on the weighted PID controller dissociates the cracked pomegranates through hydraulic methods, avoiding the loss of fruit particles caused by collision and friction with mechanical equipment. At the same time, the weighted PID controller can receive the real-time dissociation rate, and with a dissociation rate of 100% as the goal, continuously optimize the water flow rate and water pressure, thereby achieving a better dissociation effect.

[0047] Table 4. Data comparison of different dissociation schemes

[0048] In the process of designing the functional factor extraction module, the extraction scheme 1 based on the public formula and the extraction scheme 2 based on the adaptive weighted particle swarm optimization algorithm at this stage were compared. The extraction scheme 1 showed an ultrasonic frequency of 80kHz, an ultrasonic power of 180W, an extraction time of 30 minutes, a solvent concentration of 34% by volume ethanol solution, and an extraction temperature of 51°C. The extraction scheme 2 showed an ultrasonic frequency of 95kHz, an ultrasonic power of 210W, an extraction time of 46 minutes, a solvent concentration of 57% by volume ethanol solution, and an extraction temperature of 46°C. The comparison results are shown in Table 5.

[0049] Table 5. Data comparison of different extraction schemes

[0050] As can be seen from Table 5, the extraction rates of various functional factors of the extraction scheme 2 based on the adaptive weighted particle swarm optimization algorithm currently used are higher than those of the public formula, among which the extraction rates of polyphenols, flavonoids and anthocyanins are 33.68%, 5.23% and 2.31%, respectively. The results show that the adaptive weighted particle swarm optimization algorithm can optimize the extraction parameters of the existing formula, thereby providing more scientific and efficient functional factor extraction parameters.

[0051] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A parameter optimization method based on pomegranate functional factor extraction, characterized in that: include: Washing and drying the pomegranates to be tested, and conveying the pomegranates to be tested to a testing area; The first piezoelectric transducer in the detection area transmits a surface wave, which is converted into a reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; pre-processing and analyzing the reflected wave, and calculating the test spectrum intensity of the reflected wave according to the spectrum intensity calculation formula; Calculating the intensity difference between the test spectrum intensity and the standard spectrum intensity, if the intensity difference is greater than a first threshold, classifying the pomegranate to be tested as a cracked pomegranate; otherwise, classifying it as a whole pomegranate; Splitting the cracked pomegranate into pomegranate pieces and dissociating the pomegranate pieces; receiving a dissociation rate during the dissociation process through a weighted PID controller, and adjusting a water flow parameter according to the dissociation rate; When the dissociation rate reaches the dissociation threshold, collecting and sieving the pomegranate seeds and pomegranate peels; the water flow parameters include water flow velocity and water pressure; The functional factors of the pomegranate seeds are extracted, and the extraction parameters in the extraction process are updated and optimized by an adaptive weighted particle swarm optimization algorithm; after the update is completed, a functional factor extract and pomegranate seed residue are obtained, and the extraction parameters are recorded; the extraction parameters include ultrasonic frequency, ultrasonic power, extraction time, solvent concentration and extraction temperature.

2. A parameter optimization method based on pomegranate functional factor extraction according to claim 1, characterized in that, The calculated test spectrum intensity includes: The first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected; the surface wave is converted into the reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; the reflected wave is amplified by an amplifier, and the noise of the reflected wave is removed by a high-pass filter to obtain a denoised reflected wave; the denoised reflected wave is subjected to a fast Fourier transform to obtain a test frequency domain signal; the test frequency domain signal is calculated according to a spectrum intensity calculation formula to obtain the test spectrum intensity.

3. A parameter optimization method based on pomegranate functional factor extraction according to claim 1, characterized in that, The calculation formula of the weighted PID controller is: ; Wherein, t represents the time when the weighted PID controller receives the dissociation rate, represents the control quantity of the weighted PID controller, represents the water velocity weighting coefficient, represents the water pressure weighting coefficient, Indicates the PID control value of water flow rate, Indicates the water pressure PID control value.

4. A parameter optimization method based on pomegranate functional factor extraction according to claim 1, characterized in that, Updating and optimizing the extraction parameters in the extraction process by the adaptive weighted particle swarm optimization algorithm includes: S401: Determine the number of particles in the particle swarm, randomly initialize the number of particles, initial position, initial velocity, initial inertia weight, individual optimal position and global optimal position of the particles in the particle swarm, and set the maximum number of updates and update threshold; S402: Calculating the current comfort level of each particle, and updating the individual optimal position and the global optimal position of each particle according to the current comfort level; S403: updating the particle speed and particle position of each particle according to the updated individual optimal position and the global optimal position of each particle, and adjusting the initial inertia weight according to the number of updates; S404: Determine whether the update satisfies the stop condition. If so, output the extraction parameters; if not, return to S402. The stop condition includes: the number of updates is greater than the maximum number of updates or the update change of the current comfort level is less than the update threshold.

