Method for processing nfc mandarin orange juice

Abstract

The application relates to the technical field of food processing, and discloses a NFC shatian orange juice processing method, which comprises the following steps: collecting the dynamic dynamic viscosity and mass flow of NFC shatian orange original juice as feedforward parameters, dynamically calculating the carbon dioxide injection amount, constructing a supercritical high-pressure three-phase mixed fluid by pressurizing and temperature adjusting, guiding the fluid to produce self-excited oscillation and cavitation shock enzyme inactivation, synchronously collecting the transient phase signal of pressure fluctuation, driving radio frequency electromagnetic waves to carry out polarization heat generation, compensating for latent heat of phase transition, carrying out stepped gradient decompression on the processed fluid, separating phases and releasing gas through a cyclone separator provided with a cryogenic reflux condenser, and outputting finished product juice. The heat flow phase compensation mechanism reduces local temperature fluctuation caused by cavitation heat absorption, guarantees stable inactivation of pectin methylesterase, reduces the escape of characteristic aroma components through stepped decompression and cryogenic reflux, and improves the flavor retention rate of the juice product.

Description

Technical Field

[0001] This invention relates to the field of food processing technology, specifically to a method for processing NFC Wogan mandarin orange juice. Background Technology

[0002] NFC (Not From Concentrate) juice requires enzyme inactivation and sterilization during processing to inhibit pectin methylesterase activity and maintain the system's rheological stability. In existing continuous non-thermal processing schemes, a common process is the synergistic treatment of supercritical carbon dioxide combined with hydrodynamic cavitation effects. However, this process suffers from thermodynamic and hydrodynamic technical drawbacks in actual operation. When supercritical carbon dioxide undergoes expansion and cavitation phase change in the flow field, it absorbs latent heat from within the system, leading to localized temperature fluctuations and uneven spatiotemporal distribution in the multiphase fluid. This endothermic physical phenomenon disrupts the thermodynamic equilibrium of the reaction system, thereby reducing the inactivation efficiency of pectin methylesterase.

[0003] Meanwhile, the NFC Wogan mandarin orange juice system contains a continuous aqueous phase, suspended pectin, and pulp cells, and the dynamic viscosity of the material fluctuates with processing batches or states. Existing processing equipment typically uses a fixed ratio or fixed flow rate for air intake control, failing to incorporate fluid viscosity changes into the feedback adjustment mechanism. This results in a decrease in the mass transfer and dissolution efficiency of carbon dioxide in high-viscosity fluids, making it difficult to continuously construct a stable high-pressure three-phase mixed fluid.

[0004] Furthermore, in the degassing stage after fluid processing, the conventional depressurization and venting operation generates a transient gas release process due to pressure difference. When gaseous carbon dioxide is rapidly released, it triggers interfacial turbulence. This process easily carries away volatile characteristic aroma components from the juice, causing loss of flavor substances and quality degradation in the final NFC juice product. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an NFC Wogan mandarin orange juice processing method. This method solves the problems in existing non-thermal processing of NFC juice, where the endothermic phase change of supercritical carbon dioxide cavitation leads to fluid temperature fluctuations, reduces the inactivation efficiency of pectin methylesterase, and affects rheological stability. Additionally, the transient gas release during depressurization and exhaust can easily cause the loss of volatile characteristic aromas, resulting in flavor degradation.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The NFC Wogan mandarin orange blend juice is pumped into the pipeline system, and the dynamic dynamic viscosity data and mass flow rate data of the NFC Wogan mandarin orange blend juice are collected as feedforward parameters. The amount of carbon dioxide injected is calculated based on the feedforward parameters. Liquid carbon dioxide is injected into the NFC Wogan mixed juice to form a primary mixture. The primary mixture is pressurized and the temperature is adjusted to reach the basic process pressure and temperature, so that the carbon dioxide is converted into a supercritical state and dissolved and mixed to form a high-pressure three-phase mixed fluid. The high-pressure three-phase mixed fluid is guided to flow through the fluid oscillator to generate fluid self-excited oscillation, alternately forming a transient subcritical low-pressure zone, causing carbon dioxide to expand and undergo phase change, and generating cavitation shock waves during pressure recovery. The transient phase signal of the pressure fluctuation of the high-pressure three-phase mixed fluid is collected simultaneously, and radio frequency electromagnetic waves are emitted to the high-pressure three-phase mixed fluid according to the transient phase signal to perform polarization heat generation and thermal energy compensation. The high-pressure three-phase mixed fluid after polarization heat generation and thermal energy compensation is subjected to gradient decompression and gas-liquid separation to discharge gaseous carbon dioxide and output the finished NFC Wogan juice.

[0007] Furthermore, the NFC Wogan mandarin orange blend juice comprises a continuous aqueous phase, a suspended pectin phase, and a pulp cell solid phase. The carbon dioxide injection amount is calculated based on the feedforward parameters, including the following steps: Obtain a preset basic mass ratio constant, and multiply the mass flow rate data by the basic mass ratio constant to obtain the theoretical basic injection volume; The built-in empirical viscosity compensation mapping relationship is invoked, and the corresponding positive compensation coefficient is extracted based on the dynamic dynamic viscosity data. The positive compensation coefficient increases as the dynamic dynamic viscosity data increases. The theoretical injection amount is multiplied by the positive compensation coefficient to calculate the carbon dioxide injection amount. The constant flow pump is then controlled to inject liquid carbon dioxide into the NFC Wogan mixed juice according to the carbon dioxide injection amount.

[0008] Furthermore, generating fluid self-excited oscillations and producing cavitation shock waves during pressure recovery includes the following steps: The high-pressure three-phase mixed fluid is passed through a wall-attached jet element, which includes a converging nozzle, a fluid action chamber, a wedge-shaped fluid distributor, and feedback channels arranged symmetrically on both sides. The high-pressure three-phase mixed fluid is accelerated by the converging nozzle to form a main jet. After entering the fluid action chamber, the main jet adheres to the side wall and entrains the surrounding fluid, resulting in a local decrease in static pressure. The entrained and stripped fluid flows back to the root region of the converging nozzle outlet through the symmetrically arranged feedback channels on both sides and laterally impacts the main jet. Combined with the guiding effect of the wedge-shaped fluid divider, the main jet is forced to deflect alternately in the fluid action chamber to form self-excited oscillation and to form a dynamic flow field inside the fluid action chamber. The local real-time cavitation number of the dynamic flow field is calculated in real time; the basic process pressure and the shrinkage ratio of the shrinkage nozzle cross-sectional area are adjusted in a coordinated manner to maintain the local real-time cavitation number within the range of 0.1 to 0.5.

[0009] Further, the transient phase signal of pressure fluctuations in the high-pressure three-phase mixture is acquired, including the following steps: The continuous analog electrical signal of the pressure fluctuation of the high-pressure three-phase mixed fluid is obtained by a high-frequency dynamic pressure transmitter, and then discretized into a digital pressure signal sequence after low-pass anti-aliasing filtering and analog-to-digital conversion. A sliding data window is opened to perform online DC removal on the digital pressure signal sequence. The digital pressure signal sequence after DC removal is used as the real part of the analytical signal. Discrete Hilbert transform is performed on the digital pressure signal sequence to obtain conjugate feature data as the imaginary part of the analytical signal. The four-quadrant arctangent algorithm is invoked to calculate the true transient principal phase distributed in the reference interval from negative pi to positive pi based on the algebraic sign and ratio relationship between the real part and the imaginary part of the analytical signal, thus obtaining the transient phase signal.

[0010] Furthermore, after obtaining the transient phase signal, phase delay compensation is performed, including the following steps: The lead compensation angle is calculated based on the multiplicative relationship between the signal main frequency period obtained from real-time statistics, the inherent transmission delay constant of the data link measured and stored in advance, and the full-cycle radian coefficient. The lead compensation angle is superimposed on the transient principal phase to generate the target phase parameter after time delay compensation correction, which serves as the time anchor point for driving the transmission of the radio frequency electromagnetic wave.

[0011] Furthermore, the control model for polarization heat generation and thermal energy compensation based on the transient phase signal transmitting the radio frequency electromagnetic wave to the high-pressure three-phase mixed fluid includes the following steps: Obtain the preset thermodynamic isothermal compensation power, and obtain the RF power modulation depth coefficient and the optimal heat flow matching phase difference; The nonlinear cosine modulation mathematical formula is called, and the difference between the target phase parameter and the optimal heat flow matching is used as the cosine variable to calculate the cosine value. The cosine value is multiplied by the RF power modulation depth coefficient and then added by one. Finally, it is multiplied by the thermodynamic isothermal compensation power to generate the target RF output power command. The solid-state radio frequency source is controlled to generate an amplitude-modulated radio frequency wave according to the target radio frequency output power command and apply it to the high-voltage three-phase mixed fluid.

[0012] Further, obtaining the difference between the RF power modulation depth coefficient and the optimal heat flow matching includes the following steps: The initial temperature, mass flow rate data, and dynamic dynamic viscosity data of the NFC Wogan blended juice are obtained through the input layer of the feedforward backpropagation neural network model. After the hidden layer performs nonlinear calculations, the output layer optimizes and outputs the RF power modulation depth coefficient and the optimal heat flow matching phase difference.

