Microcrystal resonance refrigeration system

Abstract

The invention relates to the technical field of freezing, in particular to a microcrystal resonance freezing system which is used for solving the problems that in seafood freezing, a phase change window is difficult to judge, freezing is slow, energy consumption is high, ice crystals are too large and water damage is caused due to the fact that ultrasonic input and refrigerating capacity are not synchronous, and the microcrystal resonance freezing system comprises a freezing cavity, a refrigerating unit, a tray, an ultrasonic vibrator, a temperature detection unit and a main control unit. The refrigerating unit is in heat exchange connection with the freezing cavity, the ultrasonic vibrator comprises a plurality of ultrasonic transducers, is arranged below the tray and is coupled to seafood through the sound conducting layer, and the temperature detection unit comprises a cavity temperature sensor and a seafood center temperature probe; the ultrasonic intermittent rhythm is linked with the output of the refrigerating unit in the seafood phase change window, and the ultrasonic starting and stopping duration and power are adaptively optimized on line, so that the phase change platform and the total freezing duration are shortened, the ineffective energy consumption is reduced, the central temperature fluctuation and local supercooling are inhibited, and fine and uniform ice crystals are formed; the dripping loss is reduced; the taste of the frozen seafood is improved.

Description

Technical Field

[0001] This invention relates to the field of cryogenics, and more particularly to a microcrystalline resonance cryogenics system. Background Technology

[0002] Chinese invention patent CN112503815A discloses an ice maker and its control method for ultrasonic-assisted freezing. This ice maker incorporates ultrasonic waves during the water freezing process, utilizing the mechanical and cavitation effects of ultrasound to accelerate ice nucleus formation and growth, thereby shortening freezing time and improving ice-making efficiency. By real-time monitoring of the water temperature in the ice trays during freezing, the ultrasonic generator is activated when the temperature reaches the phase transition temperature required for nucleation. Throughout the phase transition period, the ultrasonic generator is controlled to operate at a specific frequency. Precise control of the ultrasonic wave duration not only shortens the ice-making time but also effectively reduces energy consumption. However, the ice-making object described in this invention is a single-phase water system placed in a regular ice tray, while seafood, when frozen, is a heterogeneous solid food with discrete shape and size, containing salt solutes, free water, and bound water. The core cooling and phase transition periods of seafood of different sizes within the same freezing chamber vary greatly. Frequent opening of doors or drawers to remove frozen products introduces heat and humidity loads, exacerbates frost formation, prolongs the freezing time of uncold products, and increases energy consumption. Seafood is not a liquid filling the ice tray; contact surfaces, air gaps, packaging films, and frost layers all weaken the ultrasonic energy input. A single bottom transducer cannot cover multiple trays and layers of stacked seafood. At the same time, freezing seafood requires controlling the size and distribution of ice crystals to reduce water loss and tissue damage. If the ultrasonic parameters and the action window are not matched, it can lead to local cavitation of seafood products and cell membranes being punctured by excessively large ice crystals, thus affecting the taste and juice yield of the seafood. In addition, the phase transition platform of seafood is not synchronized with the core temperature, surface temperature, and cavity temperature. It is difficult to stably determine whether seafood enters or exits the phase transition window at a single temperature point, which can easily cause the ultrasonic action time to shift, resulting in ineffective energy consumption and affecting the quality of the seafood. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a microcrystalline resonance freezing system, which links the intermittent ultrasonic beat with the output of the refrigeration unit within the phase change window of seafood, and adaptively optimizes the ultrasonic start-stop time and power online, thereby shortening the phase change platform and total freezing time, reducing ineffective energy consumption, suppressing central temperature fluctuations and local supercooling, forming fine and uniform ice crystals, reducing dripping water loss and improving the taste of frozen seafood.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A microcrystalline resonant freezing system includes a freezing chamber, a refrigeration unit, a tray, an ultrasonic transducer, a temperature detection unit, and a main control unit. The refrigeration unit is heat-exchange connected to the freezing chamber. The ultrasonic transducer includes multiple ultrasonic transducers, which are arranged under the tray and coupled to the seafood through a sound guiding layer. The temperature detection unit includes a cavity temperature sensor and a seafood center temperature probe. When the center temperature enters a preset phase change temperature zone and the temperature change rate meets a preset threshold, the main control unit controls the ultrasonic transducer to work intermittently for a preset start-up time and a preset stop time, and simultaneously modulates the output of the refrigeration unit.

[0006] It should be noted that during the seafood freezing process, the key to quality lies in the number of ice crystal nucleation points and the crystal growth rate during the phase transition stage. If there are few nucleation points and a long phase transition plateau, water will preferentially grow into large ice crystals outside the cells, compressing the cell structure, leading to increased drip loss during thawing and a deterioration in the seafood's texture. The system proposed in this invention identifies the phase transition temperature zone using a cavity temperature sensor and a seafood center temperature probe. This allows the main control unit to trigger intermittent ultrasonic action only when the center temperature enters the phase transition plateau and the temperature change rate meets the threshold. The ultrasonic transducer then passes through a tray to mechanically... Vibration and microscale disturbances introduced into the free water region of seafood tissue promote the formation of more microbubble nuclei and micro-disturbance nucleation points, increasing the instantaneous nucleation density and causing ice crystals to disperse and start on a large number of nuclei, making it difficult for them to grow. On the other hand, they weaken the local concentration gradient boundary layer and thermal boundary layer, improve the transfer and discharge efficiency of latent heat of phase change, thereby shortening the duration of the plateau and suppressing the tendency of coarse crystal formation with few nuclei and slow growth. This results in faster freezing and a finer and more uniform ice crystal structure, reducing damage to seafood tissue and drip loss, improving the quality of seafood after freezing, and also reducing energy consumption.

