A method and system for energy-saving and color-protecting quick-frozen fruits and vegetables dehydration

Abstract

The application discloses a quick-frozen fruit and vegetable color protection energy-saving dehydration method and system, belongs to the field of quick-frozen fruit and vegetable pretreatment and energy-saving dehydration, and aims at solving the problems of uneven removal of water attached to the surface and free water between tissues of fruits and vegetables after cleaning, cutting or blanching, quick-frozen block and inconsistent energy consumption control of color protection. The application realizes the technical effects of reducing ice crystal bridge connection, maintaining color and reducing invalid energy consumption through double-domain water estimation, ice bridge and color risk generation, model predictive control and segmented dehydration execution.

Description

Technical Field

[0001] This invention relates to the field of quick-freezing pretreatment and energy-saving dehydration of fruits and vegetables, and particularly to a method and system for color-preserving and energy-saving dehydration of quick-frozen fruits and vegetables. Background Technology

[0002] Blueberries, onions, bell peppers, broccoli, cauliflower, and other fruits and vegetables typically require washing, cutting, or blanching before quick-freezing to remove impurities, improve compatibility with subsequent processing, or stabilize enzyme activity. After these pretreatments, droplets and water films may remain on the surface of the materials, and free water may also be present in porous or exposed areas. If this moisture is not uniformly removed before entering the quick-freezing section, ice crystals can easily form between the materials during freezing, causing particles or lumps to stick together and clump together.

[0003] In existing technologies, dehydration before quick-freezing mainly employs methods such as natural drainage, centrifugal drainage, hot air drying, or fixed-speed dehumidification. Natural drainage is time-consuming and difficult to adapt to continuous production lines; centrifugation or strong mechanical action can easily cause blueberries to break, cauliflower heads to loosen, or damage to the cut surface tissue; while hot air can accelerate dehydration, it can easily lead to dull color, softened tissue, and a decline in taste after rehydration. Different types of fruits and vegetables have significant differences in structure, cut surface, pore size, and water content, making it difficult to ensure a consistent moisture state before quick-freezing by simply controlling time or temperature.

[0004] In addition, traditional dehydration control usually uses equipment running time, air temperature or simple weight changes as the endpoint, lacking a coordinated assessment of surface water film, interstitial free water, ice crystal bridging risk, color damage risk and unit dehydration energy consumption. This results in some batches being over-dehydrated and some batches having too much residual water, which not only affects the quality of quick-freezing but also increases unnecessary energy consumption.

[0005] Therefore, there is a need for a quick-freezing method and system for color preservation, energy saving, and dehydration of fruits and vegetables that can overcome the shortcomings of the existing technologies. Summary of the Invention

[0006] One objective of this invention is to propose a quick-freezing method and system for color protection and energy-saving dehydration of fruits and vegetables. Addressing the problems of uneven removal of surface water and interstitial free water before quick-freezing, easy ice bridging and clumping after quick-freezing, and inconsistency between color protection and energy consumption control in existing technologies, this invention proposes a technical solution that collects raw material state data, generates surface water film index and interstitial free water index, calculates ice crystal bridging risk index and color damage risk, and uses model predictive control to drive the coordinated execution of dew point gradient jet, pulsed vacuum, vibration loosening, heat pump dehumidification, conveying, and color protection micro-atomization mechanisms. This invention has the technical effects of reducing the risk of quick-freezing clumping, maintaining the color of fruits and vegetables, and reducing ineffective dehydration energy consumption.

[0007] This invention provides a method for quick-freezing fruits and vegetables to preserve color and save energy during dehydration, including:

[0008] S1. Spread the washed, cut or blanched fruit and vegetable materials in a single layer, collect data on material structure, moisture, airflow, energy consumption, dehydration, surface temperature and color difference, and determine the spreading thickness, dwell time before freezing, dwell time after pretreatment and material contact probability to form raw material state data.

[0009] S2. Input the raw material state data into the dual-domain moisture estimation unit, and generate the surface water film index and interstitial free water index according to the category parameter set;

[0010] S3. Generate an ice crystal bridging risk index based on the surface water film index, the interstitial free water index, the material contact probability, the spreading thickness, the residence time before freezing, and the airflow data. Generate a color damage risk based on the color difference, surface temperature, and residence time after pretreatment. Generate a unit dehydration energy consumption based on the energy consumption and dehydration data.

[0011] S4. The model prediction and control unit takes the ice crystal bridging risk index as the target to be within the category ice bridging threshold, and takes the color damage risk as the target to be within the color protection threshold and the unit dehydration energy consumption as the constraint to determine the jet parameters, air inlet dew point, vacuum parameters, vibration frequency, conveying speed and color protection micro-mist start and stop signals.

[0012] S5. Drive the dew point gradient jet, vibration loosening, pulse vacuum, heat pump dehumidification, conveying drive and color protection micro-mist mechanism according to the control amount determined in step S4 to remove surface droplets, attached water film and free water between tissues.

[0013] S6. When the ice crystal bridging risk index updated after processing in S5 is within the category ice bridging threshold and the color damage risk is within the color protection threshold, the fruit and vegetable materials are sent into the quick-freezing section.

[0014] Optionally, S1 includes:

[0015] Based on the material category, the particle size sampling rules, flower ball gap sampling rules, cross-section recognition area, near-infrared band group, and dielectric sampling frequency are read from the category parameter library.

[0016] The material structure data includes material type, particle size or flower ball gap and cross-sectional area; the moisture detection data includes bright spot area of ​​surface image, near-infrared moisture response, dielectric response and online weight change; the airflow status data includes inlet dew point, inlet humidity and outlet humidity.

[0017] The energy consumption data includes the electrical energy metering values ​​of the heat pump dehumidification mechanism, pulse vacuum mechanism, vibration loosening mechanism, and conveying drive mechanism, and the dehydration data is determined by the online weight change at adjacent sampling times;

[0018] The pre-freezing dwell time is the dwell time from the completion of single-layer spreading to the expected time of frozen expiration, and is updated according to the conveying speed and processing section length;

[0019] The dwell time after pretreatment is the dwell time from the completion of the last washing, cutting or blanching to the current sampling time;

[0020] The ratio of the area of ​​the bright spot region in the conveyor belt image to the visible area of ​​the material is determined as the bright spot area of ​​the surface image, and the material contact probability is determined based on the spreading thickness, the particle size or the gap between the flower balls, and the distance between the boundaries of adjacent materials in the conveyor belt image.

[0021] Optionally, S2 includes:

[0022] After normalizing the bright spot area of ​​the surface image, the near-infrared moisture response, the increase of the outlet humidity relative to the inlet humidity, and the online weight change, the surface water film index is generated according to the surface water film weight in the category parameter set.

[0023] After normalizing the dielectric response, the online weight change, the increase of the outlet humidity relative to the inlet humidity, and the material surface temperature, the interstitial free water index is generated according to the free water weight in the category parameter set.

[0024] The surface water film index is used to constrain the jet velocity and jet duty cycle, and the interstitial free water index is used to constrain the vacuum amplitude and vacuum holding time.

[0025] Optionally, S3 includes:

[0026] The surface water film index, the interstitial free water index, the material contact probability, the spreading thickness, the dwell time before freezing, and the outlet air humidity are input into the ice bridge risk mapping table to obtain the ice crystal bridging risk index.

