Method for improving nutrition density of sweet potato granules through low-temperature vacuum frying

Abstract

The invention provides a method for improving the nutrient density of sweet potato granules through low-temperature vacuum frying, and belongs to the technical field of food processing. Accurate temperature control is realized by establishing a sweet potato granule temperature gradient mathematical model and adopting a segmented temperature rise control strategy; uniform moisture migration is promoted by combining periodic vacuum degree adjustment of a pulse vacuum technology and the synergistic effect of a 40kHz ultrasonic-assisted mass transfer system, and an upper layer model with maximum temperature uniformity and a lower layer model with maximum moisture migration rate are established through a double-layer game optimization model to realize multi-target collaborative optimization. An infrared temperature monitoring system is used for monitoring a temperature uniformity evaluation index in real time and dynamically adjusting differential heating power distribution, and finally, an infrared thermal imager is used for monitoring the surface temperature stability to ensure the machining quality. The technical problem that in the low-temperature vacuum frying process of sweet potato granules, internal and external temperature distribution is not uniform, and consequently nutrient loss is serious is solved.

Description

Technical Field

[0001] This invention belongs to the field of food processing technology, and specifically relates to a method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying. Background Technology

[0002] Vacuum frying technology, as an advanced food processing method, is widely used in the production of products such as fruit and vegetable crisps and potato chips. It reduces frying temperature by lowering environmental pressure, thereby minimizing heat loss of nutrients. Traditional vacuum frying processes employ constant temperature and fixed vacuum levels, and have been widely used in processing block or sheet-like agricultural products such as sweet potato chunks, apple slices, and carrot strips, especially in industrial production lines where a relatively mature process has been established. However, traditional vacuum frying technology has significant technical shortcomings when processing block-shaped materials with complex internal structures, such as sweet potato chunks. The main issue is uneven temperature transfer during heating, resulting in excessively high temperatures on the outer layer and insufficient temperatures on the inner layer. This leads to excessive degradation of nutrients in the outer layer and insufficient processing in the inner layer. In existing technologies, due to the lack of precise temperature gradient control and moisture migration regulation mechanisms, a significant temperature difference exists between the inner and outer parts of the sweet potato chunk. Under high temperatures, the outer layer undergoes excessive Maillard reactions and lipid oxidation, resulting in the significant loss of heat-sensitive nutrients such as vitamins and carotene. Simultaneously, the inner layer, due to insufficient temperature, cannot adequately dehydrate and form an ideal tissue structure. In other words, existing technologies have a technical problem: uneven temperature distribution inside and outside sweet potato pieces during low-temperature vacuum frying leads to significant loss of nutrients. Summary of the Invention

[0003] In view of this, the present invention provides a method for improving the nutritional density of sweet potato granules by low-temperature vacuum frying, which can solve the technical problem in the prior art that the uneven internal and external temperature distribution of sweet potato granules during low-temperature vacuum frying leads to serious loss of nutrients.

[0004] This invention is implemented as follows: A method for improving the nutritional density of sweet potato granules during low-temperature vacuum frying is provided. Fresh sweet potatoes are washed, peeled, and cut into cubic granules. These granules are then soaked in a citric acid solution to remove surface starch and undergo a color-protecting pretreatment, resulting in pretreated sweet potato granules. A mathematical model of the temperature gradient of the sweet potato granules is established. An infrared temperature monitoring system is installed in the vacuum frying equipment. Segmented temperature control strategy parameters are calculated using a temperature control parameter calculation equation, and these parameters are used for temperature control. Pulsed vacuum technology is used for dynamic vacuum degree regulation. Pulsed vacuum control parameters are calculated using a vacuum degree optimization calculation equation, and the vacuum degree is adjusted according to these parameters to promote uniform moisture migration within the sweet potato granules. An ultrasonic-assisted mass transfer system is activated, operating synchronously with the pulsed vacuum to improve internal moisture mass transfer efficiency. A higher-level game theory model is established with the goal of maximizing temperature uniformity and a lower-level game theory model with the goal of maximizing moisture migration rate. The target's lower-level game model solves for the optimal heating power distribution through a two-level game optimization model, monitors the uniformity evaluation index of sweet potato grain surface temperature in real time, and adjusts the heating power distribution mode or switches to a differentiated heating mode. The moisture migration process is modeled as a shortest path problem, and the Dijkstra algorithm is used to optimize the moisture migration path. Sweet potato grain samples are captured, and the moisture content of different layers of the sample is detected by a Karl Fischer moisture analyzer, and the rate of change of moisture gradient is calculated. The vacuum degree and ultrasonic treatment time are adjusted, or heating is paused and the pulse vacuum intensity is increased. When the frying time reaches the preset total time and the average surface temperature of the sweet potato grain is stable as detected by an infrared thermal imager, the frying process is ended, and the sweet potato grain is cooled to room temperature using a gradient cooling method to obtain a sweet potato grain product with improved nutritional density. The temperature gradient mathematical model and the two-level game optimization model achieve the synergistic effect of temperature distribution uniformity control and moisture migration efficiency optimization.

[0005] The frying process ends when the frying time reaches the preset total time and the average surface temperature of the sweet potato pieces remains stable within the range of 80±3℃ for 3 minutes as detected by an infrared thermal imager.

[0006] Specifically, the step of soaking in citric acid solution involves soaking in 0.5% citric acid solution for 15 minutes.

[0007] The temperature control parameter calculation equation is used to calculate the temperature setpoint and heating rate at each stage of the segmented heating process. The inputs include the geometric size of sweet potato grains, initial temperature distribution, target temperature and thermal conductivity coefficient, and the output is the temperature control parameters for each time period.

[0008] The vacuum degree optimization calculation equation is used to determine the vacuum degree change period and amplitude during the pulsed vacuum process. The inputs include the moisture content of sweet potato grains, porosity, temperature and pressure gradient, and the output is the pulsed vacuum control parameters.

[0009] The ultrasonic-assisted mass transfer system has an ultrasonic frequency set to 40kHz and a power density of 0.3W per square centimeter.

[0010] Specifically, when the temperature uniformity evaluation index ∈ [0, 0.2), the heating power distribution mode is adjusted to a uniform heating state; when the temperature uniformity evaluation index ∈ [0.6, 1], the differential heating mode parameters are calculated through the differential heating power calculation function and the differential heating mode is switched.

[0011] In this process, three sweet potato samples are collected every 5 minutes. When the moisture gradient change rate is ∈ (15, 25]% / min, the vacuum level is reduced to -0.06MPa and the ultrasonic treatment time is extended to 1.5 times the original set value. When the moisture gradient change rate is >25% / min, heating is paused and the pulse vacuum intensity is increased.

[0012] The differentiated heating power calculation function is used to calculate the heating power distribution curves of different regions. The inputs include the temperature deviation, heat capacity, heat transfer coefficient and target temperature of each region, and the output is the heating power curve of each region.

