Dehydrated garlic drying process parameter collaborative optimization method based on swarm intelligence
By acquiring and standardizing the drying data sequence of garlic slices, evaluating the dissipation coefficient and co-health, and updating the search speed using a state feedback mechanism, the problem of the strong dependence of the drying process parameter optimization algorithm on prior parameters was solved. This enabled the efficient and low-energy drying process of garlic slices, ensuring the synergistic optimization of allicin bioactivity and energy efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-10
- Publication Date
- 2026-04-14
AI Technical Summary
Existing optimization algorithms for dehydrated garlic drying process parameters are highly dependent on prior parameters, have poor search stability under time-varying conditions, and are difficult to balance quality conservation and energy efficiency indicators. Furthermore, they lack a mathematical characterization of the thermodynamic essence of drying, resulting in a logical disconnect between the optimization objective and the physical process.
By acquiring the raw data sequence of garlic slices during drying, standard moving average filtering and linear normalization are used to evaluate the dissipation coefficient and collaborative health. Combined with the state feedback mechanism to update the search speed, the heating energy flow and moisture migration rate are matched, the drying temperature and residence time of garlic slices are adjusted, and parameter mismatch and local overheating caused by increased moisture migration resistance are reduced.
This improved the protection of allicin bioactivity during the drying process, reduced energy waste, enhanced energy utilization, and ensured the synergistic optimization of garlic slice quality and energy efficiency.
Smart Images
Figure CN121857604A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent optimization and industrial control technology. More specifically, this invention relates to a collaborative optimization method for dehydrated garlic drying process parameters based on swarm intelligence. Background Technology
[0002] Dehydrated garlic, as an important condiment and deep-processed agricultural product, has widespread applications in food processing and international trade. Its drying process is a core element in controlling the retention rate of allicin bioactivity and the unit energy efficiency ratio. In actual industrial drying production processes, garlic slices undergo complex moisture, heat, and mass exchange processes. The migration rate of moisture from the inside of the garlic slices to the surface directly determines the drying efficiency, and the thermophysical properties of garlic slices exhibit a significant non-linear shift as the real-time moisture content decreases. The internal structural characteristics of garlic slices cause the resistance to moisture overflow to increase exponentially in the later stages of drying. If the heat source supply is mismatched with the moisture migration rate, the excess heat energy flow will directly act on the material matrix, leading to thermal damage to the allicin bioactivity and even causing charring and browning of the garlic slices. Currently, in production practice, the setting of key process parameters such as drying temperature, inlet air velocity, and conveyor belt speed generally relies on empirical process procedures or fixed proportional-integral-derivative (PID) control logic. This static control method is difficult to cope with dynamic disturbances such as fluctuations in environmental humidity, differences in the bulk density of garlic slices, and uneven average thickness of garlic slices.
[0003] To improve the intelligence level of the drying process, the industry has attempted to introduce swarm intelligence algorithms to globally optimize the set of process parameters, aiming to establish a balance between energy loss and quality conservation. However, such computationally intelligent optimization models generally exhibit strong parameter sensitivity when dealing with drying conditions with complex time-varying constraints. Existing algorithmic frameworks heavily rely on pre-set learning factors, inertia weights, or evolution rates, whose values directly determine the population's exploration and development capabilities in the solution space. Once the initial moisture content of the garlic slices or the chamber load shifts, the fixed algorithm weights cannot guide the search particles to migrate efficiently in the dynamically changing fitness landscape. When the optimization process is in a stage of drastic changes in drying dynamics, this dependence on artificial parameters easily leads the computational model to get stuck in local optima or to produce violent search oscillations near the ideal target point, failing to achieve real-time and accurate adaptation of the process parameter set to the physiological state of the material. Furthermore, due to the lack of a mathematical characterization of the thermodynamic nature of drying, the algorithm often struggles to quantitatively assess the effective contribution of the heating energy flow to moisture removal during the iteration process. This results in a logical disconnect between the optimization objective and the physical drying process, making it difficult to meet the deep synergistic requirements of modern garlic processing for high quality and low energy consumption. Summary of the Invention
[0004] To address the technical problems of optimization algorithms for dehydrated garlic drying parameters, such as strong dependence on prior parameters, poor search stability under time-varying conditions, and difficulty in simultaneously considering quality conservation and energy efficiency indicators, this invention provides a collaborative optimization method for dehydrated garlic drying process parameters based on swarm intelligence. The method includes: acquiring and standardizing the raw drying data sequence of garlic slices to obtain a processed raw drying data sequence; obtaining the drying temperature change rate based on the processed raw drying data sequence; obtaining the dissipation coefficient based on the drying temperature, real-time moisture content gradient, heating energy flow, and moisture migration resistance characteristic values in the processed raw drying data sequence; acquiring a set of process parameters; obtaining a collaborative health level based on the dissipation coefficient, the process parameter set, the drying temperature change rate, and a preset critical temperature for allicin activity protection; obtaining an update search speed based on the collaborative health level and the historical highest health level; obtaining the optimal set of process parameters based on the update search speed; and obtaining heating energy flow adjustment signals and drive motor frequency adjustment commands based on the optimal set of process parameters to achieve collaborative adjustment of drying temperature and garlic slice residence time.