[0027] The brightness component, red-green component, and yellow-blue component in the color difference data, the surface temperature of the material, and the dwell time after pretreatment are input into the color constraint mapping table to obtain the color damage risk.

[0028] The ratio of the electrical energy metering value to the dehydration amount data is determined as the unit dehydration energy consumption.

[0029] Optionally, S4 includes:

[0030] The model prediction and control unit uses the ice crystal bridging risk index, color damage risk, actual unit dehydration energy consumption fed back from the previous sampling period, surface water film index, and interstitial free water index as state variables, and uses the category ice bridge threshold, color protection threshold, and energy consumption threshold as constraint boundaries to calculate the control quantity sequence within the preset prediction time domain.

[0031] The first control quantity selected from the control quantity sequence is used as the jet wind speed, jet duty cycle, air inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, conveying speed and color protection micro-mist start / stop signal for the next sampling period;

[0032] The actual unit dehydration energy consumption after execution is fed back to the model prediction and control unit to correct the control quantity sequence for the next sampling period;

[0033] Furthermore, the color-protecting micro-mist start / stop signal is determined based on the color damage risk, the material surface temperature, and the dwell time after pretreatment;

[0034] The color protection warning threshold is derived from the category parameter set and is less than the color protection threshold.

[0035] When the color damage risk reaches the color protection warning threshold but does not exceed the color protection threshold, the ice crystal bridging risk index is within the category ice bridging threshold and the surface water film index is within the surface water film threshold in the category parameter set, the color protection micro-atomization mechanism is activated to apply color protection liquid droplets to the material surface.

[0036] When the surface water film index exceeds the surface water film threshold or the ice crystal bridging risk index is not within the category ice bridging threshold, the color protection micro-fogging mechanism is suppressed and low dew point jet film removal control is prioritized.

[0037] When the risk of color damage exceeds the color protection threshold, the color protection micro-mist mechanism is suppressed, the priority of low dew point dehumidification and cooling control is increased, and the output of the freezing permit signal is prohibited until the risk of color damage obtained by the latest sampling is back within the color protection threshold.

[0038] Optionally, S5 includes:

[0039] The dew point gradient pulse jet mechanism includes a first jet zone and a second jet zone arranged sequentially along the conveying direction;

[0040] The first jet zone outputs a dehumidifying airflow with a temperature between 0 and 10 degrees Celsius and a dew point between -25 and 0 degrees Celsius, and intermittently washes the surface droplets and the attached water film according to the jet duty cycle.

[0041] The second jet zone receives fruit and vegetable materials processed by the pulse vacuum mechanism, resamples the processed fruit and vegetable materials, and obtains the updated value of the surface water film index through the dual-domain moisture estimation unit. The jet wind speed is corrected according to the updated value within the control quantity sequence determined by the model prediction control unit.

[0042] Furthermore, the phases of the pulsed vacuum mechanism and the vibration loosening mechanism are jointly determined by the vacuum holding time, the vacuum amplitude, and the vibration frequency;

[0043] During the interval after the dew point gradient pulse jet mechanism stops airflow, the pulse vacuum mechanism adjusts the processing chamber pressure to a preset vacuum pressure range and maintains the vacuum for the specified time.

[0044] During the pressure recovery phase of the processing chamber, the vibration loosening mechanism turns the material according to the vibration frequency, so that the moisture that has migrated to the surface of the material enters the removal path of the next jet cycle.

[0045] Optionally, S6 includes:

[0046] When the ice crystal bridging risk index and the color damage risk obtained from the latest sampling before the freezing permit is issued are both within the corresponding threshold in two consecutive sampling periods, the freezing permit signal is output and the fruit and vegetable materials are sent into the quick-freezing section.

[0047] The heat pump dehumidification mechanism adjusts the compressor frequency, evaporator bypass ratio, and reheat capacity based on the inlet dew point, the outlet humidity, the exhaust temperature collected by the exhaust temperature sensor, and the actual unit dehydration energy consumption generated by the energy consumption data and the dehydration amount data. The adjusted actual unit dehydration energy consumption is then fed back to the model prediction control unit.

[0048] The category parameter set includes a color protection warning threshold and a surface water film threshold. The category parameter set is updated based on the previous batch's export moisture content, feedback on clumping after quick-freezing, feedback on rehydrated taste, and feedback on color difference. The updated category ice bridge threshold, color protection warning threshold, color protection threshold, energy consumption threshold, surface water film threshold, surface water film weight, and free water weight are provided to the dual-domain moisture estimation unit, risk and energy consumption generation module, and model prediction control unit for subsequent batches.

[0049] On the other hand, the present invention also provides a quick-freezing fruit and vegetable color-preserving and energy-saving dehydration system, comprising:

[0050] The spreading and acquisition module is used to spread the washed, cut, or blanched fruits and vegetables into a single layer and generate the raw material state data; the dual-domain moisture estimation module is used to generate the surface water film index and the interstitial free water index; the risk and energy consumption generation module is used to generate the ice crystal bridging risk index, the color damage risk, and the unit dehydration energy consumption, and feeds back the actual unit dehydration energy consumption to the predictive control module and the heat pump dehumidification mechanism; the predictive control module is used to determine the jet wind speed, the jet duty cycle, the inlet dew point, and the vacuum amplitude. The system includes: the vacuum holding time, the vibration frequency, the conveying speed, and the color-protecting micro-mist start / stop signal; a segmented execution module, comprising a dew point gradient pulse jet mechanism, a vibration loosening mechanism, a pulse vacuum mechanism, a heat pump dehumidification mechanism, a conveying drive mechanism for receiving the conveying speed, and a color-protecting micro-mist mechanism for receiving the color-protecting micro-mist start / stop signal; and a quick-freezing connection module, used to send the fruit and vegetable materials into the quick-freezing section when the ice crystal bridging risk index and the color damage risk, which are both resampled and updated after processing by the segmented execution module, are within the corresponding thresholds.

[0051] The beneficial effects of this invention are:

[0052] 1. By using the surface water film index and the interstitial free water index as dehydration control parameters, the removal requirements of surface-attached water and interstitial free water can be distinguished. This allows dehydration control to no longer rely solely on a single total moisture content or a fixed treatment time, thereby improving the consistency of moisture status of different types of fruits and vegetables before they enter the quick-freezing stage.

[0053] 2. By incorporating the ice crystal bridging risk index, color damage risk, and unit dehydration energy consumption into the model's predictive control constraints, it is possible to reduce the risk of ice crystal bridging and agglomeration while avoiding color degradation caused by excessive hot air or prolonged residence, and to reduce unnecessary dehydration energy consumption.

[0054] 3. Through the coordinated control of dew point gradient jet, pulsed vacuum, vibration loosening, heat pump dehumidification, conveying drive and color protection micro-mist mechanism, it can remove surface droplets, attached water film and interstitial free water in stages under low temperature and feedback conditions, thereby improving the looseness, color stability and rehydrated taste of quick-frozen products. Attached Figure Description

[0055] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0056] Figure 1 This is a flowchart of a quick-freezing method for color preservation and energy-saving dehydration of fruits and vegetables.