[0013] Furthermore, during polarization heat generation and thermal energy compensation, isothermal boundary maintenance is performed simultaneously, including the following steps: The high-pressure three-phase mixed fluid flows inside the microwave-transparent reaction tube and receives the radio frequency electromagnetic wave polarization to generate heat. The outer surface of the microwave-transparent reaction tube is covered with an isothermal boundary maintenance component. The real-time boundary temperature of the inner wall of the microwave-transparent reaction tube is collected in real time. Based on the basic process temperature as the target control temperature, a closed-loop calculation is performed to adjust the energy output of the isothermal boundary maintenance component, so that the tube wall temperature of the microwave-transparent reaction tube and the temperature of the high-pressure three-phase mixed fluid are kept in dynamic equilibrium.

[0014] Furthermore, the high-pressure three-phase mixed fluid after polarization heat generation and thermal energy compensation is subjected to gradient decompression, including the following steps: The system controls a multi-stage servo back pressure valve group arranged in series. Based on the pressure difference between the inlet and outlet of the main pipeline and the preset pressure drop distribution algorithm, it calculates the target set pressure value of each stage of back pressure valve and dynamically adjusts the valve core opening of each stage of back pressure valve through a digital PID algorithm to form a spatial step-like pressure reduction. Before the high-pressure three-phase mixed fluid enters the gas-liquid separation process, the high-pressure three-phase mixed fluid is guided to flow through the forced cooling heat exchange component for countercurrent heat exchange, thereby reducing the temperature of the high-pressure three-phase mixed fluid to the refrigeration process temperature range of 5°C to 10°C, and obtaining a gas-liquid-solid three-phase mixture after depressurization and cooling treatment.

[0015] Further, gas-liquid separation is performed, including the following steps: The gas-liquid-solid three-phase mixture is tangentially introduced into a hydrocyclone separator to separate the liquid phase fruit juice, the solid phase fruit pulp sacs, and the gaseous carbon dioxide. A cryogenic reflux condenser is installed at the exhaust port of the cyclone separator to control the temperature of the circulating refrigerant in the cryogenic reflux condenser between 1°C and 4°C, so that the volatile aroma components entrained in the gaseous carbon dioxide are condensed and liquefied and flow back to the liquid fruit juice at the bottom of the cyclone separator under the action of gravity.

[0016] This invention provides a method for processing NFC Wogan mandarin orange juice. It has the following beneficial effects: 1. This invention uses online acquisition of the dynamic dynamic viscosity and mass flow rate of fruit juice as feedforward parameters, and combines this with a viscosity compensation mapping relationship to dynamically correct the amount of carbon dioxide injected. This overcomes the influence of material viscosity fluctuations on the mass transfer and dissolution process, ensuring the precise proportion and efficient dissolution of liquid carbon dioxide in the juice, and ensuring the continuous construction of a physically stable high-pressure three-phase mixed fluid.

[0017] 2. This invention extracts the transient pressure phase signal during cavitation generated by the self-excited oscillation of the flow field and synchronously triggers radio frequency electromagnetic waves to polarize and compensate for the heat generation of the mixed fluid. This compensates for the latent heat absorbed during the carbon dioxide expansion and cavitation process, reduces the temperature gradient and fluctuation in the reaction zone, and ensures the stable inactivation efficiency of pectin methylesterase without increasing the overall basic process temperature.

[0018] 3. This invention employs a multi-stage servo backpressure valve assembly to perform step-by-step pressure reduction on the reacted materials, combined with a cyclone separator and a cryogenic reflux condenser to perform gas-liquid separation. The step-by-step pressure reduction mitigates the interfacial turbulence caused by transient gas release, while the cryogenic reflux condenser condenses the volatile aroma components that escape with carbon dioxide and refluxes them back into the liquid phase of the fruit juice, thereby improving the retention rate of flavor substances in the final product fruit juice. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a waveform diagram of static pressure fluctuation in the flow field according to the present invention; Figure 4 This is a transient principal value phase waveform diagram of the present invention; Figure 5 This is a waveform diagram of the radio frequency power control command of the present invention. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] See attached document Figure 1 The present invention provides an NFC Wogan orange juice processing system, comprising: a feed sensing unit, a high-pressure mixing unit, a phase reaction unit, a gas release separation unit, and a controller.

[0022] The feed sensing unit is located at the beginning of the fluid pipeline of the processing system. The feed sensing unit includes a feed screw pump, a rotational viscometer, and a mass flow meter. The rotational viscometer and the mass flow meter are installed in series on the main fluid pipeline downstream of the feed screw pump.

[0023] The high-pressure mixing unit is connected to the feed sensing unit. The high-pressure mixing unit includes a fluid injector, a constant-flow pump, a plunger pump, a preheating sleeve, and a static mixer. The side inlet of the fluid injector is connected to the constant-flow pump. The plunger pump, preheating sleeve, and static mixer are connected in series downstream of the fluid injector.

[0024] The coherent reaction unit is located at the downstream fluid outlet of the static mixer. The coherent reaction unit includes a microwave-transparent reaction tube, a pressure transmitter, a high-frequency dynamic temperature sensor, an radio frequency heating array, and a fluid oscillator. The fluid oscillator is built into the cavity of the microwave-transparent reaction tube. The pressure transmitter, high-frequency dynamic temperature sensor, and radio frequency heating array are fixedly mounted on the inner and outer walls of the microwave-transparent reaction tube.

[0025] The gas release separation unit is connected after the coherent reaction unit. The gas release separation unit includes a cyclone separator and an exhaust valve at its top.

[0026] The controller establishes electrical and communication connections with the feed screw pump, rotary viscometer, mass flow meter, constant flow pump, plunger pump, preheating sleeve, pressure transmitter, radio frequency heating array, and exhaust valve, respectively.

[0027] See attached document Figure 2 This invention provides a method for processing NFC Wogan mandarin orange juice, comprising the following steps: The S100 feed screw pump pumps the NFC Wogan mandarin orange blend juice, comprising a continuous aqueous phase, suspended pectin, and pulp / cell solid phase, into the pipeline system. During fluid flow, a rotational viscometer measures the dynamic viscosity of the NFC Wogan mandarin orange blend juice flowing through the pipeline. A mass flow meter measures the mass flow rate of the NFC Wogan mandarin orange blend juice flowing through the pipeline. The rotational viscometer and mass flow meter transmit the collected data as feedforward parameters to the controller in real time. S200: Based on the received feedforward parameters, the controller calculates the corresponding carbon dioxide injection control command and sends the command to the constant flow pump. The constant flow pump executes the command, injecting liquid carbon dioxide into the main pipeline containing NFC Wogan mandarin orange blend juice through the fluid injector to form a primary mixture. The primary mixture flows through a plunger pump for mechanical pressurization, and then enters a preheating jacket for temperature regulation. The pressurization and temperature regulation operations bring the fluid to the preset basic process pressure and basic process temperature. The basic process pressure is greater than the critical pressure of carbon dioxide. In this state, carbon dioxide is converted to a high-density supercritical state and dissolves in the liquid phase of the juice and the micropores of the solid phase of the pulp. The dissolved material is then subjected to the turbulence effect of the static mixer to form a stable high-pressure three-phase mixed fluid. In S300, a high-pressure three-phase mixed fluid flows out of the static mixer and into the microwave-transparent reaction tube. The high-pressure three-phase mixed fluid flows through the geometrically constrained flow channel inside the fluid oscillator, generating self-excited oscillations and forming a transient subcritical low-pressure region locally. Carbon dioxide, in a high-density state, undergoes an expansion phase change in the low-pressure region, generating cavitation shock waves that act on pectin methylesterase and microbial cell membranes through fluid shear force. In synchronous operation, the pressure transmitter detects the transient phase signal of a sudden pressure drop inside the pipeline and sends the phase signal to the controller. Upon receiving the phase signal, the controller outputs a radio frequency (RF) trigger command to the RF heating array. The RF heating array emits RF electromagnetic waves according to the trigger command. The RF electromagnetic waves pass through the microwave-transparent reaction tube, polarizing and generating heat in the fluid flowing through the pipeline. The heat energy converted by the RF electromagnetic waves compensates for the heat absorbed by the expansion of carbon dioxide during phase change, maintaining a stable temperature environment in the cavitation region. S400: The treated fluid is output to the hydrocyclone separator. The controller sends a control signal to the exhaust valve to adjust the opening, establishing a static pressure gradient environment between the various chambers of the hydrocyclone separator. Carbon dioxide gas is released through the exhaust channel at the top of the hydrocyclone separator. The degassed NFC pulp solid phase and juice liquid phase are continuously discharged from the outlet at the bottom of the hydrocyclone separator and enter subsequent processing stages.

[0028] Based on the structure of the processing system, the physical pumping process of the original juice is carried out through the following implementation steps.

[0029] S101, In this embodiment, the input material received by the system is NFC Wogan mandarin orange mixed juice. In the general terminology of fluid dynamics classification and claims, this material is specifically embodied as a multiphase mixed fluid, which is mainly composed of a continuous aqueous phase, suspended pectin, and a granular pulp cell solid phase. Due to the lack of filtration and homogenization treatment, macroscopic non-homogeneous phase interfaces usually exist inside the fluid.