[0005] This invention obtains the original data sequence of garlic slices during drying and acquires the dissipation coefficient and collaborative health. It then uses a state feedback-based update mechanism to obtain the optimal set of process parameters and sends them to the dryer controller. This achieves the adaptation of heating energy flow and moisture migration rate, reducing parameter mismatch and local overheating problems caused by increased moisture migration resistance in the later stages of drying.
[0006] Preferably, the standardization process includes: denoising the acquired electrical signal using a standard moving average filtering algorithm; and mapping physical quantities of different dimensions to a standard numerical space using a linear normalization method to obtain the processed original drying data sequence.
[0007] This invention utilizes a standard moving average filtering algorithm to denoise the acquired electrical signals and uses a linear normalization method to map physical quantities of different dimensions to a standard numerical space. By obtaining the processed raw drying data sequence, the adverse effects of random interference from the sampled electrical signals on the accuracy of subsequent process parameter evaluation are reduced.
[0008] Preferably, the dissipation coefficient satisfies the expression: In the formula, for Time dissipation coefficient; For time indexing; for Maintain constant drying temperature; for Real-time moisture content; For time variables Real-time moisture content; for Real-time moisture content gradient; for Constant flow of heating energy; This represents the characteristic value of water migration resistance. It is an exponential function with the natural constant as its base; For integration operations; For infinitesimal time; It is the absolute value symbol.
[0009] This invention obtains the dissipation coefficient by measuring drying temperature, real-time moisture content gradient, heating energy flow, and moisture migration resistance characteristics, and evaluates the effective contribution of unit heating energy flow to moisture removal, thereby reducing energy waste caused by the increased binding force on moisture in the later stages of drying.
[0010] Preferably, the moisture migration resistance characteristic value is obtained by: using an ultrasonic sensor to obtain the average thickness of garlic slices; and obtaining the moisture migration resistance characteristic value based on the average thickness of garlic slices and a preset steady-state diffusion coefficient.
[0011] This invention obtains the average thickness of garlic slices using an ultrasonic sensor and combines it with a preset steady-state diffusion coefficient to obtain the characteristic value of moisture migration resistance. It describes the modulation effect of the physical morphology of garlic slices on the difficulty of moisture removal and enhances the adaptability of the dissipation coefficient to the drying conditions of garlic materials of different thicknesses.
[0012] Preferably, the collaborative health score satisfies the expression: In the formula, For the first The collaborative health of each search particle; For population particle indexing; This represents the total number of samples taken over time. for Conveyor belt speed at all times; for Time dissipation coefficient; for Maintain constant drying temperature; for The first derivative of drying temperature with respect to time; The critical temperature for protecting allicin activity; This refers to the rate of quality degradation. It is an exponential function with the natural constant as its base; For summation operations; It is the absolute value symbol.
[0013] This invention obtains the synergistic health status based on parameters such as conveyor belt speed, dissipation coefficient, and critical temperature for allicin activity protection. By establishing a search guide for allicin bioactivity protection within the parameter space, it reduces the damage to garlic slice quality when improving the utilization level of heating energy flow.