[0057] Figure 2This is a flowchart of the rolling optimization process of the model prediction control unit in S4 of the present invention. Detailed Implementation

[0058] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0059] refer to Figures 1-2 A method for color preservation and energy-saving dehydration of quick-frozen fruits and vegetables, comprising:

[0060] S1. Spread the washed, cut or blanched fruit and vegetable materials in a single layer, collect data on material structure, moisture, airflow, energy consumption, dehydration, surface temperature and color difference, and determine the spreading thickness, dwell time before freezing, dwell time after pretreatment and material contact probability to form raw material state data.

[0061] S2. Input the raw material state data into the dual-domain moisture estimation unit, and generate the surface water film index and interstitial free water index according to the category parameter set;

[0062] S3. Generate an ice crystal bridging risk index based on the surface water film index, the interstitial free water index, the material contact probability, the spreading thickness, the residence time before freezing, and the airflow data. Generate a color damage risk based on the color difference, surface temperature, and residence time after pretreatment. Generate a unit dehydration energy consumption based on the energy consumption and dehydration data.

[0063] S4. The model prediction and control unit takes the ice crystal bridging risk index as the target to be within the category ice bridging threshold, and takes the color damage risk as the target to be within the color protection threshold and the unit dehydration energy consumption as the constraint to determine the jet parameters, air inlet dew point, vacuum parameters, vibration frequency, conveying speed and color protection micro-mist start and stop signals.

[0064] S5. Drive the dew point gradient jet, vibration loosening, pulse vacuum, heat pump dehumidification, conveying drive and color protection micro-mist mechanism according to the control amount determined in step S4 to remove surface droplets, attached water film and free water between tissues.

[0065] S6. When the ice crystal bridging risk index updated after processing in S5 is within the category ice bridging threshold and the color damage risk is within the color protection threshold, the fruit and vegetable materials are sent into the quick-freezing section.

[0066] In this specific embodiment, S1 includes:

[0067] The cleaned, cut or blanched fruits and vegetables are fed into the breathable conveyor belt by the feeding and leveling mechanism. The materials are spread into a single layer by the height limiting scraper, the dispersing roller and the side alignment guide plate. The single layer spread state means that the outlines of adjacent materials do not overlap or obstruct each other in the conveyor belt image, and the spread thickness measured by the laser displacement sensor is not greater than the upper limit of the single layer thickness of the corresponding material category.

[0068] The controller reads particle size sampling rules, flower head gap sampling rules, cross-section recognition area, near-infrared band group, and dielectric sampling frequency from the category parameter library according to the material category. The category parameter library is a pre-calibrated data table. The index field of the data table is the material category. The parameter fields of the data table include single-layer thickness upper limit, particle size recognition threshold, flower head gap grayscale threshold, cross-section color threshold, near-infrared band group, dielectric sampling frequency, surface bright spot grayscale threshold, and contact distance threshold. The particle size sampling rules are used to determine the equivalent circle diameter within the outer contour of the material. The flower head gap sampling rules are used to identify the dark seam area between flower buds in the flower head material image and calculate the distance between adjacent dark seams. The cross-section recognition area is used to limit the color and texture recognition range of the exposed cross-section in the segmented material image. The near-infrared band group is used to collect the reflection response related to water absorption. The dielectric sampling frequency is used to collect the dielectric response related to free water in the tissue.

[0069] The image acquisition device acquires visible light images above the conveyor belt. The controller obtains the visible area of ​​the material based on the outer contour of the material and identifies the areas in the image with gray values ​​higher than the surface bright spot gray value threshold and continuous pixel areas larger than the noise removal area as bright spot areas. The ratio of the bright spot area to the visible area of ​​the material is determined as the bright spot area of ​​the surface image.

[0070] Near-infrared sensors collect near-infrared moisture responses according to the near-infrared band groups in the category parameter library; dielectric sensors collect dielectric responses according to the dielectric sampling frequency in the category parameter library; online weighing units record the weight of materials carried by the conveyor belt according to the sampling period; inlet dew point sensors, inlet humidity sensors, and outlet humidity sensors collect inlet dew point, inlet humidity, and outlet humidity respectively; power metering units record the power metering values ​​of the heat pump dehumidification mechanism, pulse vacuum mechanism, vibration loosening mechanism, and conveying drive mechanism respectively; infrared temperature sensors collect the surface temperature of materials; and online colorimeters collect color difference data, which includes luminance components, red-green components, and yellow-blue components.

[0071] The amount of water removed is determined by the change in online weight between adjacent sampling times. The controller uses the absolute value of the difference between the online weight at the current sampling time and the online weight at the previous sampling time as the amount of water removed in the current sampling cycle.

[0072] The pre-freezing dwell time is the time between the completion of single-layer spreading and the expected freezing permit output time. The controller updates the expected freezing permit output time on a rolling basis according to the conveying speed and the length of the processing section.

[0073] The dwell time after pretreatment is the time between the completion of the last washing, cutting or blanching and the current sampling time;

[0074] The controller determines the material contact probability based on the spreading thickness, particle size or ball gap, and the distance between adjacent material boundaries in the conveyor belt image. The calculation formula is as follows:

[0075] ;

[0076] in, The probability of material contact ranges from 0 to [value missing]. The spreading thickness measured by the laser displacement sensor during the current sampling period. This refers to the maximum single-layer thickness corresponding to the current material category in the category parameter library. The feature size corresponding to the current material category, when the material structure data uses particle size. For particle size, when the material structure data uses the flower ball gap... For the gaps between the flower balls, This represents the average distance between adjacent material boundaries in the conveyor belt image. This represents the number of adjacent material pairs whose boundary distance is less than the contact distance threshold. This represents the total number of adjacent material pairs participating in the statistics within the current image frame. and The contact probability weights are pre-calibrated in the category parameter library and satisfy the following conditions: Used to limit the calculation result to a range not exceeding 1;

[0077] The controller timestamps the material category, particle size or flower ball gap, cross-sectional area, bright spot area of ​​surface image, near-infrared moisture response, dielectric response, online weight change, inlet dew point, inlet humidity, outlet humidity, power metering value, dehydration data, material surface temperature, color difference data, spreading thickness, residence time before freezing, residence time after pretreatment, and material contact probability according to the same sampling time, and encapsulates them as raw material status data.

[0078] In this specific embodiment, S2 includes:

[0079] The dual-domain moisture estimation unit receives raw material status data and calls the category parameter set according to the current material category. The category parameter set is a parameter table stored in the controller after being calibrated by the same production line. Its index field is the material category, and its parameter fields include the lower and upper limits of the bright spot area of ​​the surface image, the lower and upper limits of the near-infrared moisture response, the lower and upper limits of the humidity increment, the lower and upper limits of the online weight change, the lower and upper limits of the dielectric response, the lower and upper limits of the material surface temperature, the surface water film weight, and the free water weight.

[0080] The dual-domain moisture estimation unit first filters the input quantities for validity, removing data whose timestamps are not aligned with the current sampling period. It then performs amplitude limiting and normalization processing on the bright spot area of ​​the surface image, the near-infrared moisture response, the increment of the outlet humidity relative to the inlet humidity, the online weight change, the dielectric response, and the material surface temperature. The increment of the outlet humidity relative to the inlet humidity is the difference between the outlet humidity and the inlet humidity in the current sampling period. The online weight change is the absolute value of the difference between the online weight at the current sampling time and the online weight at the previous sampling time. The amplitude limiting and normalization processing refers to subtracting the corresponding lower limit of the category parameter set from the input quantity to be processed, and then dividing by the difference between the corresponding upper limit and the corresponding lower limit, and limiting the result to the range of 0 to 1.