[0013] The temperature uniformity evaluation index refers to the ratio of the standard deviation of the temperature difference between each point on the surface of the sweet potato grain and the average temperature to the average temperature, which is used to quantify the uniformity of temperature distribution.

[0014] Optionally, the step of making cubic particles specifically involves washing and peeling fresh sweet potatoes and then cutting them into cubic particles with a side length of 8mm to 10mm.

[0015] The objective function of the upper-level game model is: The constraints are: temperature uniformity evaluation index ≥ 0.3 and total heating power ≤ set upper limit, where the coupling term is... With the lower-level model This establishes a power coupling relationship. The objective function of the lower-level game model is: The constraints are that the rate of change of moisture gradient is ≤30% / min and the vacuum degree is within the allowable range of the equipment.

[0016] Optional, the preset total time is 90 to 180 seconds.

[0017] The segmented heating control strategy refers to dividing the heating process into multiple temperature stages based on the temperature conduction characteristics inside and outside the sweet potato grains, with each stage having a different heating rate and duration.

[0018] The pulsed vacuum technology refers to periodically adjusting the vacuum level during vacuum frying, generating pressure pulsations through intermittent changes in vacuum level, which promotes the uniform migration of moisture from the inside of sweet potato grains to the surface.

[0019] The two-layer game optimization model refers to a multi-objective optimization model established using Stackelberg game theory. The upper-layer model acts as the leader to optimize temperature control, while the lower-layer model acts as the follower to optimize water migration. Overall optimization is achieved through game equilibrium.

[0020] This invention effectively solves the problem of uneven temperature distribution in traditional vacuum frying by establishing a mathematical model of the temperature gradient of sweet potato grains and adopting a segmented heating control strategy, combined with the synergistic effect of pulsed vacuum technology and an ultrasonic-assisted mass transfer system. This invention achieves simultaneous optimization of temperature uniformity and moisture migration efficiency through a two-layer game optimization model. The upper-layer model adjusts the heating power distribution through real-time monitoring of the temperature uniformity evaluation index, while the lower-layer model optimizes the vacuum degree and ultrasonic parameters through dynamic detection of the moisture gradient change rate. This ensures precise temperature control and uniform moisture migration throughout the sweet potato grains, avoiding overheating of the outer layer and underprocessing of the inner layer, thus maximizing the preservation of vitamins, carotene, dietary fiber, and other nutrients in the sweet potato. This invention optimizes the moisture migration path using the Dijkstra algorithm, combined with precise detection by a Karl Fischer moisture analyzer and temperature monitoring by an infrared thermal imager, achieving precise control of the processing and effective protection of nutrients. Technically, this invention completely solves the technical problem of severe nutrient loss due to uneven temperature distribution inside and outside the sweet potato grains in existing technologies. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method of the present invention.

[0022] Figure 2 This is a monitoring diagram of the working status of the pulse vacuum system in Example 2.

[0023] Figure 3 This is a diagram showing the convergence process of the objective function in the two-layer game model in Example 2.

[0024] Figure 4 This is a data acquisition diagram of the multi-sensor fusion monitoring system in Example 2. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0026] like Figure 1 The diagram shown is a flowchart of a method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying, provided by the present invention. This method includes the following steps:

[0027] S01. After washing and peeling fresh sweet potatoes, cut them into cubic granules with a side length of 8mm. Soak them in a 0.5% citric acid solution for 15 minutes to remove surface starch and perform color protection pretreatment to obtain pretreated sweet potato granules.

[0028] S02. Establish a mathematical model of sweet potato grain temperature gradient, install an infrared temperature monitoring system in the vacuum frying equipment, calculate the segmented heating control strategy parameters through the temperature control parameter calculation equation, and use the parameters for temperature control.

[0029] S03. Employ pulsed vacuum technology for dynamic vacuum control. Calculate pulsed vacuum control parameters using a vacuum optimization calculation equation. Adjust the vacuum level according to the pulsed vacuum control parameters to promote uniform moisture migration inside sweet potato grains.

[0030] S04. Start the ultrasonic-assisted mass transfer system. Set the ultrasonic frequency to 40kHz and the power density to 0.3W per square centimeter. Synchronize with the pulsed vacuum to improve the internal moisture mass transfer efficiency.

[0031] S05. Establish an upper-level game model with the goal of maximizing temperature uniformity and a lower-level game model with the goal of maximizing moisture migration rate. Solve the optimal heating power distribution through the two-level game optimization model. Monitor the evaluation index of temperature distribution uniformity on the surface of sweet potato grains in real time. When the evaluation index of temperature uniformity ∈ [0, 0.2), adjust the heating power distribution mode to uniform heating state. When the evaluation index of temperature uniformity ∈ [0.6, 1], calculate the parameters of the differentiated heating mode through the differentiated heating power calculation function and switch to the differentiated heating mode.

[0032] S06. The moisture migration process is modeled as a shortest path problem. The Dijkstra algorithm is used to optimize the moisture migration path. Three sweet potato samples are collected every 5 minutes. The moisture content of different layers of the sample is detected by a Karl Fischer moisture analyzer and the moisture gradient change rate is calculated. When the moisture gradient change rate ∈ (15, 25]% / min, the vacuum degree is reduced to -0.06MPa and the ultrasonic treatment time is extended to 1.5 times the original set value. When the moisture gradient change rate > 25% / min, heating is paused and the pulse vacuum intensity is increased.

[0033] S07. When the frying time reaches the preset total time and the average surface temperature of the sweet potato pieces is stable within the range of 80±3℃ for 3 minutes as detected by infrared thermal imager, the frying process is ended, and the sweet potato pieces are cooled to room temperature using a gradient cooling method to obtain a sweet potato product with improved nutritional density.

[0034] The temperature control parameter calculation equation is used to calculate the temperature setpoint and heating rate at each stage of the segmented heating process. The inputs include the geometric size of sweet potato grains, initial temperature distribution, target temperature and thermal conductivity coefficient, and the output is the temperature control parameters for each time period.

[0035] The vacuum degree optimization calculation equation is used to determine the vacuum degree change period and amplitude during the pulsed vacuum process. The inputs include the moisture content of sweet potato grains, porosity, temperature and pressure gradient, and the output is the pulsed vacuum control parameters.

[0036] The differentiated heating power calculation function is used to calculate the heating power distribution curves of different regions. The inputs include the temperature deviation, heat capacity, heat transfer coefficient and target temperature of each region, and the output is the heating power curve of each region.

[0037] The objective function of the upper-level game model is: The constraints are: temperature uniformity evaluation index ≥ 0.3 and total heating power ≤ set upper limit, where the coupling term is... With the lower-level model A power coupling relationship is formed.

[0038] The objective function of the lower-level game model is: The constraints are that the rate of change of moisture gradient is ≤30% / min and the vacuum degree is within the allowable range of the equipment.

[0039] The objective function of the upper-level game model is used to optimize the uniformity of temperature distribution. The inputs include the temperature uniformity evaluation index, average temperature, heating power, time and heat transfer, and the output is the optimal temperature control strategy.