[0007] As a further aspect of the present invention, the seafood center temperature probe is positioned at the center of a representative heat load. The representative heat load is a food-grade simulated load block of the same batch and specifications as the seafood to be frozen. The center hole is a pre-drilled hole. The seafood center temperature probe is surrounded by a food-grade sealing sleeve and sealed with food-grade sealing material. By using a food-grade simulated load block of the same batch and specifications as a reference heat load and employing a pre-drilled hole and sealing sleeve to position the center temperature probe, real and representative center phase change temperature data can be stably obtained without puncturing or contaminating the seafood to be sold, thereby achieving precise control of the timing of ultrasonic and cooling triggering.

[0008] As a further aspect of the present invention, the cavity temperature sensor is installed at the return air inlet of the freezing cavity, which can reflect the overall heat load in the freezing cavity and the temperature of the mixed air after the cooled object is reheated, thereby providing more stable control feedback for the output adjustment of the refrigeration unit and resisting local direct airflow interference.

[0009] As a further aspect of this invention, the refrigeration control unit is connected to a refrigeration process knowledge graph library. This library is a time-series causal knowledge graph for refrigeration control, constructed as stage slices according to the pre-cooling stage, sub-cooling stage, nucleation stage, phase transition stage, and cryogenic stage. This limits the control objectives and key constraints of the same batch of refrigeration processes to the corresponding stages, avoiding cross-stage parameter misuse and improving the timing accuracy of phase transition window triggering and parameter tuning. The refrigeration process knowledge graph library uses seafood category, specifications, initial core temperature, packaging method, loading layer, and the coupling state of the tray and packaging as parameters. Using nodes such as the number of parameters, target center temperature curve, phase change temperature zone parameters, temperature change rate threshold, ultrasonic start-up time, ultrasonic shutdown time, ultrasonic power, target output of the refrigeration unit, and quality quantification indicators, a unified structured expression of load attributes, coupling conditions, control parameters, and quality results is achieved. This allows for multi-objective optimization of quality indicators such as freezing speed, energy consumption, and drip loss. Each edge includes causal weights, confidence levels, and applicable domain parameters to quantify the influence of different attribute factors on phase change platform duration, temperature fluctuations, and quality loss, and to limit the applicable range of parameters, thereby improving the recommended parameters. The data transferability is improved, and the failure risk under mismatched operating conditions is reduced. Causal weights and confidence levels are incrementally updated based on the deviation of the center temperature curve of historical batches, the duration of the phase change platform, and energy consumption and quality quantification indicators. This allows the graph to adaptively calibrate the causal strength and confidence level as batches accumulate, continuously correcting parameter recommendation deviations and improving long-term stability and consistency. After receiving cavity temperature and center temperature data, the main control unit maps the attribute characteristics of the current batch of seafood to graph query conditions. It first filters the candidate parameter set based on the applicable domain and hard constraint edges, and then slices the subgraph at each stage based on graph embedding similarity. The system performs recall and sorting to eliminate parameter combinations that violate constraints such as temperature fluctuation limits, duty cycle limits, energy input limits, and quality thresholds before output, ensuring the safety and feasibility of the control process. It outputs the phase change temperature zone, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-up time, ultrasonic power, and target output parameters of the refrigeration unit for the current batch of seafood, along with their corresponding confidence levels. This provides an interpretable and traceable source of parameters for ultrasonic-refrigeration coordinated control and supports conservative backoff or online adaptive correction based on confidence levels, thereby improving the quality consistency and energy consumption controllability of batch seafood freezing.

[0010] As a further aspect of this invention, the main control unit generates a parameter set based on a collaborative recommendation mechanism. The collaborative object is the historical frozen batches. The main control unit encodes the category, specifications, packaging method, loading layer, initial center temperature, and target endpoint temperature of the current batch of seafood into user-side feature vectors. It encodes the phase change temperature zone parameters, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-up time, ultrasonic power, and target output of the refrigeration unit of each historical frozen batch into item-side feature vectors. Based on the comprehensive freezing effect score composed of the similarity of the center temperature curve and freezing time, energy consumption, and drip loss, a user-item interaction matrix is ​​constructed. The main control unit performs neighborhood collaborative filtering on the interaction matrix to obtain the predicted score of the candidate parameter set, and selects the candidate parameter set with the highest predicted score as the phase change temperature zone, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-up time, ultrasonic power, and target output of the refrigeration unit for the current batch of seafood.

[0011] As a further aspect of the present invention, the main control unit performs online adaptive adjustment of the ultrasonic start-up duration: within the phase change temperature zone, the center temperature sequence within the sliding time window is taken, and after exponential weighted filtering, the actual temperature change rate is obtained by least squares fitting. The duration of the phase change platform and the center temperature fluctuation amplitude are statistically obtained to form a first actual feature vector. The first target feature vector is obtained from the target center temperature curve corresponding to the current freezing batch, and the first feature covariance matrix is ​​obtained from historical batches similar to the current batch. The first Mahalanobis deviation is calculated. When the first Mahalanobis deviation exceeds the first threshold and the absolute value of the actual temperature change is less than the absolute value of the target temperature change, and the duration of the phase change platform is greater than the duration of the target platform, the ultrasonic start-up duration is increased by a preset step size. When the first Mahalanobis deviation exceeds the first threshold and the center temperature fluctuation amplitude reaches or exceeds the preset fluctuation upper limit, the ultrasonic start-up duration is decreased by a preset step size, and the updated ultrasonic start-up duration is used for the next control cycle.

[0012] It should be noted that the reason for using the above-mentioned online adaptive adjustment based on a sliding time window for the start-up time of the ultrasonic transducer is that the ultrasonic coupling efficiency in the batch freezing of seafood dynamically changes with the loading layer, packaging air gap, and the contact state between the frost layer and the tray. Fixing the start-up time easily leads to two types of mismatches: one is that insufficient ultrasonic action leads to a prolonged phase change plateau period and low latent heat extraction efficiency; the other is that ultrasonic superimposed cooling causes a short-term increase in the center temperature fluctuation or even a warming oscillation, thus affecting quality consistency and energy consumption. The temperature change rate, plateau duration, and fluctuation amplitude are stably extracted through exponential weighted filtering and least squares fitting, and the first Mahalanobis bias is used to separate the differences between multiple indicators. The system uses a unified measurement under covariance constraints to avoid misjudgment by a single indicator and adaptively normalizes the correlation of indicators under different operating conditions. When insufficient propulsion is detected (low temperature change rate and long phase change platform), the start-up time is increased to improve the effective nucleation and interface heat transfer enhancement time, accelerating the release of latent heat of phase change. When excessive disturbance is detected (fluctuation close to the upper limit), the start-up time is reduced to suppress temperature oscillations and local quality risks caused by ultrasound. This achieves closed-loop matching between ultrasonic energy input and actual freezing dynamics, shortens the phase change platform, reduces ineffective energy consumption, and maintains the stability of the central temperature curve and the consistency of seafood freezing quality under batch differences and coupling state drift.