[0081] After normalization, the dual-domain water estimation unit uses a dual-domain weighted estimation model to calculate the surface water film index and the interstitial free water index, respectively. The calculation formula is as follows:

[0082] ,

[0083] ;

[0084] in, The surface water film index ranges from 0 to 1. A higher value indicates a more pronounced surface water droplet and water film on the material. The interstitial free water index ranges from 0 to 1. A higher value indicates that more free water can migrate within the material's cross-section, pores, or intergranular spaces. The area of ​​the bright spot in the normalized surface image. The normalized near-infrared moisture response. This represents the normalized increase in outlet humidity relative to inlet humidity. The normalized online weight change, The normalized dielectric response, The normalized online weight change, This represents the normalized increase in outlet humidity relative to inlet humidity. This is the normalized surface temperature of the material. and The surface water film weights in the category parameter set correspond to the bright spot area in the surface image, near-infrared moisture response, humidity increment, and online weight change, respectively, and are all greater than or equal to 0, and satisfy the following conditions: and The free water weights in the category parameter set correspond to dielectric response, online weight change, humidity increment, and material surface temperature, respectively, and are all greater than or equal to 0, and satisfy the following conditions: ;

[0085] In this embodiment, the dual-domain weighted estimation model is constructed by establishing surface water film calibration samples and interstitial free water calibration samples according to material categories. The surface water film calibration samples consist of bright spot area of ​​surface image, near-infrared moisture response, humidity increment and online weight change. The interstitial free water calibration samples consist of dielectric response, online weight change, humidity increment and material surface temperature. The controller determines the corresponding weights based on the correlation coefficients between each input quantity in the calibration samples and the surface water content obtained by manual weighing and the interstitial free water content obtained by centrifugation. The weights within the same index are normalized to a sum of 1, thereby obtaining the surface water film weight and free water weight corresponding to the current material category.

[0086] The dual-domain moisture estimation unit writes the calculated surface water film index and interstitial free water index into the state cache of the current sampling period. The surface water film index serves as a state variable for the subsequent model prediction control unit to constrain the jet velocity and jet duty cycle. When the surface water film index increases, the priority of low dew point jet defilm removal control is increased. The interstitial free water index serves as a state variable for the subsequent model prediction control unit to constrain the vacuum amplitude and vacuum holding time. When the interstitial free water index increases, the priority of pulsed vacuum water release control is increased. The surface water film index, interstitial free water index, and raw material state data are then sent to the risk and energy consumption generation module.

[0087] In this specific embodiment, S3 includes:

[0088] The risk and energy consumption generation module receives the surface water film index. Interstitial free water index And receive the material contact probability after timestamp alignment in step S1. Spreading thickness , Pre-freezing residence time, exhaust air humidity, color difference data, material surface temperature, power metering value, and dehydration data;

[0089] The risk and energy consumption generation module first reads the ice bridge risk mapping table, color constraint mapping table, thickness normalization upper limit, pre-freezing residence time normalization upper limit, outlet humidity normalization upper and lower limits, color difference normalization upper and lower limits, material surface temperature normalization upper and lower limits, and pre-treatment residence time normalization upper limit from the category parameter set according to the current material category.

[0090] The ice bridge risk mapping table is a piecewise linear mapping table pre-calibrated according to the current material category. Its input fields include surface water film index, interstitial free water index, material contact probability, normalized value of spreading thickness, normalized value of residence time before freezing, and normalized value of outlet air humidity. Its output field is the ice crystal bridging risk index, which is used to characterize the probability level of ice crystal bridging between adjacent materials due to residual moisture after the current material enters the quick-freezing section.

[0091] The color constraint mapping table is a piecewise linear mapping table pre-calibrated according to the current material category. Its input fields include normalized values ​​of brightness component, red-green component, yellow-blue component, material surface temperature, and residence time after pretreatment. Its output field is color damage risk, which is used to characterize the degree of browning, loss of green color, dullness, or overall color deterioration of the current material caused by temperature, residence time, and color difference shift.

[0092] The ice bridge risk mapping table is constructed by recording the surface water film index, interstitial free water index, material contact probability, spreading thickness, residence time before freezing, and exhaust humidity in the calibration batch of the current material category, and calculating the agglomeration rate after quick freezing. The agglomeration rate is normalized to the calibration value of the ice crystal bridging risk index. Then, piecewise linear nodes are established based on the monotonic relationship between each input quantity and the calibration value, and linear interpolation is used between the nodes.

[0093] The color constraint mapping table is constructed by recording the brightness component, red-green component, yellow-blue component, material surface temperature and dwell time after pretreatment in the calibration batch of the current material category, and measuring the comprehensive color difference shift and sensory color protection level after treatment. The comprehensive color difference shift and sensory color protection level are converted into calibration values ​​of color damage risk. Then, piecewise linear nodes are established based on the monotonic relationship between each input quantity and the calibration value, and linear interpolation is used between the nodes.

[0094] In this embodiment, both the ice bridge risk mapping table and the color constraint mapping table are executed in the controller in a weighted piecewise linear form of normalized input quantities. When the sampled input falls between two adjacent calibration nodes, the controller obtains the corresponding risk increment according to the linear ratio of the adjacent nodes, and limits the weighted sum of each risk increment to the range of 0 to 1.

[0095] The risk and energy consumption generation module generates the ice crystal bridging risk index, color damage risk, and unit dehydration energy consumption according to the following formula:

[0096] ,

[0097] ,

[0098] ;

[0099] in, This is the ice crystal bridging risk index, with a value ranging from 0 to 1. The higher the value, the higher the risk of ice crystal bridging and clumping after quick-freezing. This represents the risk of color damage, with a value ranging from 0 to 1. A higher value indicates a higher risk of color deterioration. Energy consumption per unit of dehydration The surface water film index, The inter-organ free water index, This represents the probability of material contact. To spread the thickness The normalized value relative to the upper limit of thickness normalization, This is the normalized value of the pre-freezing dwell time relative to the normalized upper limit of the pre-freezing dwell time. The normalized value is the outlet air humidity obtained by normalizing the upper and lower limits of the outlet air humidity. This refers to the normalized value of the luminance component in the color difference data, obtained by normalizing the luminance component according to its upper and lower limits. These are the normalized values ​​of the red and green components in the color difference data, obtained by normalizing the red and green components according to their upper and lower limits. These are the normalized values ​​of the yellow and blue components in the color difference data, obtained by normalizing the yellow and blue components according to their upper and lower limits. This is the normalized value of the material surface temperature obtained by normalizing the upper and lower limits of the material surface temperature. This is the normalized value of the post-preprocessing dwell time relative to the normalized upper limit of the post-preprocessing dwell time. and The risk weights in the ice bridge risk mapping table correspond to the surface water film index, interstitial free water index, material contact probability, normalized spread thickness, normalized residence time before freezing, and normalized outlet air humidity, respectively, and are all greater than or equal to 0, and satisfy the following conditions: and The risk weights in the color constraint mapping table correspond to the normalized values ​​of the luminance component, red-green component, yellow-blue component, material surface temperature, and post-treatment residence time, respectively, and are all greater than or equal to 0, and satisfy the following conditions: This is a limiting function used to restrict the result within the parentheses to the range of 0 to 1. This represents the metering increment of electrical energy for the heat pump dehumidification unit during the current sampling period. This represents the incremental electrical energy measurement of the pulse vacuum mechanism within the current sampling period. This represents the incremental electrical energy consumption of the vibration-assisted loosening mechanism during the current sampling period. This represents the metering increment of electrical energy for the drive mechanism during the current sampling period. This refers to the amount of water removed, determined by the online weight changes at adjacent sampling times within the current sampling period. To prevent the generation of a positive constant for division by zero when the dehydration amount data is zero;