[0040] The objective function of the lower-level game model is used to optimize the water migration efficiency. The inputs include water migration rate, heating power, pressure difference, time and migration resistance, and the output is the optimal water migration control strategy.

[0041] The temperature uniformity evaluation index refers to the ratio of the standard deviation of the temperature difference between each point on the surface of the sweet potato grain and the average temperature to the average temperature, which is used to quantify the uniformity of temperature distribution.

[0042] The segmented heating control strategy refers to dividing the heating process into multiple temperature stages based on the temperature conduction characteristics inside and outside the sweet potato grains. Each stage is set with a different heating rate and duration to avoid the phenomenon of the outer layer being overheated while the inner layer is underheated.

[0043] Among them, pulsed vacuum technology refers to periodically adjusting the vacuum level during vacuum frying. The intermittent changes in vacuum level generate pressure pulsations, which promote the uniform migration of moisture from the inside of sweet potato grains to the surface.

[0044] Among them, the Karl Fischer moisture analyzer refers to a moisture detection device that uses the Karl Fischer titration method to accurately determine the moisture content in a sample through a chemical titration reaction, and is used to quantitatively analyze the moisture distribution at different levels of sweet potato grains.

[0045] The differentiated heating mode refers to adjusting the power distribution of the heating element according to the temperature distribution of different parts of the sweet potato, increasing the heating power in areas with lower temperatures and reducing the heating power in areas with higher temperatures.

[0046] Among them, the moisture gradient change rate refers to the percentage change in the difference between the internal and surface moisture content of sweet potato grains per unit time, which is used to determine whether there are any abnormalities in the moisture migration process.

[0047] Among them, the two-layer game optimization model refers to a multi-objective optimization model established using Stackelberg game theory. The upper-layer model acts as the leader to optimize temperature control, while the lower-layer model acts as the follower to optimize water migration. Overall optimization is achieved through game equilibrium.

[0048] The pulsed vacuum control parameters include the vacuum degree change period, the vacuum degree change amplitude, and the pulse duration, which are used to control the implementation process of pulsed vacuum technology.

[0049] The specific implementation methods of the above steps are described in detail below.

[0050] The specific implementation of step S01 is as follows: First, a soft-bristled brush is used in conjunction with running water to wash the surface of fresh sweet potatoes. During the washing process, the water temperature is controlled between 15 and 20°C. The outer skin of the sweet potatoes is removed using a mechanical peeling device, with the peeling thickness controlled between 1.5 and 2 mm. Then, a food-grade stainless steel cutting device is used to cut the peeled sweet potatoes into cubic particles with a side length of 8 mm. The cutting process adopts a constant speed cutting method, with the cutting speed set at 50 mm per second to ensure a smooth and flat cut surface. Next, the cut sweet potato particles are immediately immersed in a pre-prepared 0.5% citric acid solution. The solution temperature was maintained at 4 to 8°C, and the soaking time was 15 minutes. During the soaking process, gentle stirring was performed every 3 minutes at a stirring speed of 20 revolutions per minute. The acidic environment of citric acid inhibited the activity of polyphenol oxidase, preventing browning of sweet potato granules. At the same time, the citric acid solution could dissolve the free starch on the surface of the sweet potato granules, reducing starch gelatinization during frying. After soaking, a vibrating draining device was used to remove the citric acid solution adhering to the surface. The vibration frequency was set to 30 Hz, and the draining time was 2 minutes, resulting in pre-treated sweet potato granules with a clean surface and good color retention.

[0051] The specific implementation of step S02 is as follows: A three-dimensional temperature field distribution mathematical model of sweet potato grains is established based on Fourier's law of heat conduction. This model considers the nonlinear characteristics of the thermal properties of sweet potato grains, including specific heat capacity, thermal conductivity, and density, as a function of temperature. Twelve infrared temperature sensors are uniformly arranged within the heating chamber of the vacuum frying equipment. These sensors employ long-wave infrared detectors with wavelengths ranging from 8 to 14 micrometers, achieving a temperature measurement accuracy of ±0.5℃ and a response time of less than 100 milliseconds. The temperature control parameter calculation is performed using an equation whose input parameters include the geometric dimensions of the sweet potato grains (8 mm) and the initial temperature distribution (20 to 14 mm). With a target temperature of 80℃ and a thermal conductivity of 0.45 to 0.55 W / m Kelvin, the partial differential equations of the temperature field were solved using the finite element method. The frying process was divided into three temperature control stages: the first stage was a rapid heating period, in which the temperature rose from room temperature to 60℃ at a rate of 8℃ per minute; the second stage was a slow heating period, in which the temperature rose from 60℃ to 75℃ at a rate of 3℃ per minute; and the third stage was a constant temperature period, in which the temperature was kept stable within the range of 75 to 80℃. This segmented control strategy prevented the formation of a hard crust on the surface of the sweet potato pieces due to rapid heating, while ensuring that the inside was fully heated.

[0052] The specific implementation of step S03 is as follows: A programmable logic controller (PLC) is used to control the vacuum pump and solenoid valve to achieve pulsed vacuum technology. The vacuum degree changes periodically between -0.08 MPa and -0.04 MPa, forming a pressure pulsation effect. The pulse parameters are determined by a vacuum degree optimization calculation equation. The inputs to this equation include the initial moisture content of sweet potato grains (65-70%), porosity (0.15-0.20), current temperature, and pressure gradient. The constrained optimization problem is solved using the Lagrange multiplier method, and the optimal pulse period is found to be 30 seconds, with a high vacuum holding time of 20 seconds and a low vacuum holding time of 10 seconds. The vacuum degree change is transitioned in the form of a sine wave, and the transition time is controlled within 2 seconds to avoid damage to the sweet potato grain structure caused by rapid pressure changes. The pressure gradient driving force generated by the pulsed vacuum promotes the conversion of bound water inside the sweet potato grains into free water, accelerating the migration of water from the inside to the surface. At the same time, the periodic pressure change prevents excessive oil penetration into the sweet potato grains, reducing the oil content of the product.

[0053] The specific implementation of step S04 is as follows: a piezoelectric ceramic ultrasonic transducer array is installed at the bottom of the frying chamber. The transducer adopts a sandwich structure design with a resonant frequency of 40 kHz. An impedance matching network ensures that the energy transmission efficiency is greater than 85%. The ultrasonic power density is set to 0.3 W / cm². The ultrasonic output is controlled by pulse modulation with a pulse duty cycle of 70%, i.e., a cycle of 7 seconds of operation followed by 3 seconds of rest. The ultrasonic waves propagate in the oil, generating cavitation and acoustic flow effects. The formation and collapse of cavitation bubbles generate micro-jets on the surface of the sweet potato particles, breaking the boundary layer resistance. The acoustic flow effect forms a convection circulation in the oil, enhancing the heat and mass transfer efficiency. The ultrasonic waves and pulsed vacuum work synchronously and in coordination. When the vacuum level is high, the ultrasonic power increases by 20%. When the vacuum level is low, the ultrasonic power returns to normal. This synergistic effect increases the internal water migration rate of the sweet potato particles by 35% to 45%, while reducing the frying time by 15% to 20%.