[0030] S102, as a preferred embodiment, the feed screw pump within the feed sensing unit is activated to continuously draw the aforementioned multiphase mixed fluid into the pipeline system. Within the scope of the claims, any conveying component used to prevent the rupture of solid vesicles can be categorized as a low-shear volumetric conveying device, and the feed screw pump in this embodiment is a specific implementation of such a device. The feed screw pump utilizes a sealed chamber formed by the internal stator and rotor to propel the material forward along the axial direction. From a physical perspective, since the volume of the sealed chamber during a single push is fixed, the macroscopic fluid throughput is primarily linearly positively correlated with the rotor speed, allowing the system to precisely intervene in the fluid flux by adjusting the external motor speed.

[0031] The output flow rate of a feed screw pump is determined by its mechanical structural parameters and operating conditions. In the system control logic, the controller internally constructs a flow mapping model, transforming specific formulas into parameter calculations. This model comprehensively considers the preset macroscopic equivalent density of the multiphase mixture, the empirical volumetric efficiency of the pipeline under current fluid resistance, and the inherent combination of mechanical parameters of the pump body, including the rotor's eccentricity, cross-sectional diameter, and stator helical lead. Based on these physical parameters, the controller analyzes the target input flow rate requirement and inversely calculates the target operating speed of the feed screw pump drive motor.

[0032] After the calculation is completed, the controller outputs a frequency conversion control command to the feed screw pump, adjusting its actual operating speed by changing the power supply frequency, thereby achieving dynamic control of the material flow rate entering the pipeline system. For the frequency conversion speed control of the feed screw pump drive motor, those skilled in the art can use a general-purpose industrial frequency converter combined with a basic speed closed-loop algorithm. The specific electrical wiring method and motor drive principle are well-known technologies in this field and will not be elaborated here.

[0033] In S103, the multiphase mixed fluid flows along the main pipeline under the thrust of the feed screw pump. The volumetric displacement of the sealed chamber results in a relatively gentle fluid velocity gradient within the pipeline, ensuring that the mechanical shear force generated by the flow field is below the physical rupture threshold of the pulp cell solid phase. Here, this physical rupture threshold is directly related to the strength of the cellulose and pectin matrix of the cell wall, and can usually be obtained by conducting rheological shear failure tests on the sample beforehand. In actual operation, the system maintains the mechanical shear stress within the pipeline within a safe range of 10Pa to 50Pa by limiting the upper limit of the operating speed. Based on these mild fluid transport conditions, when the NFC Wogan mandarin orange mixed juice enters the pipeline system, its original pulp tissue structure and suspension state can be basically maintained in their original form, and the continuous aqueous phase and pulp cell solid phase are transported to the subsequent sensing and processing units in a stable physical ratio.

[0034] After a smooth transport process, the multiphase fluid mixture enters the dynamic acquisition stage of its fluid characteristic parameters. This feedforward online acquisition process is explained through the following steps.

[0035] S104, the multiphase mixed fluid travels along the main pipeline and flows through a rotational viscometer installed in series. In the general patent claims, the component used to obtain the internal frictional characteristics of the fluid is usually referred to as a rheological property sensing component. The rotational viscometer in this embodiment is a specific industrial implementation of this component. Because the NFC Wogan mandarin orange mixed juice contains pectin and pulp suspensions, it exhibits non-Newtonian fluid characteristics, and its viscosity value changes dynamically with fluctuations in local solids content. In operation, the rotational viscometer immerses its measuring probe in the central region of the fluid in the main pipeline. By detecting the fluid shear resistance torque experienced by the probe at a constant rotation speed, it measures and outputs the dynamic dynamic viscosity of the fluid in real time. For the torque detection sensing structure and basic torque-to-viscosity conversion logic inside the rotational viscometer, those skilled in the art can refer to conventional industrial online viscosity measurement equipment for configuration. Its basic hardware principles are well-known in the field and will not be elaborated here.

[0036] S105, in synchronous flow mode, the multiphase mixed fluid continues to pass through the downstream mass flow meter installed immediately. From the perspective of patent protection scope, devices used to determine the scale of material conveying can be collectively referred to as fluid flux sensing components. This embodiment selects a mass flow meter to meet the measurement needs of complex three-phase systems. In actual working conditions, conventional volumetric flow meters are easily affected by physical interference from uneven distribution of dissolved gases or fruit particles within the fluid, resulting in measurement errors. The mass flow meter, however, directly measures the true total mass of the substance flowing through the pipe based on the Coriolis force effect generated when the fluid passes through a specific vibrating pipe. It should be noted that this measurement mode can effectively reduce the volume fluctuation interference caused by uneven distribution of the internal gas or solid phases, thereby obtaining more accurate material mass flux data.

[0037] S106, as a preferred method, after the rotational viscometer and mass flow meter acquire the corresponding physical characteristic parameters, they synchronously transmit the dynamic dynamic viscosity data and mass flow data to the controller via an industrial communication link. High-frequency electrical interference may occur when the system acquires field sensor data. To reduce the impact of data spikes caused by occasional mechanical vibrations in the pipeline or abnormally large fruit particles passing through the sensor probe, the controller performs a sliding window averaging filter operation in its internal data processing unit after receiving the original electrical signal sequence.

[0038] In this data processing logic, considering the continuity of fluid transport, the data sampling period for the rotational viscometer and mass flow meter is typically set to 10ms to 50ms. The controller sets the length parameter of the sliding window to N. To effectively filter out the periodic fluctuations caused by the mechanical movement of the pump, the parameter N is set to cover at least one complete mechanical propulsion cycle of the feed screw pump, and its value range is typically between 20 and 100 sampling points. The controller calculates the arithmetic mean of N consecutive historical data within the current window in real time and uses it as the current valid output.

[0039] The smoothed data stream formed after filtering is defined by the system as feedforward parameters. This set of feedforward parameters objectively characterizes the macroscopic viscous resistance properties and actual mass load of the raw material at the current process node, and serves as the basic input condition to provide a computational benchmark for the subsequent implementation of adaptive phase construction and proportioning control by the system.

[0040] After acquiring the feedforward parameters, the process flow enters the multiphase fluid construction stage. The specific implementation steps for the dynamic calculation of carbon dioxide injection and primary mixing in this stage are as follows.

[0041] S201: After acquiring the feedforward parameters of dynamic dynamic viscosity and mass flow rate data, the controller's internal data processing unit begins dynamic calculation of the carbon dioxide injection ratio. Since the system does not employ a conventional fixed-proportion static open-loop control mode, dynamic correlation calculation logic based on rheological characteristics is required. Specifically, the controller internally presets a basic mass ratio constant for NFC Wogan orange juice. To ensure that the mixing system can provide sufficient cavitation phase change working fluid in subsequent process pipelines, the preset basic mass ratio constant is typically set within the range of 3% to 8%. The controller performs a basic product operation between the received mass flow rate data and the basic mass ratio constant to derive the theoretical basic injection volume.

[0042] Building upon this calculation, and considering that real-time fluctuations in fluid viscosity directly impact hydrodynamic mass transfer resistance, the controller further incorporates a dynamic viscosity compensation mechanism. An increase in dynamic dynamic viscosity objectively reflects a localized increase in the concentration of pectin or pulp suspended solids within the fluid. This physical phenomenon hinders the mass transfer and dissolution efficiency of the gas-liquid-solid interface under subsequent high-pressure conditions. To overcome this mass transfer resistance, the controller invokes a built-in empirical viscosity compensation mapping relationship, extracting the corresponding positive compensation coefficient based on the currently received dynamic dynamic viscosity parameter. This empirical viscosity compensation mapping relationship is typically pre-stored in the controller's memory in the form of a one-dimensional data lookup table or a polynomial fitting function. The basic logic of this mapping relationship is: as dynamic dynamic viscosity increases, the system output positive compensation coefficient increases accordingly. The positive compensation coefficient typically fluctuates between 1.0 and 1.3. The controller multiplies the previously calculated base injection volume by this positive compensation coefficient to deduce the carbon dioxide injection volume corrected for process environment parameters. It should be noted that the specific coordinate point data of the empirical viscosity compensation mapping relationship can be obtained by conducting static high-pressure dissolution mass transfer tests on Wogan mandarin orange mixed juice with different solid contents in advance.

[0043] S202, after the calculation logic is completed, the controller converts the carbon dioxide injection amount into a corresponding drive level control signal and sends the control signal to the constant flow pump. In the generalized claims text, components that perform such proportional fluid delivery actions can be collectively referred to as adjustable metering pumping equipment, and the constant flow pump in this embodiment is an industrial application of such equipment. After receiving the control command, the constant flow pump dynamically adjusts the operating frequency of the internal servo or variable frequency motor, continuously extracts liquid carbon dioxide from the external high-pressure storage tank, and pumps and pressurizes it towards the main fluid pipeline in response to the flow rate requirement of the carbon dioxide injection amount.

[0044] S203, liquid carbon dioxide arrives at the fluid injector under the mechanical thrust of a constant flow pump. In this embodiment, the fluid injector adopts a T-shaped pipe structure design, with a branch introduced at its side end forming a geometric relationship perpendicular to the axis of the main pipe. Liquid carbon dioxide, from the side inlet of the fluid injector, flows into the NFC Wogan mixed juice flowing smoothly in the main pipe in a jet form with high local dynamic pressure. The two fluids with different physical properties undergo physical mixing and local fluid shearing in the confluence area. Due to the current spatial environment of the confluence pipe not yet reaching the high temperature and high pressure critical conditions required for the supercritical phase, only a small portion of the liquid carbon dioxide injected by the side jet dissolves in the continuous aqueous phase of the juice. Under normal circumstances, the main carbon dioxide is dispersed in the juice fluid in the form of tiny droplets or fine microbubbles formed by vaporization due to local decompression and heating. This macroscopic dispersion system, which has not reached a deep thermodynamic dissolution state, composed of the continuous aqueous phase, the pulp solid phase, and the dispersed carbon dioxide, forms the primary mixture that enters the subsequent pressurization and temperature control unit.