[0100] At the end of each sampling period, the risk and energy consumption generation module writes the ice crystal bridging risk index, color damage risk, and unit dehydration energy consumption into the state cache. It also sends the ice crystal bridging risk index to the model prediction control unit as a comparison object for the category ice bridging threshold, sends the color damage risk to the model prediction control unit as a comparison object for the color protection threshold, and sends the unit dehydration energy consumption to the model prediction control unit and the heat pump dehumidification mechanism as feedback objects for energy consumption threshold and heat pump dehumidification adjustment.

[0101] In this specific embodiment, S4 includes:

[0102] The model predicts and controls the ice crystal bridging risk index. Risk of color damage The actual unit dehydration energy consumption fed back from the previous sampling period and receive the surface water film index Interstitial free water index This constitutes the state quantity for the current sampling period, and the state quantity is... ;

[0103] The model prediction control unit reads the category icing threshold from the category parameter set according to the current material category. Color protection warning threshold Color protection threshold Energy consumption threshold Surface water film threshold The upper and lower limits of jet velocity, jet duty cycle, inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, and conveying speed.

[0104] In this embodiment, the sampling period is 10s, the preset prediction time domain is 6 sampling periods, and the model prediction control unit calculates a control quantity sequence covering the next 60s in each sampling period. Each set of control quantities in the control quantity sequence includes jet wind speed, jet duty cycle, air inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, conveying speed and color protection micro-mist start and stop signal.

[0105] The model prediction control unit uses a linear incremental prediction model to predict the ice crystal bridging risk index, color damage risk, and unit dehydration energy consumption for each future sampling period. The linear incremental prediction model is constructed from the calibration batch data of the current material category. During construction, the jet wind speed, jet duty cycle, inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, and conveying speed are changed within the allowable range of the equipment. The corresponding changes in the ice crystal bridging risk index, color damage risk, and unit dehydration energy consumption are recorded. The incremental coefficient of each control quantity on risk and energy consumption is obtained through least squares identification. The incremental coefficient and the actual unit dehydration energy consumption fed back from the previous sampling period are written into the prediction model parameter table of the current material category.

[0106] The model predictive control unit determines the control quantity sequence according to the following formula:

[0107] ,

[0108] st ;

[0109] in, The objective function value of the model predicts the control unit. For the preset prediction time domain, from the first future sampling period to the... A sequence of control values ​​for a future sampling period. The number of sampling periods in the preset prediction time domain is set to 6 in this embodiment. To predict the sampling period number in the time domain, For the first A control vector for each future sampling period. The parameters include, in order, jet velocity, jet duty cycle, inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, conveying speed, and color-protecting micro-mistification start / stop signal. For the first Control vector for each sampling period This represents the lower limit vector of each control quantity in the category parameter set. This represents the upper limit vector of each control quantity in the category parameter set. For the predicted first Ice crystal bridging risk index for a future sampling period For the predicted first Risk of color damage in future sampling periods For the predicted first Energy consumption per unit of dehydration for each future sampling cycle. For product category ice bridge threshold, For color protection threshold, Energy consumption threshold Assigning risk penalty weights to ice bridges. Assigning a weight to the color risk penalty, As energy consumption penalty weight, To control the change in the quantity, the penalty weight is... , and All are non-negative numbers from the category parameter set. This represents the positive part of the function; the value inside the parentheses is taken when the value is greater than 0, and 0 when the value inside the parentheses is less than or equal to 0. The squared norm of the control vector represents the change in the control quantity vector, and is used to suppress abrupt changes in the control quantity between adjacent sampling periods;

[0110] After the model prediction control unit completes the solution, it selects the first control quantity from the control quantity sequence as the jet wind speed, jet duty cycle, air inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, conveying speed and color protection micro-mistification start / stop signal for the next sampling period, and sends it to the dew point gradient pulse jet mechanism, pulse vacuum mechanism, vibration loosening mechanism, conveying drive mechanism and color protection micro-mistification mechanism.

[0111] The start / stop signal for color protection micro-mist is determined by the model prediction and control unit after solving the continuous control quantity. The decision is based on the risk of color damage, material surface temperature, and residence time after pretreatment. This decision uses color protection early warning thresholds from the product category parameter set. Color protection threshold and surface water film threshold ,and ;

[0112] Risk of color damage Reaching the color protection warning threshold And it did not exceed the color protection threshold. Ice Crystal Bridge Risk Index At the category ice bridge threshold Inner and surface water film index At the surface water film threshold When the material surface temperature and the residence time after pretreatment are both within the allowable range of color protection specified in the category parameter set, the model prediction control unit sets the color protection micro-mist start / stop signal to the start state, so that the color protection micro-mist mechanism applies color protection liquid droplets to the material surface;

[0113] When the surface water film index Exceeding the surface water film threshold Or the risk index of ice crystal bridge Not at the category ice bridge threshold When the time is within the specified range, the model prediction control unit will set the color protection micro-mist start / stop signal to the suppressed state and increase the priority of low dew point jet film removal control in the objective function, so that the jet wind speed and jet duty cycle in the next sampling period are increased within the corresponding upper limit.

[0114] Risk of color damage Exceeding the color protection threshold At this time, the model prediction control unit sets the color protection micro-mist start / stop signal to a suppressed state, increases the priority of low dew point dehumidification and cooling control, prohibits the output of freezing permit signals, and continuously updates the color damage risk in subsequent sampling cycles until the latest sampled color damage risk is obtained. Back to the color protection threshold within;

[0115] The actual unit dehydration energy consumption after execution is fed back to the model prediction control unit by the risk and energy consumption generation module. The model prediction control unit replaces the energy consumption status of the previous sampling period in the prediction model with the actual unit dehydration energy consumption and corrects the control quantity sequence of the next sampling period.

[0116] In this specific embodiment, S5 includes:

[0117] The segmented execution module receives jet wind speed, jet duty cycle, inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, conveying speed, and color protection micro-mistification start / stop signals, and drives the dew point gradient pulse jet mechanism, pulse vacuum mechanism, vibration loosening mechanism, second jet zone, heat pump dehumidification mechanism, conveying drive mechanism, and color protection micro-mistification mechanism to operate in coordination according to the conveying direction.