[0054] The specific implementation of step S05 is as follows: A two-layer optimization model based on Stackelberg game theory is constructed. The upper-layer model, acting as the leader, aims to maximize temperature uniformity. Its objective function includes a logarithmic term for temperature uniformity, a quadratic term for average temperature, a square root term for heating power, a sine term for the time period, and a linear term for heat transfer. A sequential quadratic programming algorithm is used to solve the upper-layer optimization problem. The lower-layer model, acting as the follower, aims to maximize the water migration rate. Its objective function includes a 1.2 power term for water migration rate, a 0.8 power term for heating power, an exponential decay term for pressure difference, a cosine period term for time, and a 0.6 power term for migration resistance. The interior-point method is used to solve the lower-layer optimization problem. The two models achieve information exchange through a heating power coupling term. The process iterates until Nash equilibrium is reached, collecting temperature data from 16 measuring points on the surface of sweet potato grains in real time. The ratio of the temperature standard deviation to the average temperature is calculated as a temperature uniformity evaluation index. When the index is less than 0.2, the temperature distribution is considered too concentrated, and the system switches to uniform heating mode, maintaining consistent power for all heating elements. When the index is between 0.6 and 1, the temperature distribution is considered uneven, and a differentiated heating power calculation function is activated. This function calculates the power distribution curve based on the temperature deviation of each region, the heat capacity (1.8 to 2.2 kJ / kg Kelvin), the heat transfer coefficient, and the target temperature. For regions with temperatures below the average, the heating power is increased by 10% to 30%, while for regions with temperatures above the average, the heating power is reduced by 5% to 15%.

[0055] The specific implementation of step S06 is as follows: the internal water migration process of sweet potato grains is abstracted as a shortest path problem in graph theory. The sweet potato grains are discretized into three-dimensional mesh nodes, each node representing a micro-volume. The edge weights between nodes are determined by water migration resistance, including capillary resistance, diffusion resistance, and osmotic resistance. The Dijkstra algorithm is used to calculate the optimal migration path from internal high-moisture nodes to surface low-moisture nodes. The algorithm complexity is the logarithm of the square of the number of nodes. Every 5 minutes, three sweet potato grain samples are randomly collected using a vacuum suction device. The samples are immediately sent to a Karl Fischer moisture analyzer for layered moisture detection. The samples are divided into three parts: an outer layer, a middle layer, and an inner layer, each approximately 2.7 mm thick. The reagent uses a pyridine-free formulation, and the titration endpoint is determined by the double platinum electrode method with a measurement accuracy of 0.01%. The moisture gradient change rate is obtained by calculating the ratio of the difference in moisture content between adjacent layers to time. When the change rate is in the range of 15 to 25% per minute, it indicates that the moisture migration speed is too fast, which may lead to excessive surface dehydration. At this time, the vacuum degree is adjusted to -0.06 MPa to reduce the pressure driving force, and the ultrasonic treatment time is extended to 1.5 times the original value to enhance the mass transfer effect. When the change rate exceeds 25% per minute, it indicates that the moisture migration is too violent. Heating is paused for 30 seconds to avoid local overheating, and the pressure change amplitude of the pulse vacuum is increased by 30% to regulate the moisture distribution through stronger pressure pulsation.

[0056] The specific implementation of step S07 is as follows: the end of the frying process is determined using both time and temperature standards. When the cumulative frying time reaches the preset total time of 25 to 30 minutes, the surface temperature distribution of the sweet potato grains is continuously monitored using an infrared thermal imager. The thermal imager uses an uncooled microbolometer detector with a resolution of 320×240 pixels and a temperature resolution of 0.05℃, acquiring 30 frames of image data per second. The outline of the sweet potato grains is extracted using an image processing algorithm, and the average surface temperature is calculated. When the average surface temperature remains stable within the range of 80±3℃ for 3 consecutive minutes, and the temperature fluctuation is less than 1℃ per minute, the process is considered complete. Once the deep-frying process is complete, heating is immediately stopped and the ultrasonic system is turned off. A gradient cooling method is used for cooling. In the first stage, the temperature is naturally cooled in a vacuum environment for 5 minutes, dropping to 60°C. In the second stage, the temperature is slowly restored to normal pressure while filtered clean air is introduced, dropping to 40°C. In the third stage, the temperature continues to cool to room temperature under normal pressure. The entire cooling process takes 15 to 20 minutes. Gradient cooling avoids thermal stress cracking caused by rapid cooling and prevents moisture reabsorption. The final product is sweet potato granules with a moisture content of 3 to 5%, an oil content of 18 to 22%, and a nutrient retention rate of more than 85%.

[0057] It should be noted that the key technical ideas of this invention include the following aspects: The first key technical idea is to use a two-layer game optimization model to coordinate the contradictory relationship between temperature control and moisture migration. Traditional frying processes often optimize temperature or vacuum parameters separately, ignoring the coupling effect between the two. This invention establishes upper and lower layer optimization models through Stackelberg game theory. The upper layer aims at temperature uniformity, and the lower layer aims at moisture migration efficiency. Information interaction and collaborative optimization are achieved through heating power coupling terms. Compared with traditional methods, this can simultaneously take into account temperature distribution uniformity and moisture migration efficiency, avoiding the local optimum problem caused by single-objective optimization, and achieving global optimization of overall process parameters. The second key technical idea is the synergistic mechanism of pulsed vacuum and ultrasonic-assisted mass transfer. Traditional vacuum frying uses a constant vacuum degree, and the driving force for moisture migration is singular and easily reaches an equilibrium state. This invention generates pressure pulsations by periodically adjusting the vacuum degree, breaking the mass transfer equilibrium. At the same time, it introduces ultrasonic cavitation effect and acoustic flow effect. The synergistic effect of the two physical fields significantly improves mass transfer efficiency and reduces frying time and energy consumption. The third key technical approach is the optimization of moisture migration paths based on graph theory's shortest path algorithm. Traditional methods lack a precise description of the microscopic mechanisms of moisture migration. This invention models the moisture migration process as a network graph structure, uses Dijkstra's algorithm to find the optimal migration path, and dynamically adjusts process parameters based on path resistance distribution, achieving precise control of the moisture migration process. The synergistic effect of these technical approaches has produced significant technical results. The two-layer game model provides a global optimization framework, ensuring the balance of the two key factors of temperature and moisture. Pulsed vacuum and ultrasound provide physical means to enhance heat and mass transfer, and the shortest path algorithm provides a precise control method at the microscopic level. The three work together to form a complete process optimization system. Compared with traditional vacuum frying processes, this invention not only improves the consistency of product quality and the retention rate of nutrients, but also significantly shortens processing time, reduces energy consumption, and achieves intelligent and precise control of the food processing process.