[0045] After the primary mixture is formed, the system achieves the regulation of the critical state of the fluid and the construction of a steady-state system through the following implementation steps.

[0046] S204, the primary mixture flows along the main pipeline and enters the plunger pump. In the broader generalization of the claims, the device used to increase the static pressure of the fluid inside the system pipeline can be summarized as a mechanical booster assembly, and the plunger pump in this embodiment is a specific manifestation of this assembly. The plunger pump compresses the fluid entering the cavity through internal reciprocating motion, increasing the basic process pressure inside the pipeline to a preset value. Considering that the physical critical pressure point of carbon dioxide is approximately 7.38 MPa, in order to enable the carbon dioxide in the fluid system to cross the pressure boundary of phase transition and to reserve a certain pressure fluctuation margin for subsequent fluid resistance, the system typically sets the target control range of this basic process pressure to 8 MPa to 15 MPa.

[0047] From a control implementation perspective, the aforementioned pressurization process is not an open-loop operation. The controller receives static pressure feedback signals from a pressure transmitter installed downstream of the pipeline in real time and compares them with a preset target value for the basic process pressure. Based on the pressure deviation generated by the comparison, the controller dynamically outputs frequency conversion commands to the drive motor of the plunger pump, thereby adjusting the reciprocating frequency of the plunger and establishing a stable high-pressure physical environment within the pipeline.

[0048] S205, under the continuous pushing action of the plunger pump, the primary mixture continues to flow through the preheating jacket. Since the formation of the supercritical phase requires the simultaneous satisfaction of both pressure and temperature thermodynamic conditions, the fluid undergoes temperature regulation here. Conceptually, the preheating jacket corresponds to the heat exchange component in the claims. As a preferred embodiment, the preheating jacket employs a water bath heating or jacketed temperature control structure to transfer heat to the fluid inside the pipeline, bringing it to a preset base process temperature.

[0049] The physical critical temperature of carbon dioxide is approximately 31.1 degrees Celsius. To balance phase formation with the retention of heat-sensitive flavor compounds in NFC tangerine juice, the controller typically maintains the base process temperature within a safe range of 35°C to 45°C. In practice, a temperature sensing element inside the preheating jacket feeds real-time fluid temperature back to the controller. The controller then outputs a control signal using a basic proportional-integral-derivative algorithm to drive an electrically operated regulating valve in the external pipeline, thereby altering the heat transfer fluid flux into the preheating jacket jacket and achieving precise closed-loop maintenance of the base process temperature.

[0050] Under these thermodynamic temperature and pressure conditions, liquid or partially vaporized carbon dioxide dispersed within the juice system crosses the phase transition boundary and transforms into a high-density supercritical state. The physical properties of supercritical carbon dioxide lie between those of a gas and a liquid, possessing diffusion capabilities close to those of a gas and solubility close to that of a liquid. This allows for an effective reduction in mass transfer resistance within the fluid.

[0051] S206, a fluid system containing supercritical carbon dioxide enters a static mixer under dynamic pressure. In this embodiment, multiple sets of spatially staggered flow guide baffles are fixedly installed inside the static mixer. Based on general fluid dynamics mixing principles, this structure does not rely on external moving parts for power, but utilizes the fluid's own kinetic energy as an energy source. As the fluid flows through these internal geometries under pressure, continuous flow splitting, radial deflection, and physical convergence occur.

[0052] This passive fluid turbulence effectively reduces surface tension resistance between multiphase interfaces. Under hydromechanical shearing, supercritical carbon dioxide is further dispersed and fully dissolved in the continuous aqueous phase of the fruit juice. Based on the high permeability of the supercritical state, carbon dioxide can also penetrate biological tissues and enter the micropores of the pulp cyst solid phase. After structural reorganization of the internal flow channels of the static mixer, the primary mixture, which originally exhibited macroscopic phase distribution differences, transforms into a macroscopically uniform high-pressure three-phase mixed fluid. This high-pressure three-phase mixed fluid establishes a stable gas-liquid-solid-dissolution equilibrium at the microscopic scale, thus providing a fluid basis with a uniform working fluid distribution for the downstream core reaction unit.

[0053] After the steady-state high-pressure three-phase mixed fluid is constructed, the system enters the core stage of fluid morphology regulation and enzyme inactivation reaction. This stage achieves passive fluid self-excited oscillation and transient pressure reduction through the following steps.

[0054] S301 is a core component for the smooth flow of high-pressure three-phase mixed fluid into the interior of a microwave-transparent reaction tube. In the definition of the claims, a structure that induces periodic instability in the flow field and has no moving parts is summarized as a passive fluid deflection component; the fluid oscillator in this embodiment is a specific implementation of this component. As a preferred embodiment, the fluid oscillator adopts a typical wall-attached effect jet element structure design, with a converging nozzle, a fluid action chamber, symmetrically arranged feedback channels on both sides, and a wedge-shaped fluid distributor at the end distributed sequentially along the fluid movement direction. When the fluid flows through the converging nozzle, the flow velocity increases accordingly due to the reduction in the flow cross-sectional area. To obtain sufficient jet kinetic energy, the cross-sectional area reduction ratio of the converging nozzle is typically designed to be 3:1 to 5:1. This process converts part of the fluid's static pressure energy into dynamic pressure energy, thereby forming a main jet with high kinetic energy and injecting it into the downstream fluid action chamber.

[0055] S302. To understand the formation mechanism of the aforementioned self-excited oscillation, it is necessary to further explore the dynamic behavior of the jet within the fluid-acting chamber. As the main jet travels within the fluid-acting chamber, it is affected by the inherent small asymmetric disturbances of the flow field, causing the jet to deviate laterally and adhere to a side wall of the chamber. This wall-attached flow behavior is based on the principle of wall adhesion in fluid mechanics. After adhering to the wall, the jet will entrain surrounding fluid near that side wall, leading to a decrease in local static pressure and thus forming a lateral pressure gradient, forcing the jet to flow even closer to that side wall.

[0056] Simultaneously, a portion of the entrained fluid is guided through the feedback channel on this side to the outlet root region of the contraction nozzle. The lateral impact of the feedback fluid on the main jet disrupts the original wall-attachment equilibrium, forcing the main jet to laterally cross the central axis, deflect, and adhere to the other side wall of the chamber. On this side wall, the aforementioned wall-attachment and feedback mechanism repeat. This pure hydrodynamic feedback network, independent of external mechanical excitation sources, causes the jet to generate continuous high-frequency alternating deflections between the two outlets of the wedge-shaped fluid distributor, thereby forming a passive fluid self-excited oscillation state. In actual operation, the design of the internal geometry of the fluid oscillator and its response to the fluid oscillation frequency can be calculated and matched by those skilled in the art based on empirical hydrodynamic models related to the Slauhal number. Typically, for the rheological characteristics of the high-pressure juice system, this self-excited oscillation frequency is controlled within the range of 500Hz to 2000Hz. Its basic geometric construction and flow field parameter adjustment are well-known techniques in the field and will not be elaborated here.

[0057] S303, as the jet alternately passes through the two outlets of the fluid divider in a high-frequency oscillating state, the spatial velocity distribution within the flow field undergoes corresponding periodic fluctuations. According to Bernoulli's principle, an increase in local fluid velocity is usually accompanied by a decrease in static pressure. In this embodiment, this alternating high-speed jet generates local low-pressure zones alternately in the wake region of the fluid divider and the outlet expansion section.

[0058] It should be noted that the system controls the transient static pressure of these local low-pressure zones below the critical pressure point of carbon dioxide (approximately 7.38 MPa) by matching the initial input process pressure of the plunger pump with the shrinkage ratio of the converging nozzle cross-sectional area of ​​the fluid oscillator. To avoid unnecessary boiling and cavitation in the aqueous phase and ensure that carbon dioxide can undergo phase change, the static pressure valley of this low-pressure zone is typically designed to be controlled between 4 MPa and 6 MPa. From a thermodynamic phase perspective, this fluid dynamics abrupt change caused by geometric constraints locally and alternately constructs transient subcritical low-pressure zones within the originally stable supercritical high-pressure flow field. The high-pressure three-phase mixture flowing through this region will experience a rapid pressure drop within an extremely short timescale. This alternating transient pressure drop environment disrupts the original high-pressure phase equilibrium of carbon dioxide, providing the basic boundary conditions for the subsequent rapid expansion, phase change, and cavitation collapse physical processes of the working fluid.

[0059] After undergoing localized fluid dynamic depressurization, the microscopic flow field conditions change. Based on this, the system achieves the process of expansion phase change and cavitation micro-shock wave enzyme inactivation through the following steps.

[0060] S304, as a preferred embodiment, involves a process where, when a high-pressure three-phase mixture flows through the aforementioned transient subcritical low-pressure region, its internal static pressure drops below the critical pressure for carbon dioxide's phase transition. The supercritical carbon dioxide dissolved in the continuous aqueous phase and permeating into the micropores of the fruit pulp cells undergoes a thermodynamic transition due to the decrease in ambient pressure, changing from a supercritical state to a gaseous state. Based on the fundamental physical laws of gas-liquid phase transitions, this process is typically accompanied by a physical expansion of the phase volume.

