An AI-optimized fully automatic pet health product processing system and method

Abstract

The present invention relates to the technical field of pet health product production and processing, and discloses a fully automatic pet health product processing system and method optimized by AI. Among them, a fully automatic pet health product processing method optimized by AI includes: constructing a pulsed electric field-assisted low-temperature sterilization system; establishing a microbial electric field sensitivity database to achieve precise customization of electric field parameters; analyzing the thermal response characteristics of different active ingredients in pet health products, constructing an inverse algorithm of the heat diffusion equation to achieve precise reconstruction of the internal temperature field of the material; combining Raman spectroscopy technology with a thermal imaging system to achieve real-time monitoring of the content of active ingredients, and establishing a thermal degradation kinetic model of active ingredients; combining the multi-objective Pareto optimization method to achieve the best balance between the sterilization effect and the retention rate of different active ingredients; the present invention solves the technical problem of serious loss of active ingredients in traditional thermal sterilization processes and improves the retention rate of heat-sensitive ingredients in pet health products.

Description

Technical Field

[0001] The present invention relates to the technical field of pet health product production and processing, and more specifically, it relates to a fully automatic pet health product processing system and method optimized based on AI. Background Art

[0002] With the expansion of the pet market scale and the increasing attention of pet owners to pet health, the market demand for pet health products is continuously growing. Pet health products usually contain various active ingredients, such as vitamins, probiotics, enzyme preparations, etc. These ingredients are sensitive to temperature and are prone to inactivation or degradation during traditional thermal processing.

[0003] During the production of pet health products, traditional thermal sterilization technology is generally adopted. Although it can effectively kill microorganisms, it will significantly reduce the content of active ingredients, resulting in a decrease in product efficacy; the sensitivity of different active ingredients in pet health products to heat varies greatly, and it is difficult for traditional single temperature control strategies to ensure the activity of multiple ingredients at the same time; traditional thermal processing technology is difficult to precisely control the local temperature distribution and cannot monitor the change of active ingredient content in real time, resulting in serious loss of nutrients during the processing.

[0004] Although non-thermal sterilization technologies such as high-pressure treatment and ultraviolet irradiation can reduce the loss of active ingredients to a certain extent, they have disadvantages such as complex equipment, poor treatment uniformity, high cost, etc., and it is difficult to accurately regulate pet health products with complex formulations. Therefore, a pet health product processing method that can maximize the retention of various thermosensitive active ingredients while ensuring the sterilization effect is needed. Summary of the Invention

[0005] The present invention provides a fully automatic pet health product processing system and method optimized based on AI to solve the technical problem of the technical contradiction between pet health product sterilization and maintaining active ingredients in related technologies.

[0006] The present invention provides a fully automatic pet health product processing method optimized based on AI, including the following steps:

[0007] Construct a pulsed electric field-assisted low-temperature sterilization system, including a pulsed electric field generation unit, a temperature control unit, a product flow unit, and a control unit;

[0008] Establish a microbial electric field sensitivity database, and realize precise customization of electric field parameters according to the electric field sensitivity characteristics of common microorganisms in pet health products;

[0009] Based on the microbial electric field sensitivity database, analyze the thermal response characteristics of different active ingredients in pet health products, construct an inversion algorithm for the heat diffusion equation, and realize the precise reconstruction of the temperature field inside the material;

[0010] Using the inversion results of the heat diffusion equation, combining Raman spectroscopy technology with a thermal imaging system to achieve real-time monitoring of the active ingredient content, and establishing a thermal degradation kinetic model of the active ingredient;

[0011] Based on the thermal degradation kinetic model of the active ingredient and the microbial electric field sensitivity data, applying computational intelligence algorithms to simultaneously optimize the electric field parameters and the auxiliary temperature, and combining the multi-objective Pareto optimization method to achieve the best balance between the sterilization effect and the retention rates of different active ingredients.

[0012] In a preferred embodiment, the construction of the pulsed electric field-assisted low-temperature sterilization system includes:

[0013] Designing the electrode configuration, using a parallel plate electrode, a coaxial cylindrical electrode or a finger-shaped electrode structure;

[0014] Constructing a pulsed electric field generating unit, including a high-voltage pulsed power supply and a pulse shaping circuit;

[0015] Designing a temperature control unit, including a circulating water bath, a temperature sensor array and a PID controller;

[0016] Constructing a product flow unit, including a peristaltic pump, a pipeline system and a stirring device;

[0017] Building a control unit based on an embedded system, integrating a data acquisition module, a parameter adjustment module and a safety monitoring module to achieve full-automatic operation of the system.

[0018] In a preferred embodiment, the pulsed electric field generating unit includes a high-voltage pulsed power supply, a multi-stage electrode chamber and a parameter adjustment module;

[0019] The temperature control unit includes a precision temperature control device and a temperature sensor array;

[0020] The product flow unit includes a peristaltic pump, a fluid pipeline and a processing cavity;

[0021] The control unit includes a main controller, a data acquisition module and a human-machine interface;

[0022] Coordinating the work of each unit through the control unit to achieve the coordinated control of the electric field parameters and the temperature.

[0023] In a preferred embodiment, the pulsed electric field-assisted low-temperature sterilization system selectively destroys the microbial cell membrane through the electro-poration effect of the microbial cell membrane.

[0024] In a preferred embodiment, the establishment of the microbial electric field sensitivity database includes:

[0025] Collecting and analyzing the sensitivity data of common microorganisms in pet health products to different electric field parameters, and establishing an electric field sterilization kinetic model, which is expressed as:

[0026] ;

[0027] Among them, represents the electric field sterilization kinetic model, N represents the number of microorganisms after treatment, represents the initial number of microorganisms, k represents the inactivation rate constant of microorganisms, t represents the treatment time, E represents the electric field strength, and m and n are the exponential parameters of time and electric field strength respectively, indicates that the logarithmic ratio of the number of microorganisms decreases with the increase of treatment time and electric field strength.

[0028] In a preferred embodiment, the thermal response characteristic analysis includes:

[0029] Using differential scanning calorimetry and thermogravimetric analysis to determine the thermal stability parameters of the active ingredient, and establishing a relationship model between the thermal degradation rate and temperature:

[0030] ;

[0031] Among them, represents the relationship model between the thermal degradation rate and temperature, is the content of the active ingredient, is the frequency factor, is the activation energy, is the gas constant, is the absolute temperature, represents the dependence of the reaction on the concentration;

[0032] The inverse algorithm of the heat diffusion equation uses the nonlinear least squares method combined with finite element analysis to invert the internal temperature field distribution based on the surface temperature measurement data, and the inverse algorithm of the heat diffusion equation is realized by solving the equation:

[0033] ;

[0034] Among them, represents the inverse algorithm of the heat diffusion equation, is the thermal diffusion coefficient, is the internal heat source term, is the density, is the specific heat capacity, is the material temperature, is the Laplace operator.