[0058] It should be noted that in traditional vacuum frying, moisture migration within sweet potato grains often follows a random diffusion pattern, lacking effective path planning and mass transfer optimization. This results in low efficiency of moisture migration from the interior to the surface, leading to moisture accumulation in some areas and affecting the quality and nutrient retention of the final product. This invention models the moisture migration process as a shortest path problem and uses Dijkstra's algorithm to optimize the moisture migration path in the porous medium structure of sweet potato grains, finding the shortest mass transfer path from each point inside to the surface. This effectively reduces mass transfer resistance and improves moisture migration efficiency. Simultaneously, a Karl Fischer moisture analyzer accurately detects the moisture content at different levels and calculates the rate of change of the moisture gradient in real time. When abnormal moisture migration is detected, the pulse vacuum intensity and ultrasonic treatment time are automatically adjusted to ensure the stability and uniformity of the moisture migration process, thus solving the dehydration efficiency problem caused by unreasonable moisture migration paths in traditional technologies. Furthermore, the vacuum frying process involves multiple process parameters such as temperature, pressure, and ultrasonic power. These parameters have complex interrelationships, and traditional single-parameter control or simple multi-parameter combination control is insufficient to achieve overall process optimization. This often results in some parameters being optimized while others deviate from their optimal state, affecting processing results and nutrient retention efficiency. This invention establishes a two-layer game optimization model, transforming the complex multi-parameter coordination problem into a Stackelberg game problem. The upper-layer model acts as the leader, optimizing the temperature control strategy, while the lower-layer model acts as the follower, optimizing the moisture migration control strategy. The game equilibrium mechanism achieves coordinated optimization of each parameter. The coupling term design in the model ensures effective information transmission and parameter coordination between the upper and lower layers, avoiding mutual interference between parameters and simultaneously maximizing temperature uniformity and moisture migration efficiency. This solves the technical challenge of coordinating and optimizing multi-parameter systems.

[0059] Specifically, the principle of this invention is as follows: This invention can solve the technical problem of nutrient loss caused by uneven temperature distribution inside and outside sweet potato grains, mainly based on the following technical principles: First, by establishing a mathematical model of the temperature gradient of sweet potato grains, the heat transfer characteristics and temperature distribution law inside the sweet potato grains are accurately described, laying a theoretical foundation for achieving precise temperature control; the segmented heating control strategy decomposes the heating process into multiple stages according to the geometric size and thermal conductivity of the sweet potato grains, with different heating rates set for each stage, avoiding the problem of excessive internal and external temperature differences caused by traditional constant temperature heating methods. Second, pulsed vacuum technology generates pressure pulsations by periodically adjusting the vacuum level, effectively promoting the uniform migration of moisture inside the sweet potato grains. Combined with the cavitation effect of 40kHz ultrasound, it significantly improves mass transfer efficiency, enabling internal moisture to transfer to the surface in a timely manner and preventing the phenomenon of delayed internal temperature transfer. The core of the two-layer game optimization model lies in establishing a coupling relationship between temperature control and moisture migration. The upper-layer model aims to maximize temperature uniformity by real-time monitoring of the temperature uniformity evaluation index and dynamic adjustment of differentiated heating power to ensure temperature consistency across different parts of the sweet potato. The lower-layer model aims to maximize the moisture migration rate by accurately detecting the rate of change of moisture gradient and optimizing the pulsed vacuum parameters to ensure the uniformity and stability of the moisture migration process. The application of the Dijkstra algorithm in moisture migration path optimization transforms the complex porous media mass transfer process into a shortest path problem. By finding the optimal moisture migration path through the algorithm, mass transfer resistance is reduced and overall mass transfer efficiency is improved. The entire technical solution achieves real-time monitoring and precise control of the processing process through the collaborative work of precision detection equipment such as an infrared temperature monitoring system, a Karl Fischer moisture analyzer, and an infrared thermal imager, thereby fundamentally solving the problem of uneven temperature distribution in traditional technologies.

[0060] The following provides a specific embodiment 1 of the present invention, and the specific implementation of each step in this embodiment 1 is described in detail below.

[0061] In this embodiment, the specific implementation of step S01 is the same as described above, and will not be repeated in detail here.

[0062] The specific implementation of step S02 is to establish a three-dimensional temperature field distribution mathematical model for sweet potato grains based on Fourier's law of heat conduction. The calculation equation for the temperature control parameters of this model is as follows:

[0063] ;

[0064] In the formula, Spatial coordinates In time The temperature, in Kelvin. The coordinates are along the length of the sweet potato grains, in meters. The coordinates represent the width of the sweet potato grains, in meters. Here are the coordinates of the height of the sweet potato grains, in meters. The initial ambient temperature is defined as 293.15–298.15 Kelvin. For the first The first-order modal temperature amplitude coefficient, in Kelvin, For the first First-order mode decay coefficient, in units of per second. For the first First mode in Wave number in direction, in units of meters. For the first First mode in Wave number in direction, in units of meters. For the first First mode in Wave number in direction, in units of meters. The total number of modes can be 8 to 12. Heating time, in seconds. The parameters of the segmented heating control strategy are obtained by solving partial differential equations:

[0065] ;

[0066] in This is the thermal diffusivity, expressed in square meters per second. The thermal conductivity is taken as 0.45–0.55 W / m Kelvin. Density, expressed in kilograms per cubic meter. Specific heat capacity, measured in joules per kilogram (Kelvin). Heat output per unit volume, expressed in watts per cubic meter.

[0067] The specific implementation of step S03 is to use pulsed vacuum technology, and its vacuum degree optimization calculation equation is as follows:

[0068] ;

[0069] In the formula, For time The vacuum level, measured in megapascals (MPa). The basic vacuum level is set to -0.08 MPa. The vacuum degree variation range is set to 0.04 MPa. The pulse period is set to 30 seconds. The pulse duration is expressed in seconds. The pulse start time, in seconds. The pressure decay time constant is expressed in seconds. The pulsed vacuum control parameters include the vacuum degree change period, the vacuum degree change amplitude, and the pulse duration, which are determined using the moisture migration kinetic equation:

[0070] ;

[0071] in Moisture content, expressed in kilograms. The mass transfer coefficient is expressed in kilograms per second (kg / s) in megapascals per square meter (MPa). This is the saturated vapor pressure, measured in megapascals (MPa). This is the actual pressure, measured in megapascals (MPA). Surface area, in square meters. This represents the mass transfer time, measured in seconds.

[0072] The specific implementation of step S04 is the same as described above, and will not be repeated in detail here.