[0013] As a further aspect of the present invention, the main control unit performs online adaptive adjustment of the ultrasonic downtime: within a preset phase change temperature zone, the core temperature of the seafood is sampled by a sliding time window, the effective proportion of ultrasound within the time window is calculated, and the core temperature fluctuation amplitude and the duration of the phase change platform are obtained to form a second actual feature vector. The second target feature vector is obtained from the target core temperature curve corresponding to the current frozen batch, and the second feature covariance matrix is ​​obtained from similar historical batches. The second Mahalanobis deviation is calculated. When the second Mahalanobis deviation exceeds the second threshold and the effective duty cycle of ultrasound exceeds the preset duty cycle upper limit or the core temperature fluctuation threshold reaches or exceeds the preset fluctuation upper limit, the ultrasonic downtime is increased by a preset step size. When the second Mahalanobis deviation exceeds the second threshold and the duration of the phase change platform is greater than the target platform duration and the core temperature fluctuation amplitude is less than the target fluctuation amplitude, the ultrasonic downtime is decreased by a preset step size, and the updated ultrasonic downtime is used for the next control cycle.

[0014] It should be noted that the reason for adopting the above-mentioned online adaptive adjustment based on a sliding time window for the ultrasonic downtime is that the nucleation and latent heat release processes of seafood tissue within the phase transition temperature zone are highly nonlinear. The duty cycle of the ultrasonic pulse directly determines the energy input and disturbance intensity per unit time. A fixed downtime is prone to mismatch under different loading densities, coupling efficiencies, and cooling intensities: too short a downtime will result in an excessively high effective duty cycle, which, after superimposed with the cooling output, will cause increased core temperature fluctuations, increased rebound risk, and increased ineffective energy consumption; too long a downtime may result in an excessively large ultrasonic action interval, making it difficult to sustain nucleation and boundary layer disturbances, leading to a prolonged phase transition plateau and slower freezing progress. By simultaneously calculating the effective ultrasonic duty cycle, core temperature fluctuation amplitude, and phase transition plateau duration within the sliding time window, and utilizing... Using the second Mahalanobis bias under covariance constraints to uniformly discriminate deviations of multiple indicators can avoid misadjustment caused by a single threshold in different batches and under different coupling states. When excessive disturbance or excessively high duty cycle is detected (duty cycle exceeding the upper limit or fluctuation approaching the upper limit), the downtime is increased to reduce the duty cycle and allow convergence time for the temperature field and tissue phase transition process, thereby suppressing temperature oscillation and quality risks. When insufficient disturbance and slow progress are detected (platform too long and fluctuation below the target), the downtime is reduced to increase the frequency of ultrasonic action, making nucleation and heat transfer enhancement more continuous, thereby shortening the phase transition plateau and improving freezing efficiency. This achieves closed-loop matching between ultrasonic pulse beat and phase transition dynamics, reducing ineffective energy consumption and improving the beat stability of batch freezing while ensuring temperature stability and quality consistency.

[0015] As a further aspect of this invention, the main control unit performs online adaptive adjustment of the ultrasonic power: within a preset phase change temperature zone, the actual temperature change rate, the duration of the phase change platform, and the amplitude of the center temperature fluctuation are obtained based on a sliding time window to form a third target feature vector. The target center temperature curve corresponding to the current freezing batch is used to obtain the third target feature vector, and the third feature covariance matrix is ​​obtained from similar historical batches. The third Mahalanobis deviation is calculated, and the power sensitivity is obtained based on the changes in ultrasonic power and temperature change rate over adjacent time periods. When the third Mahalanobis deviation exceeds the third threshold, the absolute value of the actual temperature change rate is less than the absolute value of the target temperature change rate, the duration of the phase change platform is greater than the duration of the target platform, and the cumulative ultrasonic energy has not reached the preset energy upper limit, the ultrasonic power is increased by a preset power increment determined according to the power sensitivity. When the amplitude of the low fluctuation of the center temperature is greater than or equal to the preset fluctuation upper limit or the cumulative ultrasonic energy reaches the preset upper limit, the ultrasonic power is decreased by a preset power step size, and the updated ultrasonic power is used for the next control cycle.

[0016] It should be noted that the reason for adopting the above-mentioned online adaptive adjustment of ultrasonic power based on a sliding time window is that the nucleation efficiency and the degree of interfacial heat transfer enhancement of seafood in the phase transition temperature range exhibit a significant condition-dependent effect on ultrasonic energy input: affected by packaging materials, tray coupling, frost layer, and stacking height, the same power may correspond to completely different effective sound intensities in different batches. Fixed power is prone to the mismatch between insufficient power leading to slow phase transition or excessive power causing amplified temperature fluctuations and quality risks. By robustly extracting the temperature change rate, plateau duration, and fluctuation amplitude within the sliding time window, and using the third Mahalanobis bias to comprehensively judge the degree of deviation of multiple indicators under covariance constraints, more reliable deviation detection is achieved under the differences between different batches and changes in the correlation of indicators. At the same time, power sensitivity (caused by power changes) is introduced. The change rate improvement is used to determine the power increment, shifting power adjustment from blind addition and subtraction to effective regulation, avoiding excessive power increase due to low coupling efficiency and resulting in ineffective energy consumption. Specifically, when it is determined that the propulsion is insufficient and the energy efficiency can still be improved (the change rate is small, the plateau is long and the energy limit has not been reached), the ultrasonic power is increased by the increment calculated according to the power sensitivity to enhance the nucleation density and boundary layer disturbance and accelerate the extraction of latent heat, thereby shortening the phase transition plateau. When excessive disturbance or excessive energy input is detected (fluctuation reaches the upper limit or accumulated energy reaches the upper limit), the ultrasonic power is reduced to suppress temperature oscillation, avoid tissue structure disturbance and control energy consumption. This achieves closed-loop matching between ultrasonic energy input and freezing dynamics and quality constraints, improving phase transition propulsion efficiency and reducing ineffective energy consumption while ensuring temperature stability and quality consistency.