[0035] In a preferred embodiment, the real-time monitoring of the active ingredient content by the Raman spectroscopy technology and the thermal imaging system includes:

[0036] Using a portable Raman spectrometer to conduct on-line detection of the active ingredient in pet health products and collect spectral data;

[0037] The quantitative relationship between Raman spectroscopy and the content of active ingredients was established using a multivariate calibration model, and the model expression is:

[0038] ;

[0039] where, is the content of the active ingredient, is the intercept, is the regression coefficient at the th wavelength, is the th Raman spectral intensity at the th wavelength, is the error term,

[0040] The surface temperature field distribution of the product was constructed using an infrared thermal imager;

[0041] The Raman spectroscopy analysis results and the thermal field distribution information were fused to establish a thermal degradation kinetic model of the active ingredient:

[0042] ;

[0043] where, represents the natural logarithm of the active ingredient retention rate, is the initial content of the active ingredient, is the content of the active ingredient, is the frequency factor, is the processing time, is the activation energy, is the gas constant, is the absolute temperature, indicates that the content of the active ingredient decreases with the increase of the processing time;

[0044] According to the thermal degradation kinetic model of the active ingredient, the retention rate of the active ingredient under different process parameters was predicted in real time to provide feedback data for the optimization algorithm.

[0045] In a preferred embodiment, the computational intelligence algorithm simultaneously optimizes the electric field parameters and the auxiliary temperature, including:

[0046] Construct a multi-objective optimization model, and simultaneously consider the following optimization objectives:

[0047] Maximize the microbial killing rate:

[0048] ;

[0049] Maximize the active ingredient retention rate:

[0050] ;

[0051] Minimize energy consumption:

[0052] ;

[0053] wherein, represents maximizing the microbial killing rate, represents maximizing the active ingredient retention rate, represents minimizing energy consumption, represents the logarithmic function, and are the initial and treated microbial counts respectively, and are the contents of the th active ingredient after and before treatment respectively, is the weight coefficient, is the power, is the treatment time, represents the total number of active ingredients;

[0054] An improved non-dominated sorting genetic algorithm is used to solve the multi-objective optimization problem;

[0055] Based on non-dominated sorting and crowding distance calculation, the Pareto optimal solution set is obtained to provide diverse optional solutions for decision-making;

[0056] According to the product characteristics and quality requirements, the most suitable combination of electric field parameters for the current demand is selected from the Pareto front.

[0057] In a preferred embodiment, the optimization method further includes constructing an adaptive parameter optimization control system, and the adaptive parameter optimization control system includes:

[0058] Based on the Pareto optimal solution set, combined with real-time monitoring data, a deep reinforcement learning method is used to realize the dynamic optimization of processing parameters;

[0059] The state space of the control system includes the current product characteristic parameters, the real-time monitored microbial inactivation rate and active ingredient retention rate;

[0060] The action space includes the adjustment amounts of electric field parameters and auxiliary temperature;

[0061] The reward function is defined as a weighted combination of the microbial killing rate, active ingredient retention rate and energy consumption;

[0062] A deep Q-network is used to realize the mapping from state to action;

[0063] Through experience replay and target network technology, the learning stability and efficiency are improved to realize the parameter adaptive optimization of the pet health product processing process.

[0064] In a preferred embodiment, a fully automatic pet health product processing system optimized by AI is used to execute a fully automatic pet health product processing method optimized by AI, including:

[0065] A pulsed electric field generating unit for generating an adjustable high-voltage pulsed electric field, including a high-voltage pulsed power supply, a multi-stage electrode chamber, and a parameter adjustment module;

[0066] A temperature control unit for precisely adjusting and maintaining the processing temperature, including a precision temperature control device and a temperature sensor array;

[0067] A product flow unit for controlling the flow state of pet health products in the processing cavity, including a peristaltic pump, a fluid pipeline, and a processing cavity;

[0068] A monitoring and analysis unit, including a Raman spectrometer and an infrared thermal imager, for real-time monitoring of the active ingredient content and temperature field distribution;

[0069] A control unit, including a main controller, a data acquisition module, and a human-machine interface, integrating a deep reinforcement learning algorithm to achieve adaptive optimization control of system parameters.

[0070] The beneficial effects of the present invention are as follows:

[0071] Improve the retention rate of heat-sensitive active ingredients: Use pulsed electric field-assisted low-temperature sterilization to replace traditional thermal sterilization, reduce the processing temperature, and while ensuring that the microbial kill rate reaches 5 logarithmic levels, improve the retention rate of heat-sensitive active ingredients.

[0072] Realize differential protection of multiple heat-sensitive components: By constructing a kinetic model of active ingredient thermal degradation and a multi-objective optimization algorithm, solve the technical problem of simultaneously preserving multiple heat-sensitive components, and reduce the fluctuation of active ingredient content between product batches.

[0073] Improve processing efficiency and energy utilization rate: Based on the technical route combining real-time monitoring and precise control, realize the dynamic optimization of processing parameters. Compared with the traditional thermal processing technology, it reduces energy consumption and shortens the processing time.

[0074] Intelligent and precise processing during the process: Through the adaptive parameter optimization control system constructed by introducing the deep reinforcement learning method, realize the precise processing technology with the goal of "ingredient retention", transform the pet health product processing from the traditional "sterilization-centered" to "active ingredient preservation-centered", and improve the effectiveness and quality stability of the product. Brief Description of the Drawings

[0075] Figure 1 is a flow chart of a fully automatic pet health product processing method optimized by AI of the present invention. Detailed Embodiments

[0076] Reference will now be made to exemplary embodiments to discuss the subject matter described herein. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein. Without departing from the scope of protection of the content of this specification, changes can be made to the functions and arrangements of the elements discussed. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described in some examples can also be combined in other examples.

[0077] In at least one embodiment of the present invention, a fully automatic pet health product processing method optimized based on AI is disclosed. As Figure 1 shown, it includes the following steps:

[0078] Step 1, construct a pulsed electric field-assisted low-temperature sterilization system, including a pulsed electric field generation unit, a temperature control unit, a product flow unit, and a control unit;

[0079] Among them, the pulsed electric field generation unit includes a high-voltage pulsed power supply, a multi-stage electrode chamber, and a parameter adjustment module; the temperature control unit includes a precision temperature control device and a temperature sensor array; the product flow unit includes a peristaltic pump, a fluid pipeline, and a processing cavity; the control unit includes a main controller, a data acquisition module, and a human-machine interface.

[0080] The pulsed electric field generation unit may also include an impedance matching circuit for improving the energy transmission efficiency;

[0081] In some embodiments, the temperature control unit can adopt a water circulation jacket or a microwave-assisted heating method, which is selected according to the thermal sensitivity characteristics of different products; for example, for high-viscosity paste-like pet health products, a spiral extrusion type product flow unit can be used to replace the peristaltic pump to improve the flow uniformity.