[0118] The dew point gradient pulse jet mechanism includes a first jet zone and a second jet zone arranged sequentially along the conveying direction. The first jet zone is located before the pulse vacuum mechanism, and the second jet zone is located after the pulse vacuum mechanism. Both the first and second jet zones consist of a low dew point air supply duct, a jet nozzle array, a wind speed regulating valve, a dew point regulating valve, and a pulse on / off valve. The jet nozzle array faces the single layer of spread material on the breathable conveyor belt, and the distance between the nozzle outlet and the upper surface of the material is fixed by the lifting bracket within the spray distance range specified in the category parameter set.

[0119] The first jet zone receives the dehumidified airflow output from the heat pump dehumidification unit and controls the temperature of the dehumidified airflow at a certain level. to Control the intake dew point at to The lower the inlet dew point, the lower the moisture content of the airflow, and the stronger its ability to remove surface droplets and adhering water films.

[0120] The first jet zone intermittently scours the material according to the jet duty cycle, which is the ratio of the time the pulse on / off valve is in the open state within one jet cycle to the total time of the jet cycle. The jet cycle is synchronously issued by the model prediction control unit according to the sampling cycle. During the opening phase, the first jet zone impacts the material surface with jet wind speed, causing surface droplets to be blown away from the protruding parts and adjacent contact parts of the material, making the attached water film thinner and migrating towards the material edge, cross-sectional boundary or flower ball gap outlet. During the closing phase, the air outlet is stopped to reduce the surface temperature rise and energy consumption caused by continuous scouring, and to provide a pressure switching interval for the pulse vacuum mechanism to enter the water release phase.

[0121] The pulse vacuum mechanism includes a sealed processing chamber, a feed airlock, a discharge airlock, a vacuum pump, a pressure regulating valve, and a pressure sensor. The vacuum amplitude is the amount by which the pressure inside the sealed processing chamber decreases relative to atmospheric pressure, and the vacuum holding time is the time during which the sealed processing chamber maintains a preset vacuum pressure range after reaching it.

[0122] During the interval after the dew point gradient pulse jet mechanism stops airflow, the feed airlock and the discharge airlock are closed. The vacuum pump adjusts the pressure of the sealing chamber to the preset vacuum pressure range according to the vacuum amplitude and maintains the vacuum holding time, so that the free interstitial water in the cut surface, pores and flower ball gaps migrates to the material surface under the action of pressure difference.

[0123] The phase of the pulse vacuum mechanism and the vibration loosening mechanism is determined by the vacuum holding time, vacuum amplitude and vibration frequency. Specifically, the vibration loosening is stopped during the vacuum holding stage to maintain the stable spreading state of the material in the sealed processing chamber. During the pressure recovery stage, the vibration loosening mechanism is activated. The vibration loosening mechanism drives the ventilated conveyor belt or loosening lever to generate periodic small-amplitude movements according to the vibration frequency, so that the material is turned over, loosened and the contact surface is renewed, so that the moisture that has migrated to the surface of the material by the pulse vacuum is exposed in the subsequent jet path.

[0124] After the pressure recovery phase, the conveying drive mechanism feeds the material into the second jet zone according to the conveying speed. In the second jet zone, the fruit and vegetable material, after pulse vacuum and vibration loosening treatment, is resampled using visible light images, near-infrared sensors, humidity sensors, and an online weight unit. The resampled data is then sent to the dual-domain moisture estimation unit to obtain an updated value for the surface water film index. Within the control quantity sequence determined by the model prediction control unit, the second jet zone corrects the jet velocity based on the updated value of the surface water film index. The correction relationship is as follows:

[0125] ;

[0126] in, This is the corrected jet velocity for the second jet region. The jet velocity output in step S4 is the jet velocity for the current sampling period. For the jet correction coefficient in the category parameter set, The updated value of the surface water film index is obtained by resampling the second jet region and using the dual-domain moisture estimation unit. For the surface water film threshold in the category parameter set, This is a positive part function; it takes the value when the value inside the parentheses is greater than 0, and takes the value when the value inside the parentheses is less than or equal to 0. The lower limit of the allowable jet wind speed within the range of the control quantity sequence determined by the model predictive control unit. The upper limit of the allowable jet wind speed within the range of the control quantity sequence determined by the model predictive control unit. This is a limiting function used to restrict the calculation result within the parentheses to a certain value. to between;

[0127] When the updated value of the surface water film index is higher than the surface water film threshold, the second jet zone increases the jet velocity within an allowable range to remove surface moisture released by pulsed vacuum and exposed by vibration. When the updated value of the surface water film index is not higher than the surface water film threshold, the second jet zone maintains the step. The output jet velocity should be adjusted to avoid wasted energy.

[0128] The heat pump dehumidification mechanism maintains the inlet dew point during operation in the first and second jet zones, and dehumidifies and reheats the exhaust air after recovering the sensible and latent heat, so that the output airflow meets the low temperature and low dew point conditions.

[0129] The color-protecting micro-atomizing mechanism only works when the color-protecting micro-atomizing start / stop signal output in step S4 is in the start state. Its atomizing nozzle is located after the second jet zone and sprays color-protecting liquid droplets toward the material surface. The application of the color-protecting liquid droplets avoids the opening phase of the first and second jet zones to prevent the color-protecting liquid from being blown away immediately. At the same time, the atomizing pump and atomizing valve are closed when the color-protecting micro-atomizing start / stop signal is in the suppression state.

[0130] Through the segmented execution process of removing surface droplets and attached water films in the first jet zone, causing free water in the tissue to migrate outward through pulsed vacuum, loosening and renewing the material contact surface through vibration in the pressure recovery stage, removing migrated moisture in the second jet zone, and maintaining low dew point airflow through heat pump dehumidification, surface droplets, attached water films, and free water in the tissue sequentially enter the removable path, and the processed material is sent to the subsequent resampling location.

[0131] In this specific embodiment, S6 includes:

[0132] The quick-freezing connection module receives the raw material state data obtained by resampling after processing in step S5, and regenerates the surface water film index, interstitial free water index, ice crystal bridging risk index and color damage risk according to the processing methods of steps S2 and S3. The resampling position is set after the second jet zone and the color protection micro-mist mechanism and before the quick-freezing section inlet, so that the state judged by the quick-freezing connection module is the final state of the material before it is actually frozen.

[0133] The quick-freezing module reads the category ice bridge threshold from the category parameter set according to the current material category. and color protection threshold The model prediction control unit reads the freezing permit prohibition status generated in step S4. When the risk of color damage exceeds the color protection threshold, the freezing permit prohibition status remains effective, the conveying drive mechanism reduces the conveying speed and keeps the material in the low dew point dehumidification and cooling control path.

[0134] When the ice crystal bridging risk index and color damage risk obtained from the latest sampling before the freezing permit is issued are both within the corresponding thresholds in two consecutive sampling periods, and the freezing permit prohibition state has been lifted, the quick-freezing connection module outputs the freezing permit signal and controls the conveying drive mechanism to send the fruit and vegetable materials into the quick-freezing section.