[0017] As a further aspect of the present invention, when the center temperature enters the preset phase change temperature zone and the temperature change rate meets the preset threshold, the main control unit will link and modulate the output of the refrigeration unit with the intermittent beat of the ultrasonic transducer.

[0018] As a further aspect of the present invention, in the main control unit, the intermittent beat-linked modulation method of the refrigeration unit output and the ultrasonic transducer is specifically as follows: when the core temperature of the seafood enters the preset phase change temperature zone and the temperature change rate meets the preset threshold, the ultrasonic transducer is placed in a periodic intermittent working mode with a start-up duration and a stop duration, and the refrigeration unit output is segmented into start-up output and stop-up output according to the intermittent beat; within the start-up segment of each cycle, the refrigeration unit output is set to a high output level, which is achieved by increasing the duty cycle of the compressor inverter speed of the refrigeration unit, and simultaneously executing at least one of the following: lowering the evaporation temperature setting or increasing the throttle valve opening. Set the circulating fan speed to the high airflow setting; during the shutdown phase of each cycle, set the chiller unit output to the reference output setting or the slow-release output setting. The reference output setting or the slow-release output setting is achieved by reducing the duty cycle of the chiller unit's compressor inverter speed drive, and simultaneously execute at least one of the following: evaporator temperature setpoint recovery or throttle valve opening reduction. At the same time, set the circulating fan speed to the medium airflow setting; between the start-up and shutdown phases, apply upper limit constraints to the compressor speed change rate, evaporator temperature setpoint change rate, and circulating fan speed change rate, and adaptively correct the gear difference between the high output setting and the reference output setting based on the center temperature fluctuation amplitude and the phase change platform duration.

[0019] It should be noted that during the ultrasonic start-up phase, the main control unit upgrades the refrigeration unit from the reference output to the high output: specifically, it gradually increases the target variable frequency speed or drive duty cycle of the compressor from the reference value by a preset step size, and each increase is constrained by the upper limit of the speed change rate to linearly climb to the high output level; at the same time, it gradually decreases the target evaporation temperature setting from the reference value by a preset temperature step size, and the downward slope is limited by the upper limit of the evaporation temperature setting change rate, or it gradually increases the throttle valve opening from the reference value by a preset opening step size, and the increase slope is limited by the upper limit of the valve position change rate, and the upper and lower limits of superheat are used as safety constraints to avoid liquid slugging or return gas overheating; at the same time, it increases the circulating fan speed from the reference value by a preset step size, and the upward slope is limited by the upper limit of the fan speed change rate to the high air volume level.

[0020] It should be noted that after entering the ultrasonic shutdown phase, the main control unit will reduce the output of the refrigeration unit from high to the reference output or gradually decrease the output: specifically, the compressor target speed or drive duty cycle is gradually reduced in preset steps and smoothly reduced under the constraint of the upper limit of the speed change rate; the evaporation temperature target setting is gradually increased in preset temperature steps and constrained by the upper limit of the change rate, or the throttle valve opening is gradually decreased in preset opening steps and constrained by the upper limit of the valve position change rate, while the circulating fan speed is gradually reduced to the medium air volume setting in preset steps; when switching between the start-up and shutdown phases, the main control unit uniformly limits the slope of the changes of the above-mentioned settings, and adaptively corrects the difference between the high output setting and the reference output setting based on the amplitude of the center temperature fluctuation and the duration of the phase change platform: when the duration of the phase change platform is too long and the temperature fluctuation is too small, the increase of the high output setting relative to the reference output setting is increased; when the temperature fluctuation is close to the upper limit, the increase is reduced and the reduction is accelerated, so that the refrigeration output and the ultrasonic beat promote the phase change in coordination without causing temperature oscillation.

[0021] The technical advantages of the microcrystalline resonance freezing system of the present invention are as follows:

[0022] This invention improves acoustic coverage and coupling efficiency by deploying multiple transducers under the tray and using a sound guiding layer. Combined with a center temperature probe and return air cavity temperature sensor positioned at the center of the food-grade simulated heat load, it reliably identifies the core phase change window of seafood and reduces door opening intervention. Simultaneously, the main control unit uses a time-series causal process knowledge graph to recommend parameters and output confidence levels. It also implements linked modulation and online adaptive correction of ultrasonic start / stop duration and power, as well as the output of the refrigeration unit. This shortens the phase change plateau period, increases freezing speed, controls temperature fluctuations, and reduces energy consumption. Furthermore, by increasing nucleation density and uniformly extracting latent heat to form fine, uniform ice crystals, it reduces dripping water loss and tissue damage. Under complex loading and packaging conditions, it achieves consistent frozen seafood quality and improved batch stability. It also suppresses excessive local cavitation and the growth of large ice crystals, reduces frost formation and the superposition of heat and moisture loads, improves the adaptability of continuous operation and automated handling, and enhances the taste of frozen seafood. Attached Figure Description

[0023] Figure 1 This is a system diagram of the present invention;

[0024] Figure 2 This is a schematic diagram of the coupling between the cooling output and the ultrasonic start / stop of the present invention. Detailed Implementation

[0025] The technical solutions of 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.