[0073] The specific implementation of step S05 is to construct a two-layer game optimization model, where the objective function of the upper-layer game model is:

[0074] ;

[0075] In the formula, The value of the upper-level objective function is dimensionless. This is a dimensionless weighting coefficient for temperature uniformity. This is the average temperature weighting factor, expressed in squares of Kelvin. This is the heating power weighting factor, expressed as the square root of the power per watt. The time period weighting coefficient is dimensionless. This is the heat transfer weighting factor, in units of per joule. This is a dimensionless index for evaluating temperature uniformity. The average temperature is expressed in Kelvin. This refers to the heating power of the upper model, measured in watts. The game time is measured in seconds. This is a time period parameter, in seconds. The heat transfer unit is joules. The objective function of the lower-level game model is:

[0076] ;

[0077] In the formula, The value of the lower-level objective function is dimensionless. This is the water migration weighting coefficient, dimensionless. This is a dimensionless weighting coefficient for heating power. The pressure weighting coefficient is dimensionless. The time weighting coefficient is dimensionless. The resistance weighting coefficient is dimensionless. The value represents the rate of water migration, expressed in kilograms per second. This refers to the heating power of the lower-level model, measured in watts. Pressure difference, unit: megapascals (MPA). For reference pressure, the unit is megapascals and the value is 0.101325. Angular frequency, measured in radians per second. The unit of measurement is Pascal-second per meter (Pa·s). The formula for calculating the temperature uniformity evaluation index is:

[0078] ;

[0079] in For the first Temperature at each measuring point, in Kelvin. The measurement point number is set to a value from 1 to... , The average temperature is expressed in Kelvin. The total number of measuring points is set to 16. The differential heating power calculation function is as follows:

[0080] ;

[0081] in For the first Zone heating power, in watts. For the region number, This refers to the basic heating power, measured in watts. The power regulation coefficient is dimensionless and ranges from 0.1 to 0.3. The target temperature is expressed in Kelvin. For the first The actual temperature of the region, in Kelvin.

[0082] The specific implementation of step S06 involves using Dijkstra's algorithm to optimize the water migration path, discretizing the sweet potato grains into a three-dimensional grid, and establishing an adjacency matrix.

[0083] ;

[0084] In the formula, This is the weight matrix. For nodes To the node The water migration resistance weight, expressed in Pascals per second per meter. The starting node number. The target node number. The total number of grid nodes is calculated using Darcy's law:

[0085] ;

[0086] in Dynamic viscosity, measured in Pascals per second. For nodes To the node The distance, in meters. Permeability, in square meters. The cross-sectional area is expressed in square meters. The formula for calculating the rate of change of moisture gradient is:

[0087] ;

[0088] in This represents the rate of change of the moisture gradient, expressed as a percentage per minute. This refers to the moisture content of the outer layer, expressed in kilograms. This refers to the moisture content of the inner layer, expressed in kilograms. Total moisture content, in kilograms. The gradient calculation time is in seconds. The distance update formula for Dijkstra's algorithm is:

[0089] ;

[0090] in To reach the node The shortest distance, To reach the node The shortest distance, As an intermediate node, For the target node, For the edge The weight value, in units of same.

[0091] The specific implementation method of step S07 is the same as described above, and will not be repeated in detail here.

[0092] It should be explained that the method for obtaining the parameters in the temperature control parameter calculation equation is as follows: Obtained by infrared thermometer measurement The thermal conductivity of sweet potato samples was obtained by measuring the steady-state method. Density was obtained by measuring the water displacement method. Specific heat capacity was obtained by measuring the specific heat capacity using differential scanning calorimetry. Obtained through Fourier decomposition of the initial temperature distribution. Wavenumbers of each order were obtained by fitting experimental data. , , Determined by boundary conditions.

[0093] The method for obtaining the parameters of the vacuum degree optimization calculation equation is as follows: Read directly using a vacuum gauge. Configured according to equipment performance and product requirements. The optimal pulse period was determined through preliminary experiments. The pressure decay time constant was obtained by fitting experimental data. The mass transfer experiment was conducted to determine the results. Check using a vapor pressure gauge.

[0094] The parameters of the two-level game model are obtained as follows: weight coefficients , , , , The weight coefficients are obtained through optimization using a genetic algorithm. , , , , Obtained through particle swarm optimization algorithm The angular frequency is set according to the process requirements. Take standard atmospheric pressure as 0.101325 MPa. The value ranges from 0.1 to 0.3. Determined by the process cycle.

[0095] The parameters of Dijkstra's algorithm are obtained as follows: Obtained by measuring with a viscometer Through permeability experiments, the coordinates of each node were determined by dividing the data into a three-dimensional mesh, and the weight matrix was used. Dimensions The value is determined based on the mesh precision, and is typically set to 1000. Obtained by calculating the Euclidean distance from the node coordinates.

[0096] The temperature control parameter calculation equation is based on Fourier heat conduction theory, describing the complex three-dimensional temperature field distribution through the superposition of multiple modes. The core of this equation lies in the exponential decay term. Simulate the temperature change over time, cosine function term Describe the characteristics of spatial temperature distribution.

[0097] ;

[0098] The partial differential equation passes through the thermal diffusivity term. By linking the thermal properties of materials with temperature changes, this equation, compared to traditional linear heating control, can predict and control the temperature gradient inside and outside sweet potato grains, avoiding surface overheating and insufficient internal heating, thus improving temperature distribution uniformity by 35% and reducing nutrient loss by 20%. The vacuum degree optimization calculation equation combines a sine function and an exponential decay function to simulate the periodic changes and decay characteristics of pressure during pulsed vacuum processes.

[0099] ;

[0100] The sine term in the equation Generates periodic pressure pulsations, with an exponentially decaying term. By simulating the pressure decay process and precisely controlling the change in vacuum level, the uniform migration of moisture inside sweet potato grains is promoted. Compared with a constant vacuum level, the pulsed vacuum technology controlled by this equation improves the moisture migration efficiency by 40% and reduces the oil content of the product by 15%. The objective function of the two-level game model adopts a nonlinear polynomial combination, and the upper-level model uses logarithmic terms... Quadratic terms Square root term and trigonometric function terms The coupling achieves a balance between temperature uniformity and energy consumption optimization.

[0101] ;

[0102] The lower-level model uses power function terms. , Exponential function term and trigonometric function terms The combination of these factors enables coordinated control of water migration rate and resistance.

[0103] ;

[0104] Compared to single-objective optimization, this two-level game theory model improves overall process efficiency by 30% and product quality indicators by 25%. The Dijkstra algorithm transforms the water migration process into a graph-based shortest path problem by constructing a weight matrix of water migration resistance.

[0105] ;

[0106] The weighting formula is based on Darcy's law, which determines the migration resistance by the ratio of dynamic viscosity, distance, permeability and cross-sectional area, and the algorithm dynamically finds the optimal migration path.

[0107] ;

[0108] The distance update formula optimizes the path by comparing the current shortest distance with the path length through intermediate nodes. Compared with traditional empirical control, this algorithm optimizes the water migration path, reduces the migration time by 20%, and makes the temperature and moisture distribution more uniform.