[0014] Preferably, the quality decay rate is obtained by fitting a constant of color change with drying temperature based on the color difference curve of a standard garlic sample and the chemical reaction kinetic equation, and using the coefficient reflecting the thermal stability characteristics obtained from the fitting as the quality decay rate.
[0015] This invention obtains the quality decay rate reflecting thermal stability characteristics by using color difference change curves and chemical reaction kinetic equations, evaluates the browning trend of garlic slices under high temperature conditions, and improves the accuracy of co-health perception of the color change of garlic slices with temperature.
[0016] Preferably, the update search speed satisfies the expression: In the formula, For the first Individual particles Search speed at any given moment; For population particle indexing; For time indexing; For the first Individual particles Search speed at any given moment; For the first The collaborative health of each search particle; This represents the highest health level in history. To prevent constants with a denominator of zero; For the first Individual particles The set of process parameters at any given time; for Time dissipation coefficient; This refers to partial derivative operations.
[0017] This invention utilizes the ratio of collaborative health to the highest historical health to obtain the update search speed, and through a self-adjusting mechanism that dynamically derives the search step size, the search behavior is automatically adjusted according to the complexity of the environment, reducing the possibility of drastic search oscillations near the ideal target point during the optimization process.
[0018] Preferably, obtaining the optimal process parameter set includes: establishing a population composed of multiple search particles, and updating the position and search speed of each search particle in each iteration; recording the maximum value of the collaborative health of each search particle in the historical iteration process as the historical highest health; and determining the position of the search particle corresponding to the historical highest health as the optimal process parameter set in response to the fact that the increment of the collaborative health is less than a convergence threshold for a consecutive preset number of times.
[0019] This invention establishes a search particle population and records the highest historical health level. When the incremental health level of the collaborative health level is continuously less than the convergence threshold, the optimal set of process parameters is locked, realizing the adaptive evolution of the optimization process and reducing the algorithm's dependence on prior parameters such as manually preset learning factors or inertial weights.
[0020] Preferably, the coordinated adjustment of drying temperature and garlic slice residence time includes: sending the optimal set of process parameters to the dryer controller; comparing the deviation between real-time moisture content and expected target in real time during each sampling cycle through the dryer controller; outputting a heating energy flow adjustment signal to the dryer heating system based on the deviation between real-time moisture content and expected target, and outputting a transmission motor frequency adjustment command to the conveyor belt drive motor.
[0021] This invention achieves closed-loop control of drying temperature and garlic slice residence time by comparing the deviation between real-time moisture content and expected target in real time with the dryer controller and outputting heating energy flow adjustment signals and drive motor frequency adjustment commands, thereby improving the control accuracy of the final moisture content of garlic slices.
[0022] Preferably, the set of process parameters includes drying temperature, air intake velocity, and conveyor belt speed; the raw drying data sequence includes drying temperature sequence, air intake velocity sequence, drive motor frequency sequence, and real-time moisture content sequence.
[0023] The beneficial effects of this invention are as follows: This invention uses dissipation coefficient and synergistic health to correlate and evaluate the real-time moisture content change of garlic slices with the heating energy flow, thereby reducing the charring and browning phenomenon caused by the mismatch between heat source supply and moisture migration rate during the drying process.
[0024] This invention utilizes a state feedback-based update search speed mechanism to perform adaptive optimization, enabling the computational model to cope with the dynamic drift of initial moisture content or chamber load in garlic slices, thus reducing the risk of swarm intelligence algorithms getting trapped in local optima under time-varying conditions.
[0025] This invention improves energy utilization and reduces unit energy consumption during the dehydrated garlic processing process by synergistically adjusting the drying temperature and the residence time of garlic slices, while ensuring the protection of allicin bioactivity. Attached Figure Description
[0026] Figure 1 The flowchart illustrates the collaborative optimization method for dehydrated garlic drying process parameters based on swarm intelligence in this invention. Figure 2 A schematic diagram illustrating the time-varying relationship between dissipation coefficient and real-time water content; Figure 3 This diagram illustrates the convergence of the iterative optimization of collaborative health. Figure 4 The diagram illustrates the coordinated adjustment of drying temperature and conveyor belt speed. Detailed Implementation
[0027] 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, not all, of the embodiments of the present invention. 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.