[0026] Example

[0027] This embodiment provides the application process of the microcrystalline resonance freezing chamber machine proposed in this invention for the batch freezing of Litopenaeus vannamei, such as... Figure 1 As shown, the system includes a freezing chamber, a refrigeration unit, a tray, an ultrasonic transducer, a temperature detection unit, and a main control unit. The refrigeration unit is connected to the freezing chamber for heat exchange. The ultrasonic transducer includes multiple ultrasonic transducers, which are arranged under the tray and coupled to the seafood through a sound guiding layer. The temperature detection unit includes a cavity temperature sensor arranged at the return air vent of the freezing chamber and a seafood center temperature probe. When the center temperature enters the preset phase change temperature zone and the temperature change rate meets the preset threshold, the main control unit controls the ultrasonic transducer to work intermittently for a preset start-up time and a preset shutdown time, and simultaneously controls the output of the refrigeration unit to increase the nucleation density, shorten the phase change plateau, and suppress center temperature fluctuations within the phase change window, thereby forming fine and uniform ice crystals, reducing dripping water loss, and ensuring controllable energy consumption.

[0028] Table 1 Hardware and Control Parameters

[0029]

[0030] As shown in Table 1, the effective volume of the freezing chamber in this embodiment is 0.65 m³. 3 The refrigeration unit includes a variable frequency compressor (30-90Hz), an electronic expansion valve (0-100% opening), and a circulating fan (800-1600rpm). The tray is made of aluminum alloy and carries seafood packaging bags on top. A food-grade silicone sound-guiding layer with a thickness of 2.0mm is installed below the tray. An array of multiple ultrasonic transducers is arranged below the sound-guiding layer, with 4 ultrasonic transducers per tray, arranged in a 2×2 partition. Each transducer has a rated power of 20W and a frequency of 40Hz. The power is adjustable and is used to couple ultrasonic mechanical vibrations to the seafood load through the sound-guiding layer. A cavity temperature sensor is installed at the return air vent of the freezing cavity to collect the mixed return air temperature after the seafood load has been reheated, so as to obtain a stable cavity temperature feedback representing the entire heat load. The seafood core temperature probe is placed at the center of a representative heat load, which is a food-grade simulated load block of the same batch and specifications as the seafood to be frozen. The center hole is a pre-made hole. The probe is equipped with a food-grade sealing sleeve and a food-grade sealing material vent, so that the representative core temperature can be stably obtained without puncturing or contaminating the seafood to be sold.

[0031] In this embodiment, the frozen object is Litopenaeus vannamei, with a length of 40 to 50 mm, each bag weighing approximately 500 g, packaged in PE composite film, and loaded in a pallet of 20 bags, stacked in two layers with staggered seams, for a total of 2 pallets. The initial center temperature was measured to be 8°C based on the reference heat load center temperature, and the target endpoint center temperature was -18°C and maintained for 20 minutes. The coupling state parameters were recorded by batch as follows: air gap level 1, frost layer level 0, and good sound guide layer integrity.

[0032] After loading is completed, the main control unit reads the seafood category, specifications, packaging method, loading layer, initial center temperature, target endpoint temperature, and coupling state parameters of this batch of seafood, maps them to knowledge graph query conditions, first filters the candidate parameter set based on the applicable domain and hard constraint edges in the phase transition stage slice subgraph, then recalls and sorts them in the stage slice subgraph based on graph embedding similarity, and integrates the collaborative recommendation mechanism (historical batches are collaborative objects, and the comprehensive freezing effect score is the interaction matrix score) to output the initial control parameter set and confidence of the current batch.

[0033] Table 2. Recommended output parameter set for knowledge graph (confidence level 0.82)

[0034]

[0035] As shown in Table 2, the confidence level of the recommended results for this batch is 0.82. The preset phase change temperature range is [-1.4℃, -0.4℃]. The phase change platform is determined by the center temperature change rate being less than or equal to 0.03℃ / min. The ultrasonic intermittent beat is 20s for start-up, 40s for stop, and 60s for cycle. The ultrasonic power is 80W. In the refrigeration unit's baseline setting, the operating frequency is 60Hz, the evaporation temperature is -30℃, the electronic expansion valve opening is 55%, and the fan speed is 1200rpm. In the high output setting during start-up, the operating frequency is 78Hz, the evaporation temperature is -34℃, the electronic expansion valve opening is 65%, and the fan speed is 1500rpm. In the stop buffer setting, the operating frequency is 58Hz, the evaporation temperature is -29℃, the electronic expansion valve opening is 50%, and the fan speed is 1100rpm. The slope upper limit constraint (operating frequency change rate less than or equal to 1.5Hz / s) is applied to the compressor speed change rate, evaporation temperature set change rate, electronic expansion valve position change rate, and fan speed change rate. At the same time, the upper limit of short-term fluctuation amplitude of center temperature is set to 0.15℃ and the upper limit of cumulative ultrasonic energy during phase transition is set to 0.09kWh.

[0036] In this embodiment, the loading completion time is t=0min (10:00 AM). Values ​​earlier than this time are recorded as negative values, and values ​​later than this time are recorded as positive values. The staged process includes cavity precooling, loading recovery, sensible heat cooling, supercooling or nucleation preparation, phase change linkage control, cryogenic propulsion and endpoint maintenance stages. The actual data and set values ​​of each stage are shown in Table 3.

[0037] (1) Cavity pre-cooling stage (t=-15min to t=0min)

[0038] Before loading, the main control unit controls the refrigeration unit to stabilize the cavity return air temperature to -28℃. The refrigeration settings are: compressor operating frequency of 75Hz, evaporation temperature of -34℃, electronic expansion valve opening of 62%, fan speed of 1500rpm, and ultrasonic transducer off, in order to establish cold capacity reserve.

[0039] (2) Loading recovery phase (t=0min to t=10min)

[0040] After loading at t=0 min, the cavity temperature instantly rose to -15.2℃, and the core temperature of the seafood was 8.0℃. The main control unit switched the refrigeration unit to high-output recovery mode to quickly bring the cavity temperature back up. At t=15 min, the cavity temperature dropped to 23.8℃, and the core temperature dropped to 6.1℃. The absolute value of the temperature change rate was 0.38℃ / min. At t=10 min, the cavity temperature was approximately -26.5℃, and the core temperature was approximately 4.0℃. The system then switched to the baseline setting to enter stable sensible heat cooling mode.