[0082] In this step, the microbial cell membrane electroporation effect under the action of pulsed electric field is the core mechanism for realizing low-temperature sterilization. The electroporation process can be represented by the following mathematical model:

[0083] ;

[0084] Among them, represents the membrane perforation density per unit area; represents the change rate of the perforation density with time; represents the applied electric field strength; represents the activation energy of cell membrane electroporation; represents the gas constant; represents the absolute temperature; represents the rate constant; represents the electric field dependence index; represents the maximum possible perforation density; It represents the membrane perforation density per unit area; Represents an exponential function.

[0085] It includes the following sub-steps:

[0086] Step 1.1, determine the optimal electric field parameter combination;

[0087] According to one embodiment of the present application, by analyzing the electroporation characteristics of common microorganisms in pet health products, it is determined that the electric field strength range is 15-25 kV / cm, the pulse width is 1-5 μs, the frequency is 100-500 Hz, and the number of pulses is 20-50.

[0088] For certain types of pet health products, such as products containing high concentrations of fat, the electric field strength can be appropriately increased to 30kV / cm; while for products containing beneficial microorganisms such as probiotics, a lower electric field strength (10-15kV / cm) and a higher auxiliary temperature (50-55°C) can be used to achieve selective killing of harmful microorganisms.

[0089] Step 1.2, configuring a multi-level electrode structure;

[0090] A multi-level electrode structure combining parallel plates and coaxial columns can be used to ensure uniformity of electric field distribution. Medical grade titanium alloy is used as the electrode material, and the electrode spacing can be adjusted within the range of 5-15mm to adapt to different product characteristics.

[0091] In some embodiments, the electrode surface may be coated with a conductive polymer to reduce the interfacial reaction between the electrode and the product; for example, for acidic pet health products, platinum-iridium alloy electrodes may be used instead of titanium alloys to improve corrosion resistance; the electrode structure may also use an array of staggered rod electrodes, which is suitable for processing granular or block pet health products.

[0092] Step 1.3, constructing an electric field-temperature coordinated control system;

[0093] In this embodiment, the electric field parameters are integrated with the auxiliary temperature control system to form a synergistic control model to maintain the temperature in the range of 40-50°C during the treatment process, which is lower than the 70-90°C required for traditional thermal sterilization, thereby reducing the loss of heat-sensitive components.

[0094] For particularly heat-sensitive products, a pulsed temperature control strategy can be used, that is, a short-term temperature increase during the application of the electric field and rapid cooling after the electric field is removed to minimize the overall thermal history of the product.

[0095] In some embodiments, a flow field control module may be integrated into the temperature control system to further improve processing uniformity through swirl or directional shear flow.

[0096] Step 2: Establish a database of microbial electro-sensitivity, and achieve precise customization of electric field parameters according to the electro-sensitivity characteristics of common microorganisms in pet health products;

[0097] Specifically, it includes the following sub-steps:

[0098] Step 2.1: Establish a microbial electro-sensitivity model;

[0099] Collect and analyze the sensitivity data of common microorganisms in pet health products to different electric field parameters, and establish an electro-kinetic model for microbial sterilization for various microorganisms:

[0100] ;

[0101] Among them, represents the electro-kinetic model for microbial sterilization, represents the number of processed microorganisms; represents the initial number of microorganisms; represents the microbial inactivation rate constant; represents the processing time; represents the electric field strength; represents the time exponent parameter; represents the electric field strength exponent parameter; indicates that the logarithmic ratio of the microbial quantity decreases with the increase of the processing time and the electric field strength.

[0102] For microorganisms in different physiological states, an environmental factor correction term can be introduced, and the corrected model is:

[0103] ;

[0104] Among them, represents the corrected electro-kinetic model for microbial sterilization, represents the microbial inactivation rate constant; represents the processing time; represents the electric field strength; represents the time exponent parameter; represents the electric field strength exponent parameter; represents the correction coefficient of environmental conditions on the electro-sterilization effect; represents the pH value of the processing environment; represents the water activity; represents the processing temperature; indicates that the logarithmic ratio of the microbial quantity decreases with the increase of the processing time and the electric field strength.

[0105] For example, under acidic conditions, the sensitivity of most pathogenic bacteria to the electric field increases, and the same sterilization effect can be achieved by reducing the electric field strength, thereby further reducing the impact on the active ingredients.

[0106] The implementation methods of the electric field sterilization kinetic model specifically include:

[0107] Inoculate the target microbial strain in a suitable culture medium and culture it until the logarithmic growth phase;

[0108] Aliquot the bacterial solution into a specially designed electroporation sample cell and apply pulsed electric fields with different intensities (15 - 25 kV / cm), different pulse widths (1 - 5 μs), and different frequencies (100 - 500 Hz) for treatment;

[0109] Immediately perform viable cell counting after treatment and record the microbial survival rates under different treatment conditions;

[0110] Determine the model parameters through non - linear regression analysis 、 and values.

[0111] In practical applications, taking Salmonella, which is common in pet health products, as an example, the model parameters determined by the above method are: , , . According to these parameters, it can be calculated that under the conditions of an electric field strength of 20 kV / cm, a pulse width of 3 μs, and 30 pulse numbers, the killing rate of Salmonella can reach 4.5 logarithmic levels (i.e., 99.997%).

[0112] For different microbial strains, the model parameters are different. The database established in this application contains the complete parameter sets of 20 common microorganisms, providing basic data support for subsequent parameter optimization and control.

[0113] In some embodiments, this database can be connected to a cloud data center to achieve data sharing and remote update, enabling the system to cope with the challenges of new pathogenic bacteria; for example, by introducing machine learning methods, the system can predict the electric field sensitivity parameters based on the basic characteristics of new strains and the electric field response data of existing strains, thereby quickly forming a preliminary treatment plan.

[0114] Step 2.2, optimize the electrode structure to improve the treatment uniformity;

[0115] Based on the finite - element analysis method, simulate the electric field distribution under different electrode geometries and arrangements, and construct an optimized electrode structure that can achieve uniform electric field distribution according to the rheological properties of pet health products. The optimized electrode structure can control the coefficient of variation of the electric field strength within 5% in the treatment area, ensuring the consistency of the sterilization effect.

[0116] The specific implementation methods for optimizing the electrode structure of this application include:

[0117] Use COMSOL Multiphysics software to establish a three-dimensional electric field distribution model, and parametrically describe the electrode chamber, processing chamber, and product fluid characteristics;

[0118] Set multiple electrode structure parameters as variables and run the electric field distribution simulation;

[0119] By calculating the spatial distribution of the electric field intensity in the processing chamber, evaluate the electric field uniformity under different electrode configurations.

[0120] For different product forms, different electrode structures can be adopted:

[0121] For example, for liquid pet health products with good fluidity, a continuous-flow tubular electrode can be used;

[0122] For viscous gel-like products, a pulsating-flow flat electrode can be used;

[0123] For granular products, an electrode system that combines a vibrating bed and an electric field can be used.