[0061] S305, under this expansion effect, carbon dioxide is released inside the fluid, forming a large number of micron-sized cavitation microbubbles. In this embodiment, to ensure that the cavitation effect can effectively act on enzyme molecules without causing deterioration of the macroscopic physical properties of the juice, the system, based on the flow field structure parameters of the preceding stage, typically controls the initial diameter of these cavitation microbubbles within the range of 10μm to 50μm.

[0062] To quantitatively assess and control this phase change cavitation state in engineering applications, the controller incorporates the dimensionless hydrodynamic parameter of cavitation number as an evaluation criterion during the design phase. Its corresponding physical formula is as follows: ; In the formula, Represents the local real-time cavitation number of the flow field; This represents the actual ambient static pressure in the transient subcritical low-pressure zone, and its value is usually collected in real time by a high-frequency dynamic pressure sensor installed on the pipe wall. This represents the equivalent saturated vapor pressure of the multiphase mixture at the current base process temperature; represents the macroscopic equivalent density of the multiphase mixture; v represents the characteristic jet velocity of the fluid after passing through the constricting nozzle. The controller needs to acquire these parameters in real time to execute the formula calculation. To ensure the rigor of the fluid dynamics calculation, the characteristic jet velocity v is not calculated solely based on the initial juice flow rate, but rather by summing the juice mass flow rate data acquired from the feedforward and the target carbon dioxide mass flux injected by the constant flow pump, combined with the real-time macroscopic equivalent density under the current process temperature and pressure and the known flow cross-sectional area of ​​the nozzle. Equivalent saturated vapor pressure. Macroscopic equivalent density The controller then dynamically interpolates the fluid temperature based on the real-time measured fluid temperature by calling a pre-stored empirical mapping table of physical properties (such as the Antoine equation fitting table) in the storage unit.

[0063] It should be noted that, in order to induce a moderate and controllable cavitation phase change in the juice system, the system coordinates the base process pressure output by the plunger pump and the orifice diameter of the constricted nozzle of the fluid oscillator to control the cavitation number. The value is set between 0.1 and 0.5. This range ensures a sufficient cavitation bubble generation rate while preventing the flow field from entering a hypercavitation state that hinders overall material transport.

[0064] At the microscopic level, in S306, the carbon dioxide bubbles released from the pulp sacs expand and evolve, generating an outward mechanical expansion force on the cell walls and internal tissue structures. This mechanical force, to a certain extent, expands the micropores of the pulp tissue, thus providing a smoother physical channel for the deep penetration of subsequent radio frequency energy and the mass transfer and extraction of soluble solids.

[0065] As the multiphase fluid continues to travel downstream towards the wake region of the fluid action chamber and the outlet expansion section under dynamic pressure, the flow cross-sectional area of ​​the flow field gradually recovers. Due to this change in geometry, the local dynamic pressure of the fluid is converted back into static pressure, causing the ambient pressure to rapidly rise back to the aforementioned basic process pressure range.

[0066] S307, under the thermodynamic conditions of macroscopic pressure recovery, the aforementioned cavitation microbubbles cannot maintain their gas phase volume and thus undergo asymmetric collapse under the compression of the surrounding high-pressure fluid. According to general cavitation kinetics principles, at the moment of collapse, the rate at which the bubble walls collapse towards the center near the solid boundary or phase interface is not uniform. This non-uniform collapse releases micro-shock waves and high-speed micro-jets within a very small local spatial range.

[0067] This micro-shock wave can generate strong transient mechanical shear stress in the microscopic environment of a fluid. For active macromolecular proteins such as pectin methyl esterase contained in NFC tangerine juice, the mechanical shearing effect generated by the micro-shock wave helps to break the secondary bonds that maintain their spatial conformation, promoting the unfolding or depolymerization of the secondary and tertiary structures of the enzyme protein. Through this purely physical mechanical cavitation effect, the system can effectively reduce the catalytic activity of endogenous enzymes in the juice at a relatively low macroscopic process temperature, thereby achieving targeted enzyme inactivation. This mechanism can achieve the process objective of non-thermal or microthermal enzyme inactivation while maintaining the original fresh flavor of the juice.

[0068] To achieve time-domain coordination between radio frequency electromagnetic energy and hydrodynamic phase transition states in subsequent processes, the system needs to acquire real-time temporal characteristics of periodic physical changes within the flow field. In this stage, the system detects and extracts the transient pressure phase through the following steps.

[0069] S308, during the process of high-frequency alternating fluctuations of the high-pressure three-phase mixed fluid inside the fluid oscillator, its pressure change information is detected synchronously. In the definition of the claims, the hardware structure responsible for performing this detection action is summarized as a flow field state sensing component. In this embodiment, this component specifically employs a high-frequency dynamic pressure transmitter. As a preferred embodiment, the probe of the high-frequency dynamic pressure transmitter is flush-mounted on the inner wall of the expansion section at the outlet of the fluid oscillator to avoid the probe protruding into the flow field and introducing additional turbulence resistance.

[0070] The high-frequency dynamic pressure transmitter converts captured local mechanical pressure fluctuations in the flow field into a continuous analog electrical signal. In actual engineering operating environments, considering the broadband electromagnetic interference often caused by motor frequency converters or high-pressure pump sets in industrial sites, this analog electrical signal needs to pass through a pre-processing hardware conditioning circuit before being introduced into the controller for digital calculations. According to general signal processing principles, the sampling frequency of the high-speed analog-to-digital converter built into the controller is typically configured to be 5 to 10 times the aforementioned expected highest oscillation frequency. Based on this sampling rate, the low-pass filter inside the conditioning circuit sets its cutoff frequency between the highest effective signal frequency and the Nyquist frequency, thereby performing anti-aliasing filtering on the signal. This physical process aims to filter out high-frequency noise components, providing an analog baseband signal with a high signal-to-noise ratio for subsequent digital phase extraction.

[0071] S309, the hardware-filtered analog signal enters the high-speed analog-to-digital converter inside the controller and is discretized into a digital pressure signal sequence. The data processing unit inside the controller takes over this discrete sequence and starts the instantaneous phase calculation algorithm. Since the system needs to extract phase information that changes continuously over time, rather than a single frequency spectrum line, this embodiment uses digital signal processing logic based on discrete Hilbert transform.

[0072] In the specific execution of the algorithm, the data processing unit first constructs a complex-form analytic signal based on the input discrete digital pressure signal sequence. For the preprocessing of the input data, the controller opens a sliding data window in memory and calculates the arithmetic mean of the signal sequence within the current window in real time, subtracting it point by point from the original data, thereby achieving online DC removal of the signal. Subsequently, the controller uses the DC-removed discrete digital pressure signal as the real part of the analytic signal, and simultaneously uses the conjugate feature data obtained after processing the signal sequence through a discrete Hilbert transform as the imaginary part of the analytic signal.

[0073] After obtaining the real and imaginary parts of the aforementioned analytical signal, the data processing unit proceeds to the phase calculation stage. From the perspective of algorithmic logic completeness, if the basic arctangent operation is directly executed, the algorithm will fall into a dead zone of division by zero when the real part approaches zero, and the ordinary arctangent function cannot distinguish the specific quadrant of the signal. To avoid this logical anomaly, the controller directly calls the four-quadrant arctangent algorithm in its underlying code implementation. By synchronously judging the algebraic signs of the imaginary and real parts and combining their ratio, the true transient principal phase parameter distributed within the reference angle range from negative to positive pi is calculated. For the specific convolution calculation process of the discrete Hilbert transform and the order matching of the finite impulse response filter, those skilled in the art can program and configure it according to standard digital signal processing theory; its underlying implementation mechanism is well-known in the field and will not be elaborated here.

[0074] S310: After obtaining the transient principal phase distributed within the reference interval, the system does not directly use it as the control trigger reference for subsequent stages. Due to the electromechanical response of the sensor, the capacitive inductance characteristics of the analog hardware circuit, and the algorithm execution cycle, there will be an objective group delay. Directly using the currently calculated phase will usually cause a time misalignment between the actual physical flow field and the control action.

[0075] To correct this time-domain deviation, the controller introduces phase delay compensation at this stage. During factory testing or initial calibration, the inherent transmission delay constant of the entire data link is pre-measured and stored. The data processing unit calculates the lead compensation angle based on the real-time statistically obtained signal frequency period, the aforementioned inherent transmission delay constant, and the multiplicative relationship of the full-cycle radian coefficient, and directly superimposes it onto the calculated transient principal phase. The target phase parameter, corrected by time delay compensation, is refreshed into the controller's output register, serving as the precise time anchor point for driving downstream RF energy release. Through this closed-loop signal detection and digital calculation, the system transforms flow field pressure fluctuations into timing parameters that can be precisely invoked at the control level.

[0076] After obtaining the precise time-domain timing characteristics of the flow field, the system enters the energy coupling control stage. To ensure that the release of radio frequency electromagnetic energy and the phase transition process of the microscopic flow field are time-domain coordinated, the system constructs and executes a thermal-fluid coherent algorithm and control model through the following implementation steps.