[0028] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
[0029] This invention discloses a collaborative optimization method for dehydrated garlic drying process parameters based on swarm intelligence, referring to... Figure 1 This includes steps S1 to S5: S1. Obtain the raw data sequence of garlic slices after drying and perform standardization processing to obtain the processed raw data sequence of drying; obtain the drying temperature change rate based on the processed raw data sequence of drying.
[0030] It should be noted that the drying process of garlic slices involves the physical process of moisture moving from the inside of the garlic slices to the surface, and the loading of garlic slices directly affects the distribution of drying temperature. The purpose of this invention to obtain the raw drying data is to provide the necessary data foundation for subsequent evaluation of the moisture removal and heat energy flow exchange patterns inside the garlic slices, thereby eliminating the perception blind spots caused by fluctuations in operating conditions.
[0031] Specifically, this invention acquires raw drying data, which includes drying temperature, air intake speed, drive motor frequency, and real-time moisture content. This invention uses a standard moving average filtering algorithm to denoise the acquired electrical signals to eliminate random glitches during the sampling process, and uses a linear normalization method to map physical quantities of different dimensions to a standard numerical space to ensure the balance of the weights of each parameter in subsequent calculations.
[0032] S2. Obtain the dissipation coefficient based on the drying temperature, real-time moisture content gradient, heating energy flow, and moisture migration resistance characteristic value in the processed raw drying data sequence.
[0033] It should be noted that the energy absorption and utilization efficiency of garlic slices varies significantly at different drying stages, especially in the later stages of drying. As the real-time moisture content inside the garlic slices decreases, the binding force on the moisture increases significantly, leading to an increase in the characteristic value of moisture migration resistance. At this point, if the original heating intensity is maintained, the excess heating energy flow will no longer be used to drive moisture removal, but will instead cause the garlic slices to heat up, which can easily lead to local overheating and charring. Therefore, this invention establishes a specific calculation logic to obtain an index reflecting the degree of matching between heating energy flow and moisture removal in real time, aiming to evaluate the effective contribution of each unit of heating energy flow to moisture removal, thereby providing a physical basis for avoiding energy waste.
[0034] Specifically, the present invention obtains the dissipation coefficient for evaluating the utilization level of heating energy flow based on the current state of garlic slices.
[0035] The dissipation coefficient satisfies the expression:
[0036] In the formula, express Time dissipation coefficient; Indicates a time index; express Maintain constant drying temperature; express Real-time moisture content; Indicates the time variable Real-time moisture content; express Real-time moisture content gradient; express Constant flow of heating energy; This represents the characteristic value of resistance to water migration; Represents an exponential function with the natural constant as its base; Indicates integration operation; Represents the infinitesimal time; This represents absolute value operations.
[0037] In the formula, when the real-time moisture content is... As the drying process continues to decrease, the exponential integral part of the denominator, determined by the real-time moisture content, gradually stabilizes due to numerical accumulation. This term physically evaluates the characteristic value of the resistance to moisture migration in garlic slices as the real-time moisture content decreases. Simultaneously, if the real-time moisture content gradient is adjusted at this point... A drastic change indicates an increase in water excretion per unit time, causing an increase in the numerator value and ultimately driving the dissipation coefficient. It produces a non-linear, increasing change.
[0038] Dissipation coefficient The higher the value, the higher the utilization level of the heating energy flow during the drying process. This means that more of the input heating energy flow is used to drive the removal of moisture from the inside of the garlic slices. The utilization level of the heating energy flow is positively correlated with the improvement of drying energy efficiency.
[0039] It should be further added that the characteristic value of water migration resistance in this invention The moderating effect describes the effect of the average thickness of garlic slices on the difficulty of water removal; the characteristic value of water migration resistance. The method for obtaining the average thickness of garlic slices involves using an ultrasonic sensor to acquire the average thickness and then calculating the steady-state diffusion coefficient at that average thickness using relevant diffusion laws. The extracted value, reflecting the strength of the water migration barrier, is used as the characteristic value of water migration resistance. .