[0124] In some embodiments, the electrode surface can also be nano-treated to reduce the formation of bubbles on the electrode surface and improve the uniformity of the electric field distribution.

[0125] In practical applications, an "interleaved coaxial double-ring electrode structure" is optimized through the above method. The structure specifically includes: a titanium alloy ring electrode with an inner ring diameter of 45 mm and an outer ring diameter of 80 mm. The two rings are fixed by an insulating bracket, and the axial spacing between adjacent electrode pairs is 12 mm, and the polarities of adjacent electrode pairs are arranged alternately.

[0126] Step 2.3, integrate bioimpedance analysis technology for real-time monitoring;

[0127] Integrate a bioimpedance analysis module in the processing system, and measure the change in impedance parameters of the product before and after processing to monitor the microbial inactivation rate in real time.

[0128] The impedance measurement uses the four-electrode method, and the frequency range is 1 kHz - 1 MHz. A sterilization effect prediction model is established based on the impedance spectrum change:

[0129] ;

[0130] Among them, represents the impedance value after processing; represents the initial impedance value; is the proportionality coefficient; Represents the number of microorganisms before treatment; Represents the number of microorganisms after treatment; Represents the logarithmic reduction value of microorganisms.

[0131] Through the sterilization effect prediction model, the sterilization effect can be evaluated in real time without interrupting production, and the treatment parameters can be dynamically adjusted according to the evaluation results.

[0132] The specific implementation methods of bioimpedance analysis technology include:

[0133] Install four-electrode impedance measurement probes at the inlet and outlet of the treatment chamber, apply a weak alternating current with two outer electrodes, and measure the voltage response with two inner electrodes;

[0134] Use an impedance analyzer to scan and measure the impedance at different frequencies to obtain an impedance spectrum;

[0135] By extracting the impedance value changes at characteristic frequency points and combining with the pre-established relationship model between the number of microorganisms and impedance, calculate the microorganism inactivation rate.

[0136] In the bioimpedance analysis system, temperature and conductivity compensation algorithms can be introduced to eliminate the influence of non-biological factors on impedance measurement;

[0137] For example, by measuring the impedance change of the reference channel, the baseline drift caused by temperature increase can be excluded.

[0138] In some embodiments, impedance analysis can also be combined with other on-line detection technologies to construct a multi-parameter fusion microorganism activity evaluation model to further improve the monitoring accuracy.

[0139] In practical applications, by real-time monitoring of impedance changes, fluctuations in the treatment effect can be detected within 2 seconds. For example, when the impedance change rate of a certain batch of products is detected to be lower than the preset threshold (indicating insufficient sterilization effect), the system will automatically increase the electric field strength or extend the treatment time to ensure that the expected sterilization effect is always achieved, while avoiding energy waste and loss of active ingredients caused by over-treatment.

[0140] Step 3, based on the microorganism electric field sensitivity database, analyze the thermal response characteristics of different active ingredients in pet health products, construct an inverse algorithm for the heat diffusion equation, and achieve accurate reconstruction of the internal temperature field of the material;

[0141] Specifically, it includes the following sub-steps:

[0142] Step 3.1, construct a pulsed infrared thermal wave analysis system;

[0143] The pulsed infrared thermal wave analysis system includes a pulsed heat source, an infrared thermal imager, a data acquisition unit, and an analysis and processing unit.

[0144] By applying a pulsed heat source to the sample and using a high-precision infrared thermal imager (temperature resolution ≤ 0.05 °C, spatial resolution ≤ 100 μm), record the change in the surface temperature of the sample over time, with a sampling frequency of 50 - 200 Hz.

[0145] The pulsed heat source can adopt different excitation methods. For example, for liquid products, an infrared laser pulse can be used, and for solid or semi-solid products, a flash lamp or electromagnetic induction heating can be used;

[0146] In some embodiments, the thermal imager can be optionally equipped with detectors with different wavelength ranges to adapt to the detection requirements of different temperature ranges;

[0147] For example, for products that need to accurately monitor the temperature range of 40 - 60 °C, a 3 - 5 μm mid-wave infrared detector can be selected to improve the detection accuracy in the temperature range.

[0148] Step 3.2, establish an inverse algorithm for the heat diffusion equation;

[0149] Based on the heat diffusion equation:

[0150] ;

[0151] where, represents the heat diffusion equation, represents the material temperature; represents time; represents the thermal diffusivity; represents the Laplace operator; represents the heat source function; represents the material density; represents the specific heat capacity.

[0152] Apply the inverse heat conduction algorithm to inversely deduce the internal temperature distribution from the time evolution of the surface temperature field.

[0153] The specific implementation includes:

[0154] Establish a forward heat conduction model to simulate the temperature field distribution under given thermal physical parameters and thermal boundary conditions;

[0155] Construct an optimization objective function:

[0156] ;

[0157] where, represents the optimization objective function, represents the set of thermal physical parameters to be inverted, represents the measured temperature value, represents the calculated temperature value, and Respectively represent the position and time of the measurement point;

[0158] The Levenberg-Marquardt algorithm is used to solve the parameter set that minimizes the objective function and realize the reconstruction of the internal temperature field. The Levenberg-Marquardt algorithm is an optimization method for solving nonlinear least squares problems, which combines the advantages of the gradient descent method and the Gauss-Newton method.

[0159] The specific implementation of the heat diffusion equation inversion algorithm is as follows:

[0160] The heat diffusion equation is discretized by finite difference and the Crank-Nicolson implicit scheme is used to ensure numerical stability.

[0161] Establish a sensitivity matrix between surface temperature and internal temperature to indicate the influence of internal temperature change on surface temperature measurement value;

[0162] Based on the measured surface temperature time series data, the optimal thermophysical parameters and internal heat source distribution are found through iteration;

[0163] The forward heat conduction calculation is performed again using the optimized parameters to obtain the temperature distribution in the entire spatial domain.

[0164] To handle products with complex geometries, the inversion algorithm can use the finite element method instead of the finite difference method to improve spatial resolution and boundary processing capabilities;

[0165] For example, for irregularly shaped pet chews, tetrahedral meshing can more accurately describe the internal temperature distribution.

[0166] In some embodiments, a Bayesian inference framework may be introduced to consider measurement noise and model uncertainty, provide a confidence interval for temperature field reconstruction, and make control decisions more reliable.

[0167] For real-time control situations with high requirements for computing efficiency, a simplified inversion model can be selected. For example, the characteristic time method can be used to quickly estimate the effective thermal diffusion coefficient of the product, calculate the internal temperature distribution based on the analytical solution, and achieve millisecond-level temperature field reconstruction.

[0168] The heat diffusion equation inversion algorithm was used to analyze the internal temperature distribution of compound pet health products containing multiple vitamins during electric field treatment. By applying a 20ms infrared pulse heat source on the product surface and recording the surface temperature response curve within 400ms, the three-dimensional temperature field at a depth of 1mm to 5mm inside the product was successfully reconstructed using the above inversion algorithm, with a temperature reconstruction accuracy of ±0.2℃.