[0041] (3) Sensible heat cooling stage (t=10min to t=30min)

[0042] At t=20min, the cavity temperature is approximately -27.0℃, the center temperature is approximately 1.8℃, and the absolute temperature change rate is 0.11℃ / min. At t=30min, the cavity temperature is approximately -27.4℃, the center temperature is approximately 0.3℃, and the absolute temperature change rate is approximately 0.06℃ / min. The main control unit begins to perform exponential weighted filtering on the center temperature sequence and uses least squares fitting to calculate the actual temperature change rate, in preparation for phase transition window determination.

[0043] (4) Supercooling or nucleation preparation stage (t=30min to t=42min)

[0044] At t=38min, the cavity temperature is approximately -0.35℃, and the absolute value of the temperature change rate is approximately 0.03℃ / min. However, this does not yet meet the lower limit of the phase change temperature zone. The main control unit maintains the baseline operation and continues to make judgments. At t=42min, the cavity temperature is approximately -27.8℃, the center temperature enters -0.45℃, and the absolute value of the temperature change rate is 0.028℃ / min. This meets the triggering conditions for entering the preset phase change temperature zone and the temperature change rate meeting the threshold, and the phase change linkage control stage begins.

[0045] (5) Phase change linkage control stage (t=42min to t=70min)

[0046] After the phase change linkage begins, the main control unit sets the ultrasonic transducer to an intermittent beat with a start-up duration of 20s, a stop duration of 40s, and an ultrasonic power of 80W. The output of the refrigeration unit is then segmented according to the ultrasonic beat into start-up and stop-up segments. During the start-up segment, the refrigeration setting is switched to the high output level, and during the stop segment, it is switched to the slow stop buffer level. When switching between segments, the main control unit applies slope upper limit constraints to the compressor speed, evaporator temperature setting, electronic expansion valve position, and fan speed, so that the set values ​​gradually climb or fall in steps to avoid temperature oscillation and refrigeration unit shock. The main control unit also adaptively corrects the difference between the high output level and the reference level based on the center temperature fluctuation amplitude and the duration of the phase change platform.

[0047] Table 3. Data recorded in the example (sampled every 5-10 minutes)

[0048]

[0049] As can be seen from Table 3, at t=52min, the cavity temperature was -0.78℃, and the absolute value of the temperature change rate was approximately 0.014℃ / min, indicating that the phase transition plateau duration was relatively long. At t=60min, after parameter adjustment, the cavity temperature was approximately -0.95℃, and the absolute value of the temperature change rate recovered to 0.022℃ / min. At t=70min, in order to suppress fluctuations, the downtime of the ultrasonic transducer and the ultrasonic power were adjusted, and subsequently the cavity temperature rapidly dropped and left the phase transition window.

[0050] (6) Cryogenic propulsion and endpoint hold (t=85min to t=135min)

[0051] When the center temperature leaves the phase change temperature zone, the main control unit shuts down the ultrasound, and the refrigeration unit switches to deep cryogenic propulsion mode. At t=100min, the cavity temperature is approximately -10.2℃; after reaching -18.0℃ at t=115min, it enters the holding phase, and the refrigeration unit maintains a closed loop at 55-60Hz for 20 minutes before being removed from storage.

[0052] Table 4 Phase Change Cycle Log Table

[0053]

[0054] As shown in Table 4, this embodiment records the ultrasonic beat, cooling level, temperature response, and decision results sampled periodically (period = 60s) during the phase transition phase. Initially, the fluctuation amplitude A in periods 1-3 was 0.08-0.11℃, and the rate of change gradually decreased. In periods 4-5, the first Mahalanobis deviation exceeded the threshold, and the absolute value of the actual rate of change was less than the target, increasing the platform duration. This was determined to be insufficient propulsion. The main control unit increased the ultrasonic start-up time from 20s to 25s by a preset step size (effective from period 6), improving the absolute value of the temperature change rate from 0.014-0.016 to approximately 0.020-0.022℃ / min, accelerating platform propulsion. Subsequently, in period 8, the central temperature fluctuation amplitude reached 0. The temperature exceeded the preset fluctuation limit by 0.15℃ at 16℃, and the second and third Mach-law deviations increased, indicating a strong disturbance / high duty cycle. The main control unit increased the ultrasonic shutdown time from 40s to 50s and decreased the ultrasonic power P from 80W to 70W. Simultaneously, the ultrasonic start-up time was slightly reduced (compressor operating frequency adjusted from 78Hz to 76Hz, evaporation temperature increased from -34℃ to -33℃, and fan speed decreased from 1500rpm to 1450rpm). This caused the fluctuation to converge to 0.12-0.13℃ within a period of 9-10 and stably exit the phase transition stage. This process ensured that the ultrasonic energy input matched the phase transition kinetics closed loop, shortening the plateau while avoiding quality risks caused by excessive cavitation or temperature oscillations.

[0055] After freezing is completed, the main control unit generates statistical results in batches and writes them back to the knowledge graph for incremental updates of causal weights and confidence scores, as shown in Table 5:

[0056] Table 5 Batch Statistical Results

[0057]

[0058] Table 5 shows that the total freezing time (from loading to a center temperature of -18℃) was 115 min, the phase change platform time was approximately 23 min, the cumulative ultrasonic energy during the phase change stage was 0.078 kWh, which did not exceed the upper limit of 0.09 kWh; the refrigeration unit consumed a total of 1.92 kWh, the maximum short-term fluctuation of the center temperature was 0.16℃ (which dropped after a single trigger suppression), and the drip loss rate was 4.3%. The final phase change stage parameters adopted were: ultrasonic start-up time of 25 s, ultrasonic stop-up time of 50 s, ultrasonic power of 70 W, and the refrigeration setting during the ultrasonic start-up phase was a compressor inverter drive frequency of 76 Hz. The setpoint for the evaporation temperature is -33℃, the setpoint for the electronic expansion valve opening is 62%, the target speed for the circulating fan is 1450 rpm, the ultrasonic shutdown section is set to a compressor variable frequency drive frequency of 56 Hz, the setpoint for the evaporation temperature is -28℃, the setpoint for the electronic expansion valve opening is 48%, and the target speed for the circulating fan is 1050 rpm. At the same time, the peak deviation of the relative target center temperature curve is recorded as 0.35℃. The above write-back fields are used to update the causal weight parameters and confidence levels of attributes, parameters, and quality, and to provide higher hit rate parameter recommendations and more robust online adaptive boundaries for subsequent similar batches.