[0040] For example, Figure 2 This is a schematic diagram illustrating the time-varying relationship between the dissipation coefficient and real-time moisture content. The diagram shows that as the drying process progresses, the real-time moisture content of the garlic slices exhibits a non-linear decreasing trend, while the dissipation coefficient remains at a relatively high level despite fluctuations. This indicates that by real-time evaluation of the matching relationship between moisture migration resistance and heating energy flow, the system can dynamically sense the energy utilization status, ensuring an effective energy dissipation ratio is maintained even in the later stages of drying when moisture removal is difficult, thus avoiding excessive input of ineffective heat energy.
[0041] S3. Obtain the set of process parameters, and obtain the synergistic health status based on the dissipation coefficient, the set of process parameters, the rate of change of drying temperature, and the preset critical temperature for allicin activity protection.
[0042] It should be noted that optimizing the garlic slice drying process is a process involving the balancing of multiple indicators. On the one hand, it is necessary to improve the utilization level of heating energy flow, and on the other hand, it is essential to strictly limit the drying temperature to protect the bioactivity of allicin and prevent high-temperature browning of garlic slices. This invention introduces the concept of quality decay rate and organically integrates the indicators reflecting the bioactivity of allicin with the dissipation coefficient of the utilization level of heating energy flow to obtain a synergistic health score, which serves as the core indicator for evaluating the merits of the current process parameter set. The aim is to establish a search guide within the parameter space that can simultaneously guide the improvement of the utilization level of heating energy flow and the protection of allicin bioactivity, ensuring that the optimization results do not sacrifice the quality value of garlic slices.
[0043] Specifically, this invention utilizes the dissipation coefficient combined with a set of process parameters to obtain collaborative health.
[0044] Collaborative health satisfies the expression:
[0045] In the formula, Indicates the first The collaborative health of each search particle; Indicates the population particle index; Indicates the total number of samples taken over time; express Conveyor belt speed at all times; express Time dissipation coefficient; express Maintain constant drying temperature; express The first derivative of drying temperature with respect to time; This indicates the critical temperature for protecting the activity of allicin. Indicates the rate of quality degradation; Represents an exponential function with the natural constant as its base; This represents the summation operation; This represents absolute value operations.
[0046] In the formula, when the drying temperature Critical temperature for protecting allicin activity When the deviation decreases, the exponential term approaches its maximum value due to the characteristics of the exponential function, reflecting the effective protection of the biological activity of allicin in garlic slices; if the conveyor belt speed is adjusted at this time... Make it consistent with the dissipation coefficient The increase in the product indicates an improvement in the utilization level of heating energy flow while ensuring quality. Furthermore, due to the constraint of the absolute value of the rate of change of drying temperature in the denominator, the instability caused by drastic fluctuations in drying temperature is suppressed, ultimately leading to improved synergistic health. The change has increased.
[0047] Collaborative Health The larger the value, the better the synergistic effect of the process parameter set in protecting the biological activity of allicin and improving energy efficiency, which means that the current process parameter set can achieve a higher level of heating energy flow utilization while ensuring the quality of garlic slices.
[0048] It should be further noted that the quality decay rate in this invention... The regulatory role is to assess the browning trend of garlic slices under high temperature conditions and the rate of quality degradation. The method for obtaining the color difference curve is to perform spectrophotometry on standard garlic samples, fit the constant of color change with drying temperature using chemical reaction kinetic equations, and use the coefficient reflecting thermal stability characteristics obtained from the fitting as the quality degradation rate. .
[0049] S4. Obtain the update search speed based on the collaborative health and the highest historical health, and obtain the optimal set of process parameters based on the update search speed.
[0050] It should be noted that traditional swarm intelligence algorithms often rely on fixed, manually preset weights. This static configuration is prone to search stalling when faced with drastic fluctuations in real-time moisture content or nonlinear disturbances caused by equipment aging. This invention introduces a self-adjusting mechanism based on state feedback during the optimization process. By calculating the difference between the current state and the historical highest health level in real time, the search step size is dynamically derived. The purpose is to achieve parameter-free operation of the algorithm, enabling the search behavior to automatically adjust the search intensity according to the complexity of the current environment, thereby achieving rapid locking of the optimal set of process parameters.
[0051] Specifically, this invention updates the search speed of each particle in real time by calculating the ratio of the current particle cooperative health to the highest historical health.