[0169] Step 3.3, establish product thermal characteristics database;

[0170] For different pet health care product formulations, measure thermal property parameters such as their thermal diffusivity, specific heat capacity, and thermal conductivity, and establish a product thermal property database.

[0171] By combining the product thermal property database with an inversion algorithm, it is possible to predict the internal temperature field during the processing of any new formulation product with an accuracy of ±0.5°C.

[0172] The specific implementation method of the product thermal property database in this application includes:

[0173] Use the Laser Flash Method (LFM) to measure the thermal diffusivity of pet health care product samples with different component ratios and water contents. The sample thickness is 2 mm, and the measurement temperature range is 10 - 80°C;

[0174] Use Differential Scanning Calorimetry (DSC) to measure the specific heat capacity of the samples;

[0175] Calculate the thermal conductivity of the samples based on the thermal diffusivity, specific heat capacity, and density;

[0176] Establish a multiple correlation model between the thermal property parameters and the component ratio, water content, and temperature of the samples to form a parameterized thermal property database.

[0177] To improve the accuracy and scope of application of the database, an Artificial Neural Network (ANN) can be used to replace the traditional multiple regression model to better capture the non - linear interaction between components;

[0178] For example, use a three - layer feed - forward neural network with protein, fat, carbohydrate, moisture content, and temperature as input features to predict the thermal property parameters.

[0179] In some embodiments, the thermal property database can also be adaptively updated according to product batch changes, continuously optimizing the model parameters through online measured thermal response data to ensure the long - term stability of the prediction accuracy.

[0180] Step 4, utilize the inversion result of the thermal diffusion equation, combine Raman spectroscopy technology and a thermal imaging system to achieve real - time monitoring of the active ingredient content, and establish a thermal degradation kinetics model for the active ingredient;

[0181] Specifically, it includes the following sub - steps:

[0182] Step 4.1, construct an active ingredient detection system based on Raman spectroscopy;

[0183] The active ingredient detection system based on Raman spectroscopy uses a Raman spectrometer with a laser wavelength of 785 nm and a spectral resolution of ≤2 cm -1 , combined with an optical fiber probe to achieve on-line detection.

[0184] For the key active ingredients in pet health products (such as proteins, polypeptides, vitamins, amino acids, etc.), a Raman characteristic peak database is established, and a quantitative relationship between the content of active ingredients and the Raman spectral intensity is established through multivariate analysis methods:

[0185] ;

[0186] where represents the content of the th active ingredient, represents the Raman spectral intensity at the wavenumber of , represents the regression coefficient, represents the intercept, represents the number of characteristic wavenumber points used for modeling, represents the index of the wavenumber point.

[0187] For different types of pet health products, different spectral preprocessing methods can be used to improve the prediction accuracy; for example,

[0188] for products containing fat, the extended multiplicative scatter correction method can be used to eliminate the fluorescence interference of fat;

[0189] for products with a darker color, the standard normal variate transformation method can be used to eliminate the scattering effect.

[0190] In some embodiments, a variable screening algorithm can also be used to select the most representative spectral variables, reduce redundant information, and improve the robustness of the model.

[0191] The specific implementation of the Raman spectroscopy active ingredient detection system includes:

[0192] Select representative pet health product samples covering different content ranges of the target active ingredients, and use high performance liquid chromatography to accurately determine the content of each active ingredient in the samples as calibration standards;

[0193] Use a 785 nm laser Raman spectrometer to scan the samples to obtain Raman spectra in the range of 400 - 3000 cm -1 ;

[0194] Preprocess the spectral data, including operations such as baseline correction, smoothing, and normalization;

[0195] Use the partial least squares regression algorithm to establish a quantitative relationship model between the spectral intensity at the characteristic wavenumber and the content of the active ingredient.

[0196] In specific applications, this system focuses on several heat-sensitive active ingredients in pet health products: vitamin C (characteristic peak at 1690 cm -1 ), collagen polypeptide (characteristic peak at 1242 - 1272 cm -1 ), probiotics (characteristic peak at 1002 cm -1 ), and trace element chelates (characteristic peak at 550 - 600 cm -1 ).

[0197] By installing the fiber optic probe at the observation window of the processing chamber, the system can collect the Raman spectral signals of the product in real time during the electric field treatment, complete a scan and analysis every 2 seconds, and calculate the current content of each active ingredient.

[0198] Step 4.2, establish the thermal degradation kinetic model of the active ingredient;

[0199] Conduct thermal degradation experiments on various heat-sensitive active ingredients in pet health products under different temperature conditions (40 - 90 °C), measure the change of the active ingredient content over time, and establish the thermal degradation kinetic model:

[0200] ;

[0201] Among them, represents the change rate of the active ingredient concentration C with time t, represents the active ingredient content, represents time, represents the frequency factor, represents the activation energy of the thermal degradation reaction, represents the gas constant, represents the absolute temperature, represents the dependence of the reaction on the concentration, represents the exponential function.

[0202] For the degradation characteristics of different active ingredients, by solving the above equation, the retention rate of the active ingredient under any temperature-time combination can be obtained:

[0203] ;

[0204] Among them, represents the active ingredient content, represents the initial active ingredient concentration, represents the retention rate of the active ingredient, represents the frequency factor, represents the activation energy of the thermal degradation reaction, represents the gas constant, represents the absolute temperature, Indicates the duration of the thermal degradation process Indicates the effect of temperature on the reaction rate, which conforms to the Arrhenius equation

[0205] For active ingredients with some complex degradation behaviors, a multi-stage degradation model can be used to more accurately describe their thermal degradation process; for example, the degradation of vitamin B12 in aqueous solution can be represented as a two-stage process: an initial rapid degradation stage and a subsequent slow degradation stage, each stage having different kinetic parameters

[0206] In some embodiments, the influence of environmental factors such as pH value, ionic strength, and redox potential on the stability of active ingredients can also be considered to establish an extended thermal degradation model

[0207] ;

[0208] Wherein Represents the change rate of the active ingredient concentration with time Represents the content of the active ingredient Indicates the duration of the thermal degradation process Represents the frequency factor Represents the activation energy of the thermal degradation reaction Represents the gas constant Represents the absolute temperature Represents the dependence of the reaction on the concentration Is the environmental factor correction function Represents the acidity or alkalinity of the solution Represents the ionic strength Represents the redox potential Represents the exponential function

[0209] The specific implementation methods of the thermal degradation kinetic model include

[0210] Extracting the target active ingredient in the pet health product and preparing it into a solution or suspension

[0211] Sub-packaging the sample into a sealed reaction tube and placing it in a precision constant temperature water bath, and incubating it at six temperature points of 40 °C, 50 °C, 60 °C, 70 °C, 80 °C, and 90 °C respectively

[0212] Sampling at predetermined time intervals and measuring the content of the active ingredient in the sample using liquid chromatography or enzyme-linked immunosorbent assay (ELISA)

[0213] Substituting the data of the measured active ingredient content changing with time into the thermal degradation kinetic equation, and determining the kinetic parameters of each active ingredient through non-linear fitting

[0214] The thermal degradation kinetic parameters of several key active ingredients in a certain pet nutritional supplement were measured. The results showed that the activation energy of vitamin C was 68.5 kJ / mol, the frequency factor was 5.6×10 6 min -1 , and the reaction order was 1; the activation energy of taurine was 75.3 kJ / mol, the frequency factor was 2.3×10 7 min -1 , and the reaction order was 1.2; the activation energy of probiotic activity was 82.7 kJ / mol, the frequency factor was 4.5×10 8 min -1 , and the reaction order was 1.5.