[0135] The following logic is used to determine whether two consecutive sampling periods are in use:

[0136] ;

[0137] in, The current sampling time The signal for permission to enter the freezer. This indicates that fruits and vegetables are permitted to be sent to the quick-freezing section. It indicates that fruits and vegetables are prohibited from being sent into the quick-freezing section. This is an indicator function that returns 1 if all conditions within the parentheses are true, otherwise returns 0. After processing in step S5, at the current sampling time The ice crystal bridging risk index obtained from resampling, After processing in step S5, at the current sampling time Risk of color damage from resampling The previous sampling time The ice crystal bridging risk index obtained from resampling The previous sampling time Risk of color damage from resampling The sampling period is 10 seconds. The category ice bridge threshold corresponding to the current material category. The color protection threshold corresponding to the current material category. Entry into the frozen state is prohibited. This indicates that the model predicts the control unit is preventing the output of the freezing permit signal because the risk of color damage exceeds the color protection threshold. This indicates that the ban on entry permits has been lifted;

[0138] During the freezing permit determination process, the heat pump dehumidification mechanism continues to receive the inlet dew point output in step S4, as well as the outlet humidity, the exhaust temperature collected by the exhaust temperature sensor, and the actual unit dehydration energy consumption generated by the risk and energy consumption generation module based on energy consumption data and dehydration amount data. The heat pump dehumidification mechanism uses the inlet dew point as the dehumidification target, the outlet humidity as the moisture load feedback, the exhaust temperature as the heat recovery and reheat control feedback, and the actual unit dehydration energy consumption as the energy saving feedback to adjust the compressor frequency, evaporator bypass ratio, and reheat capacity.

[0139] When the outlet air humidity is higher than the target range for the outlet air humidity corresponding to the current material category and the actual unit dehydration energy consumption does not exceed the energy consumption threshold, the heat pump dehumidification mechanism increases the compressor frequency and reduces the evaporator bypass ratio to reduce the inlet air dew point and enhance the dehumidification capacity.

[0140] When the outlet humidity is within the target range but the actual unit dehydration energy consumption exceeds the energy consumption threshold, the heat pump dehumidification mechanism reduces the compressor frequency, increases the evaporator bypass ratio, and reduces reheat to reduce ineffective dehumidification energy consumption.

[0141] When the exhaust temperature is lower than the lower limit of the exhaust temperature corresponding to the current material category, the heat pump dehumidification mechanism increases the reheat capacity to maintain the inlet air temperature of the first and second jet zones at a certain level. to ;

[0142] After the heat pump dehumidification mechanism completes the adjustment, it feeds back the actual unit dehydration energy consumption after adjustment to the model prediction and control unit, so that the model prediction and control unit can correct the control quantity sequence in the next sampling period.

[0143] The category parameter set is updated by the parameter update module after each production batch. The production batch is a group of fruit and vegetable materials that enter the system continuously under the same material category, the same pretreatment method, and the same specifications. The parameter update module collects the previous batch outlet moisture, quick-freezing clumping feedback, rehydration taste feedback, and color difference feedback for the production batch. The previous batch outlet moisture is the amount of water attached to the material surface and the amount of free water between the tissues measured by sampling before the quick-freezing section enters. The quick-freezing clumping feedback is the mass ratio of clumping in a unit mass of material after quick-freezing. The rehydration taste feedback is the grade value calculated by hardness, brittleness, and tissue integrity after rehydration. The color difference feedback is the offset of the brightness component, red-green component, and yellow-blue component relative to the standard sample before and after quick-freezing.

[0144] The parameter update module associates the clumping feedback after quick-freezing with the category ice bridge threshold, the color difference feedback with the color protection warning threshold and the color protection threshold, the actual unit dehydration energy consumption with the energy consumption threshold, the previous batch outlet moisture with the surface water film threshold, the surface water film weight and the free water weight, and the rehydrated taste feedback as a correction item to limit excessive dehydration. It also uses an exponential smoothing method with a limit to update the category parameter set so that abnormal feedback in a single batch will not cause sudden changes in thresholds and weights.

[0145] The updated category parameter set includes category ice bridge threshold, color protection warning threshold, color protection threshold, energy consumption threshold, surface water film threshold, surface water film weight, and free water weight. The parameter update module writes the updated category parameter set into the category parameter library and provides it to the dual-domain moisture estimation unit, risk and energy consumption generation module, and model prediction control unit at the beginning of subsequent batches. This allows subsequent batches to continue to perform closed-loop correction based on the historical outlet moisture, quick-freezing clumping feedback, rehydrated taste feedback, and color difference feedback of the same material category.

[0146] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0147] This invention transforms the pre-treatment data of fruits and vegetables, including structure, moisture, airflow, energy consumption, dehydration amount, surface temperature, and color difference, into surface water film index, interstitial free water index, ice crystal bridging risk index, color damage risk, and unit dehydration energy consumption. This allows the pre-freezing dehydration process to be controlled by closed-loop control based on material and risk states, rather than fixed process parameters. This more effectively addresses issues such as uneven residual water, ice bridging, dull color, and high energy consumption.

[0148] This invention combines risk generation, energy consumption feedback, category parameter updates, and color protection micro-mist safety priorities, making the color protection treatment subject to ice bridge risk and surface water film threshold constraints. This allows heat pump dehumidification and model predictive control to be subsequently corrected based on actual unit dehydration energy consumption, thereby improving the stability and adaptability of the continuous quick-freezing pretreatment process for multiple categories of fruits and vegetables.

Claims

1. A method for quick-freezing fruits and vegetables to preserve color and save energy during dehydration, characterized in that, include: S1. Spread the washed, cut or blanched fruit and vegetable materials in a single layer, collect data on material structure, moisture, airflow, energy consumption, dehydration, surface temperature and color difference, and determine the spreading thickness, dwell time before freezing, dwell time after pretreatment and material contact probability to form raw material state data. S2. Input the raw material state data into the dual-domain moisture estimation unit, and generate the surface water film index and interstitial free water index according to the category parameter set; S3. Generate an ice crystal bridging risk index based on surface water film index, interstitial free water index, material contact probability, spreading thickness, residence time before freezing, and airflow data. Generate color damage risk based on color difference, surface temperature, and residence time after pretreatment. Generate unit dehydration energy consumption based on energy consumption and dehydration data. S4. The model prediction and control unit aims to ensure that the ice crystal bridging risk index is within the category ice bridging threshold, and is constrained to ensure that the color damage risk is within the color protection threshold and the unit dehydration energy consumption does not exceed the energy consumption threshold. It determines the jet parameters, air inlet dew point, vacuum parameters, vibration frequency, conveying speed and color protection micro-mist start and stop signals. S5. Drives dew point gradient jet, vibration loosening, pulse vacuum, heat pump dehumidification, conveying drive and color protection micro-mist mechanism according to control amount to remove surface droplets, attached water film and free water between tissues; S6. When the ice crystal bridging risk index updated after processing by S5 is within the category ice bridging threshold and the color damage risk is within the color protection threshold, the fruit and vegetable materials are sent to the quick-freezing section.

2. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 1, characterized in that, S1 includes: According to the material category, the particle size sampling rules, flower ball gap sampling rules, cross-section recognition area, near-infrared band group, and dielectric sampling frequency are read from the category parameter library; the material structure data includes material category, particle size or flower ball gap, and cross-section area; the moisture detection data includes surface image bright spot area, near-infrared moisture response, dielectric response, and online weight change; the airflow status data includes inlet dew point, inlet humidity, and outlet humidity; the energy consumption data includes the electrical energy metering values ​​of the heat pump dehumidification mechanism, pulse vacuum mechanism, vibration loosening mechanism, and conveying drive mechanism; the dehydration amount data is determined by the online weight change at adjacent sampling times; the pre-freezing residence time is the residence time from the completion of single-layer spreading to the expected permissible output time for freezing, and is updated according to the conveying speed and processing section length; the post-pre-treatment residence time is the residence time from the completion of the last washing, cutting, or blanching to the current sampling time. The ratio of the area of ​​the bright spot region in the conveyor belt image to the visible area of ​​the material is determined as the bright spot area of ​​the surface image, and the material contact probability is determined based on the spreading thickness, the particle size or the gap between the flower balls, and the distance between the boundaries of adjacent materials in the conveyor belt image.

3. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 2, characterized in that, S2 includes: After normalizing the bright spot area of ​​the surface image, the near-infrared moisture response, the increment of the outlet humidity relative to the inlet humidity, and the online weight change, the surface water film index is generated according to the surface water film weight in the category parameter set; after normalizing the dielectric response, the online weight change, the increment of the outlet humidity relative to the inlet humidity, and the material surface temperature, the interstitial free water index is generated according to the free water weight in the category parameter set; the surface water film index is used to constrain the jet velocity and jet duty cycle, and the interstitial free water index is used to constrain the vacuum amplitude and vacuum holding time.

4. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 3, characterized in that, S3 includes: inputting the surface water film index, the interstitial free water index, the material contact probability, the spreading thickness, the dwell time before freezing, and the outlet air humidity into an ice bridge risk mapping table to obtain the ice crystal bridging risk index; inputting the brightness component, red-green component, and yellow-blue component in the color difference data, the material surface temperature, and the dwell time after pretreatment into a color constraint mapping table to obtain the color damage risk; and determining the ratio of the electrical energy metering value to the dehydration amount data as the unit dehydration energy consumption.

5. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 4, characterized in that, S4 includes: The model prediction control unit uses the ice crystal bridging risk index, color damage risk, actual unit dehydration energy consumption from the previous sampling period, surface water film index, and interstitial free water index as state variables, and the category ice bridge threshold, color protection threshold, and energy consumption threshold as constraint boundaries to calculate a control quantity sequence within a preset prediction time domain. The first control quantity selected from the control quantity sequence is used as the jet velocity, jet duty cycle, inlet dew point, vacuum amplitude, vacuum holding time, vibration frequency, conveying speed, and color protection micro-mist start / stop signal for the next sampling period. The actual unit dehydration energy consumption after execution is fed back to the model prediction control unit to correct the control quantity sequence for the next sampling period.

6. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 5, characterized in that, The dew point gradient pulse jet mechanism in step S5 includes a first jet zone and a second jet zone arranged sequentially along the conveying direction; the first jet zone outputs a dehumidifying airflow with a temperature between 0 and 10 degrees Celsius and a dew point between -25 degrees Celsius and 0 degrees Celsius, and intermittently washes the surface droplets and the attached water film according to the jet duty cycle; the second jet zone receives the fruit and vegetable materials processed by the pulse vacuum mechanism, resamples the processed fruit and vegetable materials, and obtains an updated value of the surface water film index through the dual-domain moisture estimation unit, and corrects the jet wind speed according to the updated value within the control quantity sequence determined by the model prediction control unit.

7. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 6, characterized in that, The phase of the pulse vacuum mechanism and the vibration loosening mechanism is determined by the vacuum holding time, the vacuum amplitude, and the vibration frequency. During the interval after the dew point gradient pulse jet mechanism stops airflow, the pulse vacuum mechanism adjusts the processing chamber pressure to a preset vacuum pressure range and maintains the vacuum holding time. During the pressure recovery phase of the processing chamber, the vibration loosening mechanism turns the material according to the vibration frequency, so that the moisture that has migrated to the surface of the material enters the removal path of the next jet cycle.

8. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 5, characterized in that, The color protection micro-mist start / stop signal is determined based on the color damage risk, the material surface temperature, and the residence time after pretreatment; the color protection early warning threshold comes from the category parameter set and is less than the color protection threshold. When the risk of color damage reaches the color protection warning threshold but does not exceed the color protection threshold, the ice crystal bridging risk index is within the category ice bridging threshold, and the surface water film index is within the surface water film threshold of the category parameter set, the color protection micro-mist mechanism is activated to apply color protection liquid droplets to the material surface; when the surface water film index exceeds the surface water film threshold or the ice crystal bridging risk index is not within the category ice bridging threshold, the color protection micro-mist mechanism is suppressed and low dew point jet film removal control is prioritized; when the risk of color damage exceeds the color protection threshold, the color protection micro-mist mechanism is suppressed, the priority of low dew point dehumidification and cooling control is increased, and the output of freezing permit signal is prohibited until the latest sampled risk of color damage is again within the color protection threshold.

9. The quick-frozen fruit and vegetable color-protecting, energy-saving, and dehydration method according to claim 5, characterized in that, S6 include: When the ice crystal bridging risk index and the color damage risk obtained from the latest sampling before the freezing permit is issued are both within the corresponding thresholds for two consecutive sampling periods, the freezing permit signal is output and the fruit and vegetable materials are sent into the quick-freezing section. The heat pump dehumidification mechanism adjusts the compressor frequency, evaporator bypass ratio, and reheat capacity based on the inlet dew point, the outlet humidity, the outlet temperature collected by the outlet temperature sensor, and the actual unit dehydration energy consumption generated by the energy consumption data and the dehydration amount data. The adjusted actual unit dehydration energy consumption is then fed back to the model prediction control unit. The category parameter set includes a color protection warning threshold and a surface water film threshold. The category parameter set is updated based on the previous batch's outlet moisture, quick-freezing clumping feedback, rehydrated taste feedback, and color difference feedback. The updated category ice bridge threshold, color protection warning threshold, color protection threshold, energy consumption threshold, surface water film threshold, surface water film weight, and free water weight are provided to the dual-domain moisture estimation unit, risk and energy consumption generation module, and model prediction control unit for subsequent batches.

10. A quick-freezing fruit and vegetable color-protecting and energy-saving dehydration system, used to perform the quick-freezing fruit and vegetable color-protecting and energy-saving dehydration method according to any one of claims 1 to 9, characterized in that, include: The spreading and acquisition module is used to spread the washed, cut or blanched fruit and vegetable materials in a single layer and form the raw material state data. A dual-domain moisture estimation module is used to generate the surface water film index and the interstitial free water index; a risk and energy consumption generation module is used to generate the ice crystal bridging risk index, the color damage risk and the unit dehydration energy consumption, and feeds back the actual unit dehydration energy consumption to the predictive control module and the heat pump dehumidification mechanism. The predictive control module is used to determine the jet wind speed, the jet duty cycle, the inlet dew point, the vacuum amplitude, the vacuum holding time, the vibration frequency, the conveying speed, and the color protection micro-mist start / stop signal; the segmented execution module includes a dew point gradient pulse jet mechanism, a vibration loosening mechanism, a pulse vacuum mechanism, a heat pump dehumidification mechanism, a conveying drive mechanism that receives the conveying speed, and a color protection micro-mist mechanism that receives the color protection micro-mist start / stop signal; The quick-freezing connection module is used to send the fruit and vegetable materials into the quick-freezing section when the ice crystal bridging risk index and the color damage risk, which are both resampled and updated after being processed by the segmented execution module, are within the corresponding thresholds.

Images