[0052] The search speed satisfies the expression:
[0053] In the formula, Indicates the first Individual particles Search speed at any given moment; Indicates the first Individual particles Search speed at any given moment; Indicates the first The collaborative health of each search particle; This indicates the highest health level in history; This represents a constant to prevent the denominator from being zero. Indicates the first Individual particles The set of process parameters at any given time; express Time dissipation coefficient; This represents partial derivative operations.
[0054] In the formula, when the current collaborative health level Nearly the highest health level in history When the proportion of the former term in the fraction increases, the search speed is improved. This allows for more subtle probing near the optimal solution by preserving the evolutionary inertia from the previous moment. If the current cooperative health is low, indicating that the particle is in an inefficient region, the latter's proportion increases. In this case, the gradient term's guidance and dissipation coefficient are combined. Energy utilization characteristics drive search speed It shifts towards the high-energy-efficiency region.
[0055] Search speed It can guide the optimization process to quickly lock onto the set of process parameters that meet the requirements of quality and energy efficiency, and the dynamic fluctuation of its values ensures that the algorithm can maintain extremely high search accuracy in different drying stages.
[0056] For example, Figure 3 This is a schematic diagram illustrating the convergence of the collaborative health iteration. As the number of iterations increases, the collaborative health rapidly rises and tends to stabilize, indicating that the search speed update mechanism based on state feedback effectively guides the particle swarm to converge toward a solution space that balances allicin activity protection and high energy efficiency, thus achieving rapid locking of the process parameter set.
[0057] S5. Based on the optimal process parameter set, obtain the heating energy flow adjustment signal and the drive motor frequency adjustment command to achieve coordinated adjustment of drying temperature and garlic slice residence time.
[0058] It should be noted that the optimization of the process parameter set is not limited to theoretical calculations. Its ultimate goal is to implement it in the closed-loop control of the physical drying process to achieve precise adjustment of the residence time and drying temperature of garlic slices. This invention transforms the optimized process parameter set into specific hardware adjustment instructions, which can dynamically adjust the input of heating energy flow and the frequency of the drive motor according to the real-time physiological state of the garlic slices. This ensures that the final garlic slices can retain the biological activity of allicin to the greatest extent while meeting the real-time moisture content standard.
[0059] Specifically, the present invention sends the optimal set of process parameters, such as drying temperature, air inlet velocity, and conveyor belt speed of each temperature zone, obtained through optimization, to the dryer controller; the system compares the deviation between the real-time moisture content and the expected target in real time during each sampling cycle.
[0060] Furthermore, this invention achieves coordinated regulation of drying temperature and garlic slice residence time by using an optimal set of process parameters, sending a heating energy flow adjustment signal to the heating system through the dryer controller, and simultaneously sending a transmission motor frequency adjustment command to the conveyor belt drive motor.
[0061] For example, Figure 4 This is a schematic diagram illustrating the coordinated adjustment of drying temperature and conveyor belt speed. The diagram shows the dynamic adjustment of drying temperature and conveyor belt speed over time under the drive of the optimal set of process parameters. As drying progresses, the system automatically lowers the drying temperature to adapt to changes in the heat sensitivity of the garlic slices, while simultaneously adjusting the conveyor belt speed to alter the material residence time, thus completing the moisture removal task while ensuring the bioactivity of allicin.
Claims
1. A collaborative optimization method for dehydrated garlic drying process parameters based on swarm intelligence, characterized in that, include: The raw data sequence of dried garlic slices was obtained and standardized to obtain the processed raw data sequence of dried garlic slices. The drying temperature change rate was obtained based on the processed raw drying data sequence. The dissipation coefficient is obtained based on the drying temperature, real-time moisture content gradient, heating energy flow, and moisture migration resistance characteristic value in the processed original drying data sequence. The process parameter set is obtained, and the synergistic health is obtained based on the dissipation coefficient, the process parameter set, the drying temperature change rate, and the preset critical temperature for allicin activity protection. The update search speed is obtained based on the collaborative health and the highest historical health, and the optimal set of process parameters is obtained based on the update search speed. Based on the optimal set of process parameters, the heating energy flow regulation signal and the drive motor frequency regulation command are obtained to achieve coordinated regulation of