[0109] To better understand and implement this invention, the following is a specific application scenario of this invention, Example 2:

[0110] Traditional deep-frying techniques, under high temperatures, can cause the outer layer of sweet potato cubes to burn while the inner layer remains uncooked, thus destroying vitamins. Nutrients such as carotene are largely lost, making it difficult for product quality to meet market demands. The technical team decided to solve this problem using the low-temperature vacuum frying nutrient density enhancement technology of this invention.

[0111] The technical team first pre-processed the sweet potato raw materials. Fresh red-fleshed sweet potatoes were selected, each weighing between 200 and 300g, with a sugar content of 8.5%. After washing and peeling, the sweet potatoes were cut into 8mm cubes using precision cutting equipment, ensuring a size deviation of no more than ±0.5mm. A 0.5% citric acid solution was prepared, and the cut sweet potato cubes were completely submerged in it for a strictly controlled soaking time of 15 minutes. During soaking, the cubes were gently stirred every 3 minutes to ensure full contact between the citric acid solution and the surface of the sweet potato cubes. After soaking, the cubes were rinsed twice with clean water and drained, yielding approximately 5000g of pre-processed sweet potato cubes.

[0112] Next, a mathematical model of the temperature gradient of the sweet potato kernels was established. The technical team installed 12 infrared temperature sensors inside the vacuum frying equipment, distributed at different locations within the heating chamber, to monitor the temperature distribution in real time. Based on the physical properties of the sweet potato kernels, the thermal conductivity coefficient was determined to be 0.52. Its specific heat capacity is 3.48. Its density is 1150. A segmented temperature control strategy was determined by calculating the temperature control parameters. The first stage involved raising the temperature from 25℃ to 45℃ at a rate of 2℃ / min for 10 minutes. The second stage involved raising the temperature from 45℃ to 65℃ at a rate of 1.5℃ / min for 13.3 minutes. The third stage involved raising the temperature from 65℃ to 80℃ at a rate of 1℃ / min for 15 minutes.

[0113] The implementation of pulsed vacuum technology is a crucial step. Based on the vacuum degree optimization calculation equation and considering the initial moisture content of sweet potato grains (72%) and porosity (38%), the pulsed vacuum control parameters were calculated. The vacuum degree variation cycle was set to 120 seconds, periodically changing between -0.08 MPa and -0.05 MPa, with a vacuum degree variation amplitude of 0.03 MPa, and each pulse lasting 60 seconds. Figure 2As shown, the pulsed vacuum system achieves precise adjustment of the vacuum level through the coordinated control of the solenoid valve group.

[0114] The ultrasonic-assisted mass transfer system was started synchronously. The ultrasonic generator frequency was set to 40kHz, and the power density was precisely controlled at 0.3. The ultrasonic transducers are evenly distributed at the bottom of the heating chamber, with a total of 16 transducers arranged in a 4×4 matrix. The ultrasonic pulses are synchronized with the vacuum pulses. When the vacuum level decreases, the ultrasonic intensity increases, and when the vacuum level increases, the ultrasonic intensity decreases accordingly, forming a synergistic mass transfer effect.

[0115] Temperature control employs a two-layer game theory optimization model. The upper-layer game theory model aims to maximize temperature uniformity, calculating a temperature uniformity evaluation index based on real-time monitored temperature data. At the 15-minute mark of the frying process, the temperature uniformity evaluation index is 0.15, falling within the range of [0, 0.2). The system automatically adjusts to a uniform heating state, with power evenly distributed across all heating elements. At the 28-minute mark, due to faster heat dissipation in the edge areas, the temperature uniformity evaluation index rises to 0.68, falling into the range of [0.6, 1]. The system then calculates the power distribution parameters for each area using a differentiated heating power calculation function, reducing the power in the central area to 65% of the rated power and increasing the power in the edge areas to 120% of the rated power.

[0116] The Dijkstra algorithm was used to optimize the path for monitoring the moisture migration process. Every 5 minutes, three sweet potato grains were randomly selected from the heating chamber, and the moisture content of the surface, middle, and inner layers was measured using a Karl Fischer moisture analyzer. Table 1 shows the changes in moisture gradient at different time points.

[0117] Table 1. Data on moisture gradient monitoring of sweet potato grains

[0118]

[0119] At the 20-minute mark, the moisture gradient change rate was 22.1% / min, falling within the range of (15, 25)% / min. The system immediately reduced the vacuum to -0.06 MPa and extended the ultrasonic treatment time to 1.5 times the original set value, i.e., the duration of each ultrasonic cycle was extended from 60 seconds to 90 seconds. At the 25-minute mark, the moisture gradient change rate reached 32.1% / min, exceeding the 25% / min threshold. The system immediately suspended heating and increased the pulse vacuum intensity, raising the vacuum change amplitude from 0.03 MPa to 0.05 MPa.

[0120] In solving the two-layer game model, the upper-layer model optimizes the uniformity of temperature distribution to obtain the optimal temperature control strategy parameters. The lower-layer model aims to maximize water migration efficiency to obtain the optimal water migration control strategy. The two layers form a game equilibrium through power coupling, achieving overall performance optimization. Figure 3 As shown, the convergence of the objective function during the two-layer game optimization process indicates that the model can quickly reach a stable state.

[0121] The entire frying process lasted 45 minutes, with the surface temperature distribution of the sweet potato pieces continuously monitored using an infrared thermal imager. At the 42-minute mark, the average surface temperature of the sweet potato pieces stabilized at 82℃, indicating good temperature uniformity. From the 43rd to the 46th minute, the average surface temperature remained stable within the range of 80±3℃, meeting the termination criteria.

[0122] Immediately after frying, a gradient cooling program is initiated. The first stage reduces the temperature from 80℃ to 60℃ at a rate of 3℃ / min. The second stage reduces the temperature from 60℃ to 40℃ at a rate of 2℃ / min. The third stage reduces the temperature from 40℃ to 25℃ at a rate of 1.5℃ / min. The entire cooling process lasts 35 minutes, yielding 4200g of final sweet potato granules with a moisture content reduced to 15%, a golden-yellow color, and a crispy texture.

[0123] Product quality testing results show that vitamins The retention rate reached 78%, and the carotene retention rate reached 82%, far exceeding the 45% and 52% of traditional high-temperature frying processes. The product's crispness value was 285N, and its elastic modulus was 1.25. The texture is crispy and moderately crunchy. (Color parameters) The value is 68.5. The value is 12.3. The value is 45.2, presenting an ideal golden yellow color.

[0124] To further verify the rationality of the process parameters, the technical team conducted a sensitivity analysis on the key control parameters. As shown in Table 2, different parameters have varying degrees of impact on product quality.

[0125] Table 2. Results of Sensitivity Analysis of Process Parameters

[0126]

[0127] Comparative analysis revealed that the heating rate had the most significant impact on nutrient retention, while ultrasonic power density had the greatest impact on texture quality, and temperature uniformity threshold primarily affected color quality. These data provide important insights for subsequent process optimization.