[0059] like Figure 2 As shown, the horizontal axis represents time, and the vertical axis represents the normalized amplitude. The blue square wave indicates that the ultrasonic transducer switches between the start-up phase (Ton, value 1) and the stop phase (Toff, value 0) according to the cycle. The orange curve indicates that the normalized output of the refrigeration unit changes synchronously with the cycle. After the core temperature of the seafood enters the phase change temperature zone, the main control unit links the output of the refrigeration unit with the intermittent cycle of the ultrasonic transducer in a segmented manner. That is, in the ultrasonic start-up phase, the refrigeration unit is switched to the high output level. By increasing the compressor frequency, lowering the evaporation temperature setting, or increasing the throttle valve opening, and increasing the air volume of the circulating fan, the latent heat extraction and the ultrasonic nucleation effect are superimposed in phase, accelerating the phase change and refining the ice crystals. In the ultrasonic stop phase, the refrigeration unit is returned to the baseline or slow-down level. By reducing the compressor frequency, raising the evaporation temperature setting, or reducing the valve position and reducing the air volume, the temperature field converges and stabilizes, and the core temperature oscillation and local supercooling are suppressed. The smooth transition of the orange curve also reflects the limiting control of the rate of change of each execution quantity, avoiding equipment shock and reducing energy consumption. In the Ton segment, the output is increased to a high level to enhance latent heat extraction and, together with ultrasonic mechanical disturbance, promote nucleation and refine ice crystals. In the Toff segment, the output drops back to the baseline or gradually decreases to suppress temperature oscillations and avoid overcooling and system shock. The rise and fall of the orange curve is a smooth exponential transition rather than a step, reflecting that the execution quantities such as compressor inverter speed, evaporator temperature setting, throttle valve opening, and circulating fan air volume are affected by the slope limitation and equipment thermal inertia. This achieves a cyclical progression of strong action - convergence stabilization - strong action again, making the ultrasonic energy input and cooling capacity output adapt and coordinate within the phase change window, improving freezing efficiency while balancing quality and energy consumption.

[0060] In summary, the system proposed in this invention, by enabling the intermittent ultrasonic action of the multi-transducer array only within the phase change window and linking it with the output of the refrigeration unit in segments according to the ultrasonic beat (Ton segment for cooling, Toff segment for convergence), can still improve the nucleation density and enhance latent heat extraction, and shorten the phase change plateau time, even when there is an air gap in loading and packaging. At the same time, by utilizing the Marsh deviation and the upper limit of fluctuation, the online adaptive Tn (ultrasonic start-up time) / Tm (ultrasonic stop-up time) / P (ultrasonic frequency) is achieved, which enhances the effect when the propulsion is insufficient and weakens the effect when the disturbance is too strong, thereby suppressing the core temperature oscillation, reducing ineffective energy consumption, and avoiding tissue damage caused by excessive local cavitation. Ultimately, it forms fine and uniform ice crystals, reduces dripping water loss, and improves the consistency of batch freezing quality.

[0061] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0062] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A microcrystalline resonance freezing system, characterized in that, It includes a freezing chamber, a refrigeration unit, a tray, an ultrasonic transducer, a temperature detection unit, and a main control unit. The refrigeration unit is connected to the freezing chamber for heat exchange. The ultrasonic transducer includes multiple ultrasonic transducers, which are arranged under the tray and coupled to the seafood through a sound guiding layer. The temperature detection unit includes a cavity temperature sensor and a seafood center temperature probe. When the center temperature enters the preset phase change temperature zone and the temperature change rate meets the preset threshold, the main control unit controls the ultrasonic transducer to work intermittently for a preset start-up time and a preset stop time, and simultaneously modulates the output of the refrigeration unit.

2. The microcrystalline resonance freezing system according to claim 1, characterized in that, The seafood center temperature probe is placed at the center of a representative heat load, which is a food-grade simulated load block of the same batch and specifications as the seafood to be frozen. The center hole is a pre-made hole. The seafood center temperature probe is equipped with a food-grade sealing sleeve and sealed with food-grade sealing material.

3. The microcrystalline resonance freezing system according to claim 1, characterized in that, The cavity temperature sensor is installed at the return air vent of the refrigeration cavity.

4. The microcrystalline resonance freezing system according to claim 1, characterized in that, The refrigeration control unit is connected to a refrigeration process knowledge graph library, which is a time-series causal knowledge graph for refrigeration control. It is constructed as stage-sliced ​​subgraphs according to the pre-cooling stage, subcooling stage, nucleation stage, phase transition stage, and cryogenic stage. The refrigeration process knowledge graph library uses seafood category, specifications, initial core temperature, packaging method, loading layer, tray-packaging coupling parameters, target core temperature curve, phase transition temperature zone parameters, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-down time, ultrasonic power, refrigeration unit target output, and quality quantification indicators as nodes. Each edge includes causal weights, placement... The confidence and applicable domain parameters, causal weights and confidence levels are updated incrementally by the deviation of the center temperature curve of historical batches, the duration of the phase change platform, and the quantitative indicators of energy consumption and quality. After receiving the cavity temperature and center temperature data, the main control unit maps the attribute characteristics of the current batch of seafood to the graph query conditions. First, it filters the candidate parameter set according to the applicable domain and hard constraint edges, and then recalls and sorts them in the stage slice subgraph according to the graph embedding similarity. It outputs the phase change temperature zone, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-up time, ultrasonic power and the parameter set of the target output of the refrigeration unit for the current batch of seafood, as well as the corresponding confidence levels.