[0077] S311, the transient principal phase after time delay compensation reflects the periodic alternation of the fluid inside the transmissive reaction tube between low-pressure expansion and high-pressure collapse. In the superordinate generalization of the claims, the module responsible for coordinating energy output according to this fluid state is summarized as a thermal-fluid coherent control component. In this embodiment, this component is implemented by the algorithm execution unit inside the controller. From a physical mechanism perspective, when the fluid is in the low-pressure expansion phase, the micropores of the pulp open and a large number of cavitation microbubbles are generated, and the macroscopic equivalent dielectric constant of the multiphase mixed fluid changes dynamically; while when the fluid enters the high-pressure collapse phase, the shock wave release generates mechanical shearing. Conventional constant power radio frequency heating is prone to causing local hot spots or energy reflection. Based on the above-mentioned changes in microscopic characteristics, the system establishes a control strategy that makes the radio frequency output power periodically modulated to follow the fluid pressure phase.

[0078] S312, the execution unit calculates the corresponding RF output power command in real time based on the received transient principal phase. As the core energy coherent coordination mechanism of this scheme, the controller uses the following nonlinear cosine modulation mathematical formula to construct the control model: ; In the formula, This represents the target RF output power command calculated at discrete sampling time n; This represents the reference heating power required to maintain the basic process temperature. It should be noted that, due to the expansion and phase change of carbon dioxide bubbles within the flow field, which absorbs a significant amount of latent heat from the surrounding fluid, the reference heating power is essentially a thermodynamic isothermal compensation power to prevent a sharp drop in the microenvironment temperature that could affect enzyme inactivation. Its value is obtained through closed-loop calculation by the upstream basic PID temperature loop based on the deviation between the preset basic process temperature and the real-time fluid temperature obtained by the temperature probe. This represents the dimensionless RF power modulation depth coefficient, with a value range of 0.1 to 0.6, used to limit the peak-to-valley difference in power fluctuations to protect the solid-state RF source. This represents the transient principal phase after time delay compensation correction; The optimal heat flow matching phase difference represents the phase offset angle between the peak RF power and the low-pressure trough of the flow field, and its value is constrained within the reference electrical angle range of [-π, π]. This formula establishes a precise time-domain mapping relationship between the periodic motion of fluid mechanics and the emitted energy of RF electromagnetic waves, enabling the RF energy to reach its peak during the microscopic window period of bubble expansion and pore opening, thereby achieving deep energy penetration.

[0079] S313. In actual industrial processing environments, the initial physical properties of Wogan mandarin orange juice vary objectively depending on the harvesting batch. This variation leads to a highly nonlinear dynamic dielectric response of the flow field. To dynamically optimize the key control parameters in the aforementioned core formula under different operating conditions, as an optimal approach, the controller embeds a pre-trained feedforward-backward propagation neural network model.

[0080] The internal hierarchical structure of this neural network model is designed as a fully connected topology consisting of an input layer, a single hidden layer, and an output layer. Data flows unidirectionally from the input layer to the output layer, with no reverse data feed. In a specific business scenario, the input layer of this model includes three physical items: the initial temperature and mass flow rate of the juice obtained from the feedforward, and the dynamic viscosity, which objectively represents the state of pectin and suspended vesicle solids within the fluid. In the preprocessing logic, all input node data are mapped to the dimensionless interval [0,1] using a minimum-maximum normalization algorithm to eliminate interference from different dimensions in the calculation of network weights. The hidden layer contains eight neurons and uses the ReLU activation function to capture nonlinear features. The output layer contains two nodes, and the corresponding output business physical state is the modulation depth coefficient in the aforementioned control formula. Difference from optimal heat flux matching The dynamic correction reference value.

[0081] S314. To ensure that the neural network model can provide parameters with engineering guidance significance in the control system, this specification discloses its construction and training process for reproduction by those skilled in the art. Network training is conducted using offline supervised learning. The training sample set is derived from the physical property sensor data matrix collected in previous batches of Wogan orange juice processing experiments; the corresponding label data is the optimal manually adjusted data recorded by the system under experimental conditions, where the targeted enzyme inactivation rate meets the process standards and the volatile flavor compound retention is at its highest. and Parameter group.

[0082] In the model training step, the algorithm uses mean squared error as the loss function to evaluate the deviation between the network's predicted output and the labeled data. The optimizer employs gradient descent, calculating the error gradient layer by layer through backpropagation and continuously updating the connection weights and bias terms between nodes according to a set learning rate of 0.01. When the validation error on the continuous test set falls below a preset convergence threshold (usually set to 10), the algorithm continues training. -4 After a certain number of iterations (on a certain scale), training stops, and the network weights are solidified and burned into the controller's non-volatile memory.

[0083] S315, relying on matching parameters dynamically issued by a neural network, generates a target RF output power command that oscillates continuously and synchronously with the flow field phase, according to the aforementioned core coherent control formula. This numerical command is then converted into a pulse-width modulated drive signal with a corresponding duty cycle and transmitted to the next-stage solid-state RF source. Based on this signal, the RF source generates an amplitude-modulated RF wave, typically set to 27.12 MHz, which, after passing through a dynamic impedance matching network, is applied to the transmitting electrode outside the transparent reactor tube. This synergistic control logic enables the RF volumetric heating effect and the flow field cavitation shock wave enzyme-inactivating effect to produce a synergistic physical effect at specific spatiotemporal points, thereby helping to improve the overall quality of the juice under mild macroscopic temperature conditions.

[0084] Upon receiving an electrical signal that is synchronously modulated with the flow field phase, energy is converted into uniform thermal energy within the multiphase fluid. During this stage, the system achieves polarization heat generation and isothermal boundary maintenance through the following steps.

[0085] S316, the high-frequency amplitude-modulated radio frequency wave output from the solid-state RF source is transmitted to the transmitting electrode located on the outside of the transparent reactor tube via a dynamic impedance matching network. In this embodiment, the transparent reactor tube is made of polytetrafluoroethylene (PTFE), which has low dielectric loss and high mechanical strength. The material selection aims to enable the RF electromagnetic wave to penetrate the tube wall with a low attenuation rate and act on the multiphase fluid flowing inside. When an alternating electromagnetic field with a frequency of 27.12 MHz is applied to the Wogan orange juice volume phase, water molecules and other charged groups in the juice, as typical representatives of polar molecules, undergo high-frequency alternating polarization motion under the drive of the alternating high-frequency electric field. At the same time, soluble free ions inside the juice undergo periodic sliding motion with the direction of the electric field. At the microscopic level, the alternating polarization friction of polar molecules and the alternating conduction of ions directly convert electromagnetic energy into macroscopic thermal energy of the multiphase mixture, completing the polarization heat generation process. Since this volumetric heating method does not rely on the traditional solid-phase medium wall conduction, it usually exhibits a faster heating rate and a more uniform internal temperature distribution within the fluid.

[0086] S317. In continuous flow industrial production, due to the boundary layer effect of fluid mechanics, the fluid velocity adjacent to the inner wall of the microwave-transparent reaction tube is usually lower than the fluid velocity at the tube axis. If the boundary heating state is not intervened, this velocity difference will cause a dual thermodynamic contradiction: on the one hand, the low-velocity boundary layer fluid, under the long-term action of a radio frequency electric field, is prone to thermal degradation of the fruit juice components due to localized heat accumulation; on the other hand, the natural heat conduction and dissipation from the microwave-transparent reaction tube wall to the external environment will form a relatively low-temperature zone on the inner side of the tube wall, thereby weakening the enzyme-inactivating effect in that region. Based on the above-mentioned coupling characteristics of fluid mechanics and thermophysical properties, this invention incorporates an isothermal boundary maintenance component encapsulated on the outside of the microwave-transparent reaction tube.

[0087] As a preferred embodiment, the isothermal boundary maintenance component specifically employs a flexible silicone heating strip tightly fitted to the outside of the microwave-transparent reaction tube, or a double-layered jacket structure with constant-temperature thermally conductive oil flowing inside. To minimize contact thermal resistance and prevent localized overheating, a highly thermally conductive insulating medium (such as thermally conductive silicone grease) is typically uniformly applied or filled between the inner surface of the heating component and the outer wall of the microwave-transparent reaction tube, thereby constructing a continuous and uniform radial heat conduction path.

[0088] S318, to maintain the stability of the pipe wall boundary conditions, the controller adjusts the energy output of the isothermal boundary maintenance component based on the real-time boundary temperature collected by the high-frequency dynamic temperature sensor installed on the inner wall of the transparent reaction tube. To avoid the sensor probe from protruding into the flow channel and disrupting the microscopic flow field structure of the fluid, the high-frequency dynamic temperature sensor in this embodiment is specifically a miniature fiber optic grating temperature sensor flush with the inside of the pipe wall or a thin-film patch thermocouple.

[0089] In specific operation control, the controller sets the boundary target control temperature to the aforementioned basic process temperature. Through closed-loop calculation, the controller outputs a corresponding duty cycle drive signal to adjust the power output of the heating elements inside the jacket, so that the solid wall temperature of the microwave-transparent reaction tube and the temperature of the internal fluid body are kept in dynamic equilibrium. This maintenance of the isothermal boundary helps to suppress spontaneous heat dissipation from the tube wall to the external environment to a large extent, causing the normal heat flux between the fluid body and the tube wall to approach zero. For the specific pulse width modulation and allocation of the power supply to the heating elements inside the isothermal jacket, those skilled in the art can configure it according to the conventional thermal closed-loop control principle. Its parameter tuning logic is a well-known technology in the field and will not be elaborated here. By reducing the temperature gradient in the radial direction of the microwave-transparent reaction tube, the juice on each streamline in the flow channel can obtain a more uniform temperature rise process and phase change coherent period during the flow process, thereby providing homogeneous thermodynamic boundary support for the stable enzyme inactivation of materials throughout the flow field.