[0128] In actual production, the technical team also established a real-time monitoring system, which uses multi-sensor fusion technology to achieve comprehensive monitoring of temperature, pressure, moisture, and ultrasonic parameters, such as... Figure 4 As shown, the system can automatically identify abnormal situations and adjust process parameters in a timely manner to ensure the stability and consistency of product quality. The monitoring system's response time is less than 2 seconds, and the control accuracy is within ±2% of the set value.

[0129] After seven consecutive days of production verification, the technical solution demonstrated good stability and repeatability. The quality index fluctuation range of each batch of products was controlled within ±5%, production efficiency was increased by 23% compared with traditional processes, and equipment energy consumption was reduced by 18%. The product shelf life was extended from the original 15 days to 25 days, greatly enhancing its commercial value.

[0130] This invention represents a significant technological advancement over traditional frying methods. First, by establishing a mathematical model of the temperature gradient and employing a segmented heating control strategy, it solves the problem of burnt exterior and undercooked interior caused by traditional high-temperature frying, achieving uniform temperature transfer between the inside and outside. Second, the introduction of pulsed vacuum technology overcomes the limitations of traditional constant vacuum, promoting uniform moisture migration through periodic pressure changes and preventing localized excessive dehydration. Third, the synergistic mechanism of the ultrasonic-assisted mass transfer system and pulsed vacuum accelerates the mass transfer process at the molecular level, improving mass transfer efficiency while maintaining the stability of nutritional components. Fourth, the application of a two-layer game optimization model enables intelligent coordination of temperature control and moisture migration. The multi-objective optimization framework established through Stackelberg game theory can find optimal solutions under complex process conditions, which is impossible with traditional single-objective control methods. Fifth, moisture migration path optimization based on the Dijkstra algorithm transforms the complex mass transfer process into a mathematical shortest path problem, achieving precise control and prediction of the moisture migration path. Finally, the integration of real-time monitoring and adaptive control systems enables the entire process to be dynamically adjusted according to actual conditions, overcoming the rigidity problem of traditional fixed parameter control and significantly improving the stability of product quality and the adaptability of the process.

[0131] It should be noted that the variables involved in this invention are explained in detail in Table 3.

[0132] Table 3. Variable Explanation Table

[0133]

[0134] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for improving the nutrient density of sweet potato granules by low-temperature vacuum frying, characterized by, The fresh sweet potato is washed and peeled, and then cut into cubic particles, and then soaked in a citric acid solution for treatment to remove surface starch and perform color protection pretreatment, to obtain pretreated sweet potato particles; a temperature gradient mathematical model of the sweet potato particles is established, an infrared temperature monitoring system is installed in a vacuum frying equipment, temperature control parameter calculation equations are used to calculate segmented temperature control strategy parameters, and the parameters are used for temperature control; a pulse vacuum technique is used for dynamic regulation of vacuum degree, vacuum degree optimization calculation equations are used to calculate pulse vacuum control parameters, the vacuum degree is adjusted according to the pulse vacuum control parameters, and internal water migration of the sweet potato particles is promoted to be uniform; An ultrasonic assisted mass transfer system is started, and is used in synchronization with the pulse vacuum to improve internal water mass transfer efficiency; an upper game model with maximization of temperature uniformity as an objective and a lower game model with maximization of water migration rate as an objective are established, an optimal heating power distribution is solved through a double-layer game optimization model, a real-time monitoring of a surface temperature distribution uniformity evaluation index of the sweet potato particles is performed, a heating power distribution mode is adjusted or a differentiated heating mode is switched to; a water migration process is modeled as a shortest path problem, a Dijkstra algorithm is used to optimize a water migration path, sweet potato particle samples are grabbed, different levels of water content of the samples are detected through a Karl Fischer moisture meter, and a water gradient change rate is calculated, the vacuum degree and ultrasonic action time are adjusted or heating is paused and pulse vacuum intensity is enhanced; when the frying time reaches a preset total time and the average surface temperature of the sweet potato particles is stable through infrared thermal imaging detection, the frying process is ended, a gradient cooling method is used for cooling treatment to room temperature, and a sweet potato particle product with improved nutritional density is obtained, and the temperature distribution uniformity control and water migration efficiency optimization are realized through the temperature gradient mathematical model and the double-layer game optimization model.

2. The method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying according to claim 1, characterized in that, When the frying time reaches the preset total time and the average surface temperature of the sweet potato particles is stable within the range of 80±3℃ for 3 minutes through infrared thermal imaging detection, the frying process is ended.

3. The method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying according to claim 2, characterized in that, The step of soaking in the citric acid solution is specifically soaking in 0.5% citric acid solution for 15 minutes.

4. The method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying according to claim 3, characterized in that, The temperature control parameter calculation equation is used to calculate temperature set values and heating rates in each stage of the segmented heating process, and the input includes sweet potato particle geometric size, initial temperature distribution, target temperature and heat conduction coefficient, and the output is temperature control parameters in each time period.

5. The method of claim 4, wherein the sweet potato pieces are subjected to the low-temperature vacuum frying process for 5 to 20 minutes. The vacuum degree optimization calculation equation is used to determine a vacuum degree change period and amplitude in the pulse vacuum process, and the input includes sweet potato particle water content, porosity, temperature and pressure gradient, and the output is pulse vacuum control parameters.

6. The method of claim 5, wherein the sweet potato pieces are subjected to the low-temperature vacuum frying process for 5 to 20 minutes. The ultrasonic assisted mass transfer system has an ultrasonic frequency of 40 kHz and a power density of 0.3 W per square centimeter.

7. The method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying according to claim 6, characterized in that, When the temperature uniformity evaluation index ∈ [0, 0.2), the heating power distribution mode is adjusted to a uniform heating state, and when the temperature uniformity evaluation index ∈ [0.6, 1], differentiated heating mode parameters are calculated through a differentiated heating power calculation function and the differentiated heating mode is switched to.

8. The method for improving the nutritional density of sweet potato granules through low-temperature vacuum frying according to claim 7, characterized in that, Every 5 minutes, 3 sweet potato samples were taken, when the moisture gradient change rate ∈ was (15, 25] / %min, the vacuum degree was reduced to-0.06MPa and the ultrasonic wave action time was extended to 1.5 times of the original setting, when the moisture gradient change rate >25% / min, the heating was suspended and the pulse vacuum intensity was increased.

9. The method of claim 8, wherein the sweet potato pieces are subjected to the low-temperature vacuum frying process for 5 to 10 minutes. The differential heating power calculation function is used to calculate the heating power distribution curve of different regions, and the input includes the temperature deviation, heat capacity, heat transfer coefficient and target temperature of each region, and the output is the heating power curve of each region.

10. The method of claim 9, wherein the sweet potato pieces are subjected to the low-temperature vacuum frying process for 5 to 20 minutes. The temperature uniformity evaluation index refers to the ratio of the standard deviation of the temperature difference between each point on the surface of the sweet potato and the average temperature to the average temperature, which is used to quantify the uniformity of the temperature distribution.

Images