[0215] Based on these parameters, the system can accurately predict the retention rates of each active ingredient under different temperature-time combinations;

[0216] Under the low-temperature electric field-assisted sterilization conditions (45°C, 2 minutes) of this application, the retention rate of vitamin C can reach 92%, the retention rate of taurine reaches 96%, and the retention rate of probiotics reaches 85%. This accurate prediction ability provides a theoretical basis for optimizing the processing parameters of the system and maximizes the retention rate of active ingredients.

[0217] Step 4.3, integrate thermal imaging and Raman spectroscopy data to achieve dynamic monitoring of active ingredients;

[0218] Combine the temperature field reconstruction results with the thermal degradation model to calculate the spatial distribution and content changes of active ingredients in the product in real time. The system completes a full-field scan and calculation every 5 seconds, and the monitoring accuracy reaches ±3%, providing real-time feedback for optimizing subsequent processing parameters.

[0219] For pet health products with complex formulations, a data fusion algorithm can be used to integrate multi-source information; for example, the Kalman filtering technique is used to combine the direct measurement values of Raman spectroscopy and the predicted values based on the temperature field to generate a more accurate estimate of the active ingredient distribution. In some embodiments, edge computing technology can also be introduced to transfer some data processing tasks to the acquisition end, reduce communication delays, and improve the system response speed; for example, a small processor is integrated into the Raman spectroscopy acquisition module to complete data preprocessing and feature extraction, and only the key feature values are transmitted to the central control system.

[0220] The specific implementation methods of the active ingredient dynamic monitoring system in this application include:

[0221] Import the reconstructed three-dimensional temperature field data and the parameters of the established active ingredient thermal degradation kinetic model into the real-time calculation module;

[0222] The real-time calculation module calculates the current content of the active ingredient at each spatial point based on the instantaneous temperature and cumulative heat history of each spatial point, and generates a spatial distribution map of the active ingredient inside the product;

[0223] The Raman spectroscopy system measures the content of the active ingredient in real time at several key points on the product surface, and these measured values are used to correct and verify the temperature field calculation results;

[0224] The system forms a more accurate active ingredient distribution model through the fusion of these two types of data and updates it in real time.

[0225] The active ingredient dynamic monitoring system is applied to the production process of canine liquid compound vitamin health products. When the health product flows through the electric field treatment area, the system calculates the concentration distribution of various water-soluble vitamins such as vitamin B1, B2, B6, and C inside the product in real time. Through the three-dimensional visualization interface, the operator can intuitively see which areas inside the product have more loss of active ingredients (usually shown as red areas).

[0226] Step 5, based on the active ingredient thermal degradation kinetic model and microbial electric field sensitivity data, apply computational intelligence algorithms to optimize the electric field parameters and auxiliary temperature simultaneously, and combine the multi-objective Pareto optimization method to achieve the best balance between the sterilization effect and the retention rate of different active ingredients;

[0227] Specifically, it includes the following sub-steps:

[0228] Step 5.1, construct a multi-objective optimization model;

[0229] In the embodiment of the present application, for the pet health product processing process, the following optimization objectives are defined:

[0230] Maximize the microbial kill rate:

[0231] ;

[0232] Maximize the active ingredient retention rate:

[0233] ;

[0234] Minimize energy consumption:

[0235] ;

[0236] Among them, represents maximizing the microbial kill rate, represents maximizing the active ingredient retention rate, represents minimizing energy consumption, represents the logarithmic function, and are the initial and treated microbial numbers respectively, and are the contents of the th active ingredient after and before treatment, is the weight coefficient, is the power, is the treatment time, represents the total number of active ingredients.

[0237] The decision variables include:

[0238] Electric field strength (15 - 25 kV / cm) represents the magnitude of the voltage applied per unit distance;

[0239] Pulse width (1 - 5 μs) represents the duration of each electrical pulse;

[0240] Pulse frequency (100 - 500 Hz) represents the number of pulses generated per second;

[0241] Number of pulses (20 - 50) represents the total number of pulses applied during the treatment;

[0242] Auxiliary temperature (40 - 55 °C) represents the ambient temperature maintained during the treatment.

[0243] According to the characteristics and quality requirements of different types of pet health products, the optimization objectives can be adjusted appropriately; for example, for high-end pet functional health products, the weight of the active ingredient retention rate can be increased, and even a fourth optimization objective: active ingredient uniformity, can be introduced to ensure the consistency of the active ingredient content in each part of the product.

[0244] In some embodiments, the objective function can also consider the sensory characteristics of the product; for example, for some non-enzymatic browning reactions that affect the color or flavor of the product, the reaction degree can be incorporated into the optimization objective to minimize the quality loss caused by adverse reactions.

[0245] The specific implementation methods of the multi-objective optimization model include:

[0246] Establish a relationship model between the decision variables and the objective function, specifically including: the relationship model between the electric field parameters and the microbial killing rate, the relationship model between the temperature and the active ingredient retention rate, and the relationship model between the treatment parameters and the energy consumption;

[0247] Normalize each objective function so that its value range is unified between 0 - 1;

[0248] Define the comprehensive objective function in the form of a weighted sum of each sub-objective function:

[0249] ;

[0250] Among them, represents maximizing the microbial killing rate, represents maximizing the active ingredient retention rate, represents minimizing the energy consumption, , , are the weight coefficients for maximizing the microbial killing rate, maximizing the active ingredient retention rate, and minimizing the energy consumption, respectively.

[0251] In specific applications, for a pet liquid nutritional supplement containing multiple B vitamins and probiotics, the system automatically sets the weight coefficients according to the product characteristics: (sterilization effect weight), (active ingredient retention weight), (energy consumption weight).

[0252] The system also automatically adjusts the parameters in the sub-objective function according to the characteristics of the main target pathogens (Escherichia coli and Salmonella) and key active ingredients (vitamin B12 and Bifidobacterium) of the product.