[0090] After undergoing the aforementioned polarization-induced heat generation and coherent cavitation-induced enzyme inactivation processes, the multiphase mixed fluid is under high pressure and at the set macroscopic process temperature. To safely convert it into an atmospheric pressure finished product while preserving its original flavor, the system performs gradient pressure field gas release and solid-liquid aroma preservation output operations through the following steps.

[0091] S401. If high-pressure gaseous materials are directly discharged into an atmospheric pressure environment, the large pressure gradient within the fluid will typically trigger the instantaneous expansion and boiling of dissolved carbon dioxide. Based on the general principles of chemical mass transfer, this gas escape process will produce a physical effect similar to gas-phase stripping, thereby entraining some low-boiling-point volatile flavor compounds from the juice. Considering the above physical mechanism, a gradient pressure reduction component is configured in the downstream pipeline of the microwave-transparent reaction tube, aiming to smooth the mass transfer kinetics of the gas-liquid separation process through a step-by-step pressure release.

[0092] In S402, the module responsible for performing pressure relief is defined as a gradient pressure reduction control unit in the superordinate summary of the claims. In this embodiment, this unit is specifically implemented using a multi-stage servo back pressure valve group arranged in series. As a preferred embodiment, the system is equipped with first-stage, second-stage, and third-stage servo back pressure valves sequentially along the fluid flow direction. To establish a control closed loop, pressure transmitters are installed in the pipe sections before and after each stage of the back pressure valve to collect the pipeline pressure in real time. The controller calculates the target set pressure value of each stage of the valve based on an internal pressure drop distribution algorithm. In actual operation, this distribution algorithm divides the inlet and outlet pressure difference of the main pipeline into multiple progressively decreasing pressure intervals, and dynamically adjusts the valve core opening of each stage of the back pressure valve through a digital PID algorithm. This gradient pressure field distribution on a spatial scale helps to promote the slow precipitation of carbon dioxide in the form of microbubbles inside the multiphase fluid, thereby reducing the probability of flavor substances being mechanically entrained by the gas.

[0093] S403, while achieving multi-stage pressure reduction in the spatial flow channel, also maintains a low bulk fluid temperature, which is of engineering significance for inhibiting the volatilization of flavor substances. Before entering the back pressure valve assembly, the mixed fluid is typically guided through a forced cooling heat exchange component. Considering that NFC Wogan juice contains a large number of three-dimensional pulp cells, to avoid solid matter accumulation and blockage in the narrow flow channel, this forced cooling heat exchange component specifically adopts a large-channel shell-and-tube heat exchanger or a scraped-plate heat exchanger. Inside this component, the low-temperature circulating cooling medium exchanges heat with the juice material in a counter-current manner. The controller reduces the outlet temperature of the juice material to the refrigeration process temperature range of 5℃~10℃ by adjusting the opening of the proportional regulating valve in the cooling medium circuit. The cooling process reduces the saturated vapor pressure of characteristic aroma substances such as alcohols, esters, and aldehydes in the system, which, from a thermodynamic perspective, is beneficial for preserving the original flavor components of the juice.

[0094] S404, the gas-liquid-solid three-phase mixture after depressurization and cooling is then introduced into the downstream gas-liquid separation component. In this embodiment, the gas-liquid separation component is specifically a cyclone separator with a tangential feed inlet. After the material enters the tank tangentially, under the combined action of centrifugal force and gravity, the liquid phase of juice and the solid phase of pulp cells swirl down along the tank wall and collect at the bottom of the degassing tank. At the same time, the released gaseous carbon dioxide, due to its lower density, gathers towards the central axis and top of the tank.

[0095] S405, to further improve the recovery rate of characteristic flavor substances, a cryogenic reflux condenser is integrated at the top exhaust port of the degassing tank. When carbon dioxide gas carrying a small amount of water vapor and volatile aroma molecules flows through this condenser, the volatile components undergo a transphase transition on the condenser tube wall. To balance condensation efficiency and prevent water vapor from freezing on the tube wall surface due to excessively low temperature, thus blocking the exhaust passage, the circulating refrigerant temperature of this cryogenic reflux condenser is typically controlled within the range of 1℃ to 4℃. Under these boundary conditions, the gaseous components re-liquefy into droplets. These droplets, rich in flavor substances, flow back and drip into the main body of juice at the bottom of the tank under gravity, while the uncondensed carbon dioxide gas is discharged from the system through the exhaust valve. The solid-liquid mixed finished juice, after the above degassing and aroma recovery, is finally pumped by the aseptic output pump at the bottom of the tank to the next stage of aseptic filling, thus completing the material output of the entire processing flow.

[0096] Specific application examples: The freshly extracted NFC Wogan mandarin orange juice, containing pulp, was initially heated to 20°C after initial pressing and coarse filtration. A screw pump controlled the flow rate within the pipeline at a shear force of 30 Pa (far below the pulp rupture threshold). A rotational viscometer measured the dynamic viscosity of the current batch at 45 mPa·s, and a mass flow meter measured the mass flow rate at 500 kg / h.

[0097] The controller is set to a carbon dioxide base ratio of 5%. Due to the high viscosity (45 mPa·s reflects a high solids content and mass transfer resistance), the controller uses a mapping table to obtain a positive compensation coefficient of 1.1. The final calculated target injection rate is 500 × 5% × 1.1 = 27.5 kg / h. The constant flow pump injects liquid CO2 at this flux.

[0098] The plunger pump increases the basic process pressure of the pipeline to 12MPa, and the preheating sleeve precisely raises the basic process temperature to 38℃. At this point (12MPa>7.38MPa, 38℃>31.1℃), CO2 is converted to a supercritical state and, under the action of the static mixer, deeply dissolves and enters the aqueous phase of the juice and the micropores of the pulp.

[0099] Fluid enters a fluid oscillator inside a microwave-transparent reaction tube. The geometry induces alternating deflections of the fluid at a frequency of 1200 Hz. A transient subcritical low-pressure region is generated in the local jet wake, where the pressure drops sharply to 4.5 MPa (below the critical pressure of CO2). The dissolved CO2 expands instantaneously, generating cavitation bubbles with a diameter of approximately 20 μm, which then violently collapse when the pressure rises back to 12 MPa, releasing micro-shock waves that mechanically tear the conformation of pectin methyl esterase (PME).

[0100] The high-frequency dynamic pressure transmitter captures pressure fluctuations at a sampling rate of 10kHz, and the controller extracts the continuous transient phase through Hilbert transform.

[0101] The neural network outputs control parameters based on the initial temperature (20℃), flow rate (500 kg / h), and viscosity (45 mPa·s): modulation depth. ,difference .

[0102] The radio frequency source (27.12MHz) is based on the formula. Output pulse power. The radio frequency power reaches its peak the instant the microbubbles expand and the pulp pores open. The radio frequency electromagnetic waves penetrate and heat without obstruction, instantly compensating for the latent heat absorbed by CO2 vaporization. The macroscopic temperature of the fluid is stabilized at 38℃ (the isothermal boundary components are activated simultaneously).

[0103] The processed juice passes through a three-stage servo back pressure valve, where the pressure is stepped down from 12MPa to 8MPa, then to 4MPa, and finally to atmospheric pressure. Simultaneously, it is cooled to 8°C by a heat exchanger.

[0104] The material enters a hydrocyclone separator, with a top cryogenic reflux condenser set at 2°C. Gases carrying characteristic flavors of Wogan tangerines (such as d-limonene and linalool) condense into droplets and flow back into the juice, while pure CO2 is discharged. Finally, chilled, enzyme-inactivated, and flavorful NFC Wogan tangerine juice is pumped out from the bottom.

[0105] Experimental verification and effect comparison: To verify the advancement of this invention, three comparative experiments were conducted in a laboratory environment using the same batch of NFC Wogan mandarin orange juice: Group A (Traditional Heat Sterilization (HTST)): 90℃, hold for 30 seconds.

[0106] Group B (Ultra-high Pressure Cold Sterilization HPP): 600MPa, pressure held for 3 minutes.

[0107] Group C (System of this invention): Working pressure 12MPa, macroscopic temperature 38℃, radio frequency coherent cavitation treatment.

[0108] Table 1. Comparison of Experimental Data

[0109] Conclusion: Refer to Appendix Figure 3 -Appendix Figure 5 Addressing the issues of traditional NFC juice processing requiring high temperatures for enzyme inactivation, which destroys flavor, and extremely high pressure for preservation, resulting in expensive and energy-intensive equipment, this system induces periodic pressure oscillations across the supercritical CO2 boundary (7.38 MPa) at a median pressure of 12 MPa to release cavitation shock wave mechanical force. It precisely extracts the principal pressure phase to drive radio frequency power through strictly corresponding nonlinear coherent modulation, thereby instantaneously and precisely injecting heat energy during the microbubble expansion period. This time-domain precisely coherent acoustic-electric coupling mechanism efficiently positions energy at the microscopic interface between microorganisms and enzymes. Under the mild macroscopic conditions of only 12 MPa and 38°C, this system not only achieves excellent enzyme inactivation effects approaching the traditional 90°C high temperature but also achieves nutrient and flavor retention rates comparable to 600 MPa ultra-high pressure. Furthermore, it avoids the useless power consumption generated by global fluid boiling and extremely high-pressure compression, reducing the overall energy consumption of the equipment to an extremely low level. This achieves the dual optimization of quality and cost in continuous processing of high-quality NFC juice.