5. The microcrystalline resonance freezing system according to claim 1, characterized in that, The main control unit generates a parameter set based on a collaborative recommendation mechanism. The collaborative object is the historical frozen batches. The main control unit encodes the category, specifications, packaging method, loading layer, initial center temperature, and target endpoint temperature of the current batch of seafood into user-side feature vectors. It encodes the phase change temperature zone parameters, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-up time, ultrasonic power, and target output of the refrigeration unit of each historical frozen batch into item-side feature vectors. Based on the comprehensive freezing effect score composed of center temperature curve similarity, freezing time, energy consumption, and drip loss, the main control unit constructs a user-item interaction matrix. The main control unit performs neighborhood collaborative filtering on the interaction matrix to obtain the predicted score of the candidate parameter set, and selects the candidate parameter set with the highest predicted score as the phase change temperature zone, temperature change rate threshold, ultrasonic start-up time, ultrasonic stop-up time, ultrasonic power, and target output of the refrigeration unit for the current batch of seafood.

6. The microcrystalline resonance freezing system according to claim 1, characterized in that, The main control unit performs online adaptive adjustment of the ultrasonic start-up duration: within the phase change temperature zone, the center temperature sequence within the sliding time window is taken, and after exponential weighted filtering, the actual temperature change rate is obtained by least squares fitting. The duration of the phase change platform and the center temperature fluctuation amplitude are statistically obtained to form the first actual feature vector. The first target feature vector is obtained from the target center temperature curve corresponding to the current freezing batch, and the first feature covariance matrix is ​​obtained from historical batches similar to the current batch. The first Mahalanobis deviation is calculated. When the first Mahalanobis deviation exceeds the first threshold and the absolute value of the actual temperature change is less than the absolute value of the target temperature change, and the duration of the phase change platform is greater than the duration of the target platform, the ultrasonic start-up duration is increased by a preset step size. When the first Mahalanobis deviation exceeds the first threshold and the center temperature fluctuation amplitude reaches or exceeds the preset fluctuation upper limit, the ultrasonic start-up duration is decreased by a preset step size, and the updated ultrasonic start-up duration is used for the next control cycle.

7. The microcrystalline resonance freezing system according to claim 1, characterized in that, The main control unit performs online adaptive adjustment of the ultrasonic downtime: within the preset phase change temperature zone, the core temperature of the seafood is sampled by a sliding time window, the effective proportion of ultrasound within the time window is calculated, and the core temperature fluctuation amplitude and the duration of the phase change platform are obtained to form a second actual feature vector. The second target feature vector is obtained from the target core temperature curve corresponding to the current frozen batch, and the second feature covariance matrix is ​​obtained from similar historical batches. The second Mahalanobis deviation is calculated. When the second Mahalanobis deviation exceeds the second threshold and the effective ultrasonic duty cycle exceeds the preset duty cycle upper limit or the core temperature fluctuation threshold reaches or exceeds the preset fluctuation upper limit, the ultrasonic downtime is increased by a preset step size. When the second Mahalanobis deviation exceeds the second threshold and the duration of the phase change platform is greater than the target platform duration and the core temperature fluctuation amplitude is less than the target fluctuation amplitude, the ultrasonic downtime is decreased by a preset step size, and the updated ultrasonic downtime is used for the next control cycle.

8. A microcrystalline resonance freezing system according to claim 1, characterized in that, The main control unit performs online adaptive adjustment of ultrasonic power: within the preset phase change temperature zone, the actual temperature change rate, the duration of the phase change platform, and the amplitude of the center temperature fluctuation are obtained based on a sliding time window to form a third target feature vector. The target center temperature curve corresponding to the current freezing batch is used to obtain the third target feature vector, and the third feature covariance matrix is ​​obtained from similar historical batches. The third Mahalanobis bias is calculated. At the same time, the power sensitivity is obtained based on the changes in ultrasonic power and temperature change rate over adjacent time periods. When the third Mahalanobis bias exceeds the third threshold, the absolute value of the actual temperature change rate is less than the absolute value of the target temperature change rate, the duration of the phase change platform is greater than the duration of the target platform, and the cumulative ultrasonic energy has not reached the preset energy upper limit, the ultrasonic power is increased by a preset power increment determined according to the power sensitivity. When the amplitude of the low fluctuation of the center temperature is greater than or equal to the preset fluctuation upper limit or the cumulative ultrasonic energy reaches the preset upper limit, the ultrasonic power is decreased by a preset power step size, and the updated ultrasonic power is used for the next control cycle.

9. A microcrystalline resonance freezing system according to claim 1, characterized in that, When the center temperature enters the preset phase change temperature zone and the temperature change rate meets the preset threshold, the main control unit will link the output of the refrigeration unit with the intermittent beat of the ultrasonic transducer.

10. A microcrystalline resonance freezing system according to claim 1, characterized in that, In the main control unit, the intermittent beat linkage modulation method between the refrigeration unit output and the ultrasonic transducer is as follows: when the core temperature of the seafood enters the preset phase change temperature zone and the temperature change rate meets the preset threshold, the ultrasonic transducer is placed in a periodic intermittent working mode with a start-up duration and a stop duration, and the refrigeration unit output is set to start-up segment output and stop segment output according to the intermittent beat; in the start-up segment of each cycle, the refrigeration unit output is set to high output level, which is achieved by increasing the duty cycle of the compressor variable frequency speed of the refrigeration unit, and at least one of the following is executed simultaneously: the evaporation temperature setting is lowered or the throttle valve opening is increased, and the circulating fan speed is set to high air volume level. During the shutdown phase of each cycle, the output of the refrigeration unit is set to either the reference output level or the slow-release output level. The reference output level or the slow-release output level is achieved by reducing the duty cycle of the compressor inverter speed drive of the refrigeration unit, and at least one of the following is executed simultaneously: evaporator temperature setpoint recovery or throttle valve opening reduction. At the same time, the circulating fan speed is set to the medium air volume level. Between the start-up and shutdown phases, upper limit constraints are applied to the compressor speed change rate, evaporator temperature setpoint change rate, and circulating fan speed change rate, respectively. The difference between the high output level and the reference output level is adaptively corrected based on the center temperature fluctuation amplitude and the duration of the phase change platform.

Images