[0253] The optimized electric field treatment parameters are: electric field strength 22 kV / cm, pulse width 2.5 μs, pulse frequency 300 Hz, pulse number 35, and auxiliary temperature 45 °C. While achieving a 5-log reduction in the sterilization effect, this set of parameters ensures that the retention rate of vitamin B12 reaches 91% and the active retention rate of Bifidobacterium reaches 85%, which are 30% and 70% higher than those of the traditional thermal sterilization process, respectively.

[0254] Step 5.2, implement multi-objective Pareto optimization;

[0255] Use the improved Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to solve the above multi-objective optimization problem. The steps include:

[0256] Initialize the population: randomly generate parameter combinations that meet the constraints;

[0257] Evaluate the individual fitness: use the models established in Steps 2 to 4 to calculate the objective function values corresponding to each set of parameters;

[0258] Non-dominated sorting: sort the individuals according to the objective function values to form the Pareto front;

[0259] Selection, crossover, and mutation: generate a new generation of population;

[0260] Repeat the above steps until the termination condition is met.

[0261] In this embodiment, the algorithm parameters are set as follows: the population size is 50, the maximum number of iterations is 100, the crossover probability is 0.8, and the mutation probability is 0.1.

[0262] For real-time optimization scenarios, an optimization algorithm with higher computational efficiency can be adopted;

[0263] For example, use the multi-objective particle swarm optimization algorithm or the multi-objective differential evolution algorithm. These algorithms have a faster convergence speed when dealing with high-dimensional decision spaces and are suitable for online optimization applications.

[0264] In some embodiments, constraint handling techniques can also be introduced to better handle various actual constraint conditions in the production of pet health products; for example, adopt an improved constraint domination principle to handle hard constraints such as the maximum temperature not exceeding a specific threshold and the minimum sterilization rate must meet food safety standards.

[0265] The specific implementation of the improved NSGA-II algorithm in this application includes:

[0266] Use real number coding to represent decision variables, and design special crossover and mutation operators according to the characteristics of electric field parameters;

[0267] Introduce a selection mechanism based on crowding distance to ensure the diversity of solutions on the Pareto front;

[0268] Design an adaptive mutation rate strategy, use a larger mutation rate in the early stage of iteration to increase the search range, and reduce the mutation rate in the later stage to improve the convergence speed;

[0269] Adopt an elitist retention strategy to ensure that the optimal individuals will not be lost during the evolution process.

[0270] In practical applications, the non-dominated sorting genetic algorithm has optimized the parameters for different types of pet health products. For example, for heat-sensitive liquid probiotic products, the algorithm pays more attention to the retention rate of active ingredients, and the obtained optimal parameter combination is low temperature (42 °C), high electric field strength (24 kV / cm), and short treatment time (60 s);

[0271] For the special nutrient solution for infants and young puppies that requires higher sterilization requirements, the algorithm pays more attention to the sterilization effect, and the obtained optimal parameter combination is medium temperature (48 °C), medium electric field strength (20 kV / cm), and longer treatment time (120 s). Different solutions on the Pareto front can meet the specific needs of different products and provide flexible choices for production.

[0272] Step 5.3, construct an adaptive parameter optimization control system;

[0273] Based on the Pareto optimal solution set, combined with real-time monitoring data, an adaptive control system is constructed using deep reinforcement learning methods to achieve dynamic optimization of processing parameters.

[0274] The state space of the control system includes: current product characteristic parameters, real-time monitored microbial inactivation rate and active ingredient retention rate;

[0275] The action space includes: adjustment amounts of electric field parameters and auxiliary temperature;

[0276] The reward function is defined as: a weighted combination of microbial killing rate, active ingredient retention rate, and energy consumption.

[0277] For different control strategy requirements, different types of deep reinforcement learning algorithms can be selected; for example, for scenarios pursuing stability, a conservative policy gradient algorithm can be used to limit the step size of each policy update to avoid drastic fluctuations in control parameters; for scenarios that need to quickly adapt to product formula changes, a model-based reinforcement learning method can be used to accelerate the policy learning process by constructing an environmental dynamics model. In some embodiments, transfer learning techniques can also be introduced to transfer the policies trained on a certain type of pet health product to new products, significantly reducing the policy optimization time for new products.

[0278] The deep Q-network is used to implement the mapping from state to action, and the network structure is:

[0279] Input layer (state dimension) → fully connected layer with 256 neurons → fully connected layer with 128 neurons → fully connected layer with 64 neurons → output layer (action dimension). Through experience replay and target network techniques, the learning stability and efficiency are improved.

[0280] The specific implementation methods of the deep Q-network include:

[0281] Define the state space to include 12 features (including physical characteristic parameters such as product temperature, conductivity, pH value, flow rate, etc., as well as the real-time monitored microbial quantity, contents of 4 key active ingredients, current electric field parameters, and cumulative energy consumption);

[0282] Discretize the action space, and define 5 adjustment levels (substantially increase, slightly increase, remain unchanged, slightly decrease, substantially decrease) for electric field strength, pulse width, frequency, and auxiliary temperature respectively;

[0283] Design the reward function as:

[0284] ;

[0285] Among them, 、 、 respectively represent the importance weights of bactericidal effect, active ingredient retention, and energy consumption in the total reward; represents the number of microorganisms before treatment; represents the number of microorganisms after treatment; represents the logarithm of the reduction in microorganisms; represents the content of the represents the content of the represents the retention rate of the represents the weight coefficient of the represents the total number of active ingredients; represents the system power; represents the treatment time; represents the total energy consumption of the treatment process;

[0286] By interacting with the simulation environment, state transition samples are collected using an experience replay buffer, and the Q-value estimate is updated through temporal difference learning.

[0287] In practical applications, the deep Q-network control system was tested on the production line of a certain pet health product factory.

[0288] The system was pre-trained for 1 million steps in the simulation environment and then put into use and continuously optimized in the actual production environment. Compared with the fixed parameter control strategy, the deep Q-network control system can dynamically adjust the processing parameters according to the product characteristics and processing effects detected in real time.

[0289] The above describes the embodiments of the present invention, but these embodiments are not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative and not restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make more equivalent embodiments in various forms, all of which fall within the protection scope of this embodiment.

Claims

1. A fully automatic processing method for pet health products optimized by AI, characterized in that, The steps include: Construct a pulsed electric field assisted low-temperature sterilization system, including a pulsed electric field generation unit, a temperature control unit, a product flow unit, and a control unit; Establish a microbial electric field sensitivity database, aiming at the electric field sensitivity characteristics of common microorganisms in pet health products, and realizing the precise customization of electric field parameters; Based on the microbial electric field sensitivity database, analyze the thermal response characteristics of different active ingredients in pet health products, construct an inverse algorithm for the heat diffusion equation, and realize the accurate reconstruction of the internal temperature field of the material; Utilize the inversion results of the heat diffusion equation, combine Raman spectroscopy technology and a thermal imaging system to realize the real-time monitoring of the content of active ingredients, and establish a thermal degradation kinetic model of active ingredients; Based on the thermal degradation kinetic model of active ingredients and the microbial electric field sensitivity data, apply computational intelligence algorithms to simultaneously optimize the electric field parameters and the auxiliary temperature, and combine the multi-objective Pareto optimization method to achieve the best balance between the sterilization effect and the retention rate of different active ingredients.

2. The fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that, The construction of the pulsed electric field assisted low-temperature sterilization system includes: Design the electrode configuration, adopting a parallel plate electrode, a coaxial cylindrical electrode, or a finger-shaped electrode structure; Construct a pulsed electric field generation unit, including a high-voltage pulsed power supply and a pulse shaping circuit; Design a temperature control unit, including a circulating water bath, a temperature sensor array, and a PID controller; Construct a product flow unit, including a peristaltic pump, a pipeline system, and a stirring device; Build a control unit based on an embedded system, integrating a data acquisition module, a parameter adjustment module, and a safety monitoring module, to realize the full-automatic operation of the system.

3. A fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that, The pulsed electric field generation unit includes a high-voltage pulsed power supply, a multi-stage electrode chamber, and a parameter adjustment module; The temperature control unit includes a precision temperature control device and a temperature sensor array; The product flow unit includes a peristaltic pump, a fluid pipeline, and a processing cavity; The control unit includes a main controller, a data acquisition module, and a human-machine interface; Coordinate the work of each unit through the control unit to realize the coordinated control of electric field parameters and temperature.

4. A fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that The pulsed electric field assisted low-temperature sterilization system selectively destroys the microbial cell membrane through the electropermeabilization effect of the microbial cell membrane.

5. A fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that, The establishment of the microbial electric field sensitivity database includes: Collect and analyze the sensitivity data of common microorganisms in pet health products to different electric field parameters, and establish an electric field sterilization kinetic model, which is expressed as: ； Among them, represents the electric field sterilization kinetic model, N represents the number of microorganisms after treatment, represents the initial number of microorganisms, k represents the microbial inactivation rate constant, t represents the treatment time, E represents the electric field strength, and m and n are the exponential parameters of time and electric field strength respectively, indicates that the logarithmic ratio of the number of microorganisms decreases with the increase of treatment time and electric field strength.

6. The full-automatic processing method of pet health products optimized based on AI according to claim 1, characterized in that The thermal response characteristic analysis includes: Use differential scanning calorimetry and thermogravimetric analysis to determine the thermal stability parameters of active ingredients, and establish a relationship model between the thermal degradation rate and temperature: ； Among them, represents the relationship model between the thermal degradation rate and temperature, is the active ingredient content, is the frequency factor, is the activation energy, is the gas constant, is the absolute temperature, represents the degree of dependence of the reaction on concentration; The inverse algorithm for the heat diffusion equation adopts the nonlinear least squares method combined with finite element analysis, and inverses the internal temperature field distribution based on the surface temperature measurement data. The inverse algorithm for the heat diffusion equation is realized by solving the equation: ； Among them, represents the inverse algorithm of the heat diffusion equation, is the heat diffusion coefficient, is the internal heat source term, is the density, is the specific heat capacity, is the material temperature, is the Laplace operator.

7. A fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that, The real-time monitoring of the content of active ingredients by the Raman spectroscopy technology and the thermal imaging system includes: Use a portable Raman spectrometer to conduct on-line detection of the active ingredients in pet health products and collect spectral data; Apply a multivariate calibration model to establish a quantitative relationship between Raman spectroscopy and the content of active ingredients, and the model expression is: ； in, is the active ingredient content, is the intercept, For the The regression coefficient at the wavelength is For the The Raman spectrum intensity at wavelengths is is the error term, is the number of spectral data points included in the model; Adopt an infrared thermal imager to construct the surface temperature field distribution of the product; Fuse the Raman spectroscopy analysis results and the thermal field distribution information to establish a thermal degradation kinetic model of active ingredients: ； Wherein, represents the natural logarithm of the active ingredient retention rate, is the initial active ingredient content, is the active ingredient content, is the frequency factor, is the treatment time, is the activation energy, is the gas constant, is the absolute temperature, indicates that the active ingredient content decreases with the increase of treatment time; Predict the retention rate of the active ingredient under different process parameters in real time according to the thermal degradation kinetic model of the active ingredient, and provide feedback data for the optimization algorithm.

8. A fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that, The computational intelligence algorithm simultaneously optimizes the electric field parameters and the auxiliary temperature, including: Construct a multi-objective optimization model, and simultaneously consider the following optimization objectives: Maximize the microbial kill rate: ； Maximize the retention rate of the active ingredient: ； Minimize energy consumption: ； Among them, represents maximizing the microbial kill rate, represents maximizing the active ingredient retention rate, represents minimizing the energy consumption, represents the logarithmic function, and are the initial and treated microbial counts respectively, and are the contents of the th active ingredient after and before treatment respectively, is the weight coefficient, is the power, is the treatment time, represents the total number of active ingredients; Use an improved non-dominated sorting genetic algorithm to solve the multi-objective optimization problem; Based on non-dominated sorting and crowding distance calculation, obtain the Pareto optimal solution set to provide diverse optional solutions for decision-making; According to the product characteristics and quality requirements, select the electric field parameter combination that best meets the current needs from the Pareto front.

9. A fully automatic pet health product processing method optimized based on AI according to claim 1, characterized in that, The optimization method further includes constructing an adaptive parameter optimization control system, and the adaptive parameter optimization control system includes: Based on the Pareto optimal solution set, combined with real-time monitoring data, use the deep reinforcement learning method to realize the dynamic optimization of the processing parameters; The state space of the control system includes the current product characteristic parameters, the real-time monitored microbial inactivation rate and the retention rate of the active ingredient; The action space includes the adjustment amounts of the electric field parameters and the auxiliary temperature; The reward function is defined as a weighted combination of the microbial kill rate, the retention rate of the active ingredient and the energy consumption; Use a deep Q-network to realize the mapping from state to action; Through experience replay and target network technologies, improve the learning stability and efficiency, and realize the parameter adaptive optimization of the pet health product processing process.

10. A fully automatic pet health product processing system optimized by AI, which is used to execute a fully automatic pet health product processing method optimized by AI according to any one of claims 1-9, and is characterized in that, Including: A pulsed electric field generating unit for generating an adjustable high-voltage pulsed electric field, including a high-voltage pulsed power supply, a multi-stage electrode chamber and a parameter adjustment module; A temperature control unit for precisely adjusting and maintaining the processing temperature, including a precision temperature control device and a temperature sensor array; A product flow unit for controlling the flow state of the pet health product in the processing chamber, including a peristaltic pump, a fluid pipeline and a processing chamber; A monitoring and analysis unit, including a Raman spectrometer and an infrared thermal imager, for real-time monitoring of the active ingredient content and the temperature field distribution; A control unit, including a main controller, a data acquisition module and a human-machine interface, integrating a deep reinforcement learning algorithm to realize the adaptive optimization control of the system parameters.

Images