Claims

1. A method for processing NFC Wogan mandarin orange juice, characterized in that, Includes the following steps: The NFC Wogan mandarin orange blend juice is pumped into the pipeline system, and the dynamic dynamic viscosity data and mass flow rate data of the NFC Wogan mandarin orange blend juice are collected as feedforward parameters. The amount of carbon dioxide injected is calculated based on the feedforward parameters. Liquid carbon dioxide is injected into the NFC Wogan mixed juice to form a primary mixture. The primary mixture is pressurized and the temperature is adjusted to reach the basic process pressure and temperature, so that the carbon dioxide is converted into a supercritical state and dissolved and mixed to form a high-pressure three-phase mixed fluid. The high-pressure three-phase mixed fluid is guided to flow through the fluid oscillator to generate fluid self-excited oscillation, alternately forming a transient subcritical low-pressure zone, causing carbon dioxide to expand and undergo phase change, and generating cavitation shock waves during pressure recovery. The transient phase signal of the pressure fluctuation of the high-pressure three-phase mixed fluid is collected simultaneously, and radio frequency electromagnetic waves are emitted to the high-pressure three-phase mixed fluid according to the transient phase signal to perform polarization heat generation and thermal energy compensation. The high-pressure three-phase mixed fluid after polarization heat generation and thermal energy compensation is subjected to gradient decompression and gas-liquid separation to discharge gaseous carbon dioxide and output the finished NFC Wogan juice.

2. The NFC Wogan mandarin orange juice processing method according to claim 1, characterized in that, The NFC Wogan mixed juice comprises a continuous aqueous phase, a suspended pectin phase, and a pulp vesicle solid phase. The carbon dioxide injection amount is calculated based on the feedforward parameters, including the following steps: Obtain a preset basic mass ratio constant, and multiply the mass flow rate data by the basic mass ratio constant to obtain the theoretical basic injection volume; The built-in empirical viscosity compensation mapping relationship is invoked, and the corresponding positive compensation coefficient is extracted based on the dynamic dynamic viscosity data. The positive compensation coefficient increases as the dynamic dynamic viscosity data increases. The theoretical injection amount is multiplied by the positive compensation coefficient to calculate the carbon dioxide injection amount. The constant flow pump is then controlled to inject liquid carbon dioxide into the NFC Wogan mixed juice according to the carbon dioxide injection amount.

3. The NFC Wogan mandarin orange juice processing method according to claim 1, characterized in that, Generating fluid self-excited oscillations and generating cavitation shock waves during pressure recovery includes the following steps: The high-pressure three-phase mixed fluid is passed through a wall-attached jet element, which includes a converging nozzle, a fluid action chamber, a wedge-shaped fluid distributor, and feedback channels arranged symmetrically on both sides. The high-pressure three-phase mixed fluid is accelerated by the converging nozzle to form a main jet. After entering the fluid action chamber, the main jet adheres to the side wall and entrains the surrounding fluid, resulting in a local decrease in static pressure. The entrained and stripped fluid flows back to the root region of the converging nozzle outlet through the symmetrically arranged feedback channels on both sides and laterally impacts the main jet. Combined with the guiding effect of the wedge-shaped fluid divider, the main jet is forced to deflect alternately in the fluid action chamber to form self-excited oscillation and to form a dynamic flow field inside the fluid action chamber. The local real-time cavitation number of the dynamic flow field is calculated in real time; the basic process pressure and the shrinkage ratio of the shrinkage nozzle cross-sectional area are adjusted in a coordinated manner to maintain the local real-time cavitation number within the range of 0.1 to 0.

5.

4. The NFC Wogan mandarin orange juice processing method according to claim 1, characterized in that, Acquiring the transient phase signal of the pressure fluctuation of the high-pressure three-phase mixture includes the following steps: The continuous analog electrical signal of the pressure fluctuation of the high-pressure three-phase mixed fluid is obtained by a high-frequency dynamic pressure transmitter, and then discretized into a digital pressure signal sequence after low-pass anti-aliasing filtering and analog-to-digital conversion. A sliding data window is opened to perform online DC removal on the digital pressure signal sequence. The digital pressure signal sequence after DC removal is used as the real part of the analytical signal. Discrete Hilbert transform is performed on the digital pressure signal sequence to obtain conjugate feature data as the imaginary part of the analytical signal. The four-quadrant arctangent algorithm is invoked to calculate the true transient principal phase distributed in the reference interval from negative pi to positive pi based on the algebraic sign and ratio relationship between the real part and the imaginary part of the analytical signal, thus obtaining the transient phase signal.

5. The NFC Wogan mandarin orange juice processing method according to claim 4, characterized in that, After obtaining the transient phase signal, phase delay compensation is performed, including the following steps: The lead compensation angle is calculated based on the multiplicative relationship between the signal main frequency period obtained from real-time statistics, the inherent transmission delay constant of the data link measured and stored in advance, and the full-cycle radian coefficient. The lead compensation angle is superimposed on the transient principal phase to generate the target phase parameter after time delay compensation correction, which serves as the time anchor point for driving the transmission of the radio frequency electromagnetic wave.

6. The NFC Wogan mandarin orange juice processing method according to claim 5, characterized in that, The control model for polarization heat generation and thermal energy compensation based on the transient phase signal transmitting the radio frequency electromagnetic wave to the high-pressure three-phase mixed fluid includes the following steps: Obtain the preset thermodynamic isothermal compensation power, and obtain the RF power modulation depth coefficient and the optimal heat flow matching phase difference; The nonlinear cosine modulation mathematical formula is called, and the difference between the target phase parameter and the optimal heat flow matching is used as the cosine variable to calculate the cosine value. The cosine value is multiplied by the RF power modulation depth coefficient and then added by one. Finally, it is multiplied by the thermodynamic isothermal compensation power to generate the target RF output power command. The solid-state radio frequency source is controlled to generate an amplitude-modulated radio frequency wave according to the target radio frequency output power command and apply it to the high-voltage three-phase mixed fluid.

7. The NFC Wogan mandarin orange juice processing method according to claim 6, characterized in that, Obtaining the difference between the RF power modulation depth coefficient and the optimal heat flux matching includes the following steps: The initial temperature, mass flow rate data, and dynamic dynamic viscosity data of the NFC Wogan blended juice are obtained through the input layer of the feedforward backpropagation neural network model. After the hidden layer performs nonlinear calculations, the output layer optimizes and outputs the RF power modulation depth coefficient and the optimal heat flow matching phase difference.

8. The NFC Wogan mandarin orange juice processing method according to claim 1, characterized in that, When performing polarization heat generation and thermal energy compensation, isothermal boundary maintenance is performed simultaneously, including the following steps: The high-pressure three-phase mixed fluid flows inside the microwave-transparent reaction tube and receives the radio frequency electromagnetic wave polarization to generate heat. The outer surface of the microwave-transparent reaction tube is covered with an isothermal boundary maintenance component. The real-time boundary temperature of the inner wall of the microwave-transparent reaction tube is collected in real time. Based on the basic process temperature as the target control temperature, a closed-loop calculation is performed to adjust the energy output of the isothermal boundary maintenance component, so that the tube wall temperature of the microwave-transparent reaction tube and the temperature of the high-pressure three-phase mixed fluid are kept in dynamic equilibrium.

9. The NFC Wogan mandarin orange juice processing method according to claim 1, characterized in that, The gradient decompression of the high-pressure three-phase mixed fluid after polarization heat generation and thermal energy compensation includes the following steps: The system controls a multi-stage servo back pressure valve group arranged in series. Based on the pressure difference between the inlet and outlet of the main pipeline and the preset pressure drop distribution algorithm, it calculates the target set pressure value of each stage of back pressure valve and dynamically adjusts the valve core opening of each stage of back pressure valve through a digital PID algorithm to form a spatial step-like pressure reduction. Before the high-pressure three-phase mixed fluid enters the gas-liquid separation process, the high-pressure three-phase mixed fluid is guided to flow through the forced cooling heat exchange component for countercurrent heat exchange, thereby reducing the temperature of the high-pressure three-phase mixed fluid to the refrigeration process temperature range of 5°C to 10°C, and obtaining a gas-liquid-solid three-phase mixture after depressurization and cooling treatment.

10. A method for processing NFC Wogan mandarin orange juice according to claim 9, characterized in that, Gas-liquid separation includes the following steps: The gas-liquid-solid three-phase mixture is tangentially introduced into a hydrocyclone separator to separate the liquid phase fruit juice, the solid phase fruit pulp sacs, and the gaseous carbon dioxide. A cryogenic reflux condenser is installed at the exhaust port of the cyclone separator to control the temperature of the circulating refrigerant in the cryogenic reflux condenser between 1°C and 4°C, so that the volatile aroma components entrained in the gaseous carbon dioxide are condensed and liquefied and flow back to the liquid fruit juice at the bottom of the cyclone separator under the action of gravity.

Images