5. A parameter optimization method based on pomegranate functional factor extraction according to claim 1, characterized in that, The method further comprises: collecting the pomegranate peel and the pomegranate seed residue, extracting functional components in the pomegranate peel and pomegranate seed oil in the pomegranate seed residue, wherein the functional components include polyphenols, tannins, alkaloids and organic acids.

6. A parameter optimization system based on pomegranate functional factor extraction, characterized in that: include: A pomegranate pretreatment module, which cleans and dries the pomegranates to be tested, and transfers the pomegranates to be tested to a testing area; The pomegranate sorting module comprises: a first piezoelectric transducer in the detection zone transmits a surface wave, the surface wave is converted into a reflected wave after propagating along the pomegranate to be detected, and a second piezoelectric transducer in the detection zone receives the reflected wave; the reflected wave is preprocessed and analyzed, and the test spectrum intensity of the reflected wave is calculated according to a spectrum intensity calculation formula; Calculating the intensity difference between the test spectrum intensity and the standard spectrum intensity, if the intensity difference is greater than a first threshold, classifying the pomegranate to be tested as a cracked pomegranate; otherwise, classifying it as a whole pomegranate; A cracked pomegranate dissociation module is used to split the cracked pomegranate into pomegranate pieces and dissociate the pomegranate pieces; a dissociation rate in the dissociation process is received by a weighted PID controller, and a water flow parameter is adjusted according to the dissociation rate; When the dissociation rate reaches the dissociation threshold, collecting and sieving the pomegranate seeds and pomegranate peels; the water flow parameters include water flow velocity and water pressure; The functional factor extraction module extracts the functional factors of the pomegranate seeds, and updates and optimizes the extraction parameters in the extraction process through an adaptive weighted particle swarm optimization algorithm; after the update is completed, the functional factor extract and pomegranate seed residue are obtained, and the extraction parameters are recorded; the extraction parameters include ultrasonic frequency, ultrasonic power, extraction time, solvent concentration and extraction temperature.

7. A parameter optimization system based on pomegranate functional factor extraction according to claim 6, characterized in that: It is characterized in that The calculated test spectrum intensity includes: The first piezoelectric transducer in the detection area is started to generate a pulse signal as a surface wave to act on the pomegranate to be detected; the surface wave is converted into the reflected wave after propagating along the pomegranate to be detected, and the second piezoelectric transducer in the detection area receives the reflected wave; the reflected wave is amplified by an amplifier, and the noise of the reflected wave is removed by a high-pass filter to obtain a denoised reflected wave; the denoised reflected wave is subjected to a fast Fourier transform to obtain a test frequency domain signal; the test frequency domain signal is calculated according to a spectrum intensity calculation formula to obtain the test spectrum intensity.

8. A parameter optimization system based on pomegranate functional factor extraction according to claim 6, characterized in that: The calculation formula of the weighted PID controller is: ; Wherein, t represents the time when the weighted PID controller receives the dissociation rate, represents the control quantity of the weighted PID controller, represents the water velocity weighting coefficient, represents the water pressure weighting coefficient, Indicates the PID control value of water flow rate, Indicates the water pressure PID control value.

9. A parameter optimization system based on pomegranate functional factor extraction according to claim 6, characterized in that: Updating and optimizing the extraction parameters in the extraction process by the adaptive weighted particle swarm optimization algorithm includes: S401: Determine the number of particles in the particle swarm, randomly initialize the number of particles, initial position, initial velocity, initial inertia weight, individual optimal position and global optimal position of the particles in the particle swarm, and set the maximum number of updates and update threshold; S402: Calculating the current comfort level of each particle, and updating the individual optimal position and the global optimal position of each particle according to the current comfort level; S403: updating the particle speed and particle position of each particle according to the updated individual optimal position and the global optimal position of each particle, and adjusting the initial inertia weight according to the number of updates; S404: Determine whether the update satisfies the stop condition. If so, output the extraction parameters; if not, return to S402. The stop condition includes: the number of updates is greater than the maximum number of updates or the update change of the current comfort level is less than the update threshold.

10. A parameter optimization system based on pomegranate functional factor extraction according to claim 6, characterized in that: The system further comprises: The residual material utilization module is used to collect the pomegranate peel and the pomegranate seed residue, and extract the functional components in the pomegranate peel and the pomegranate seed oil in the pomegranate seed residue, wherein the functional components include polyphenols, tannins, alkaloids and organic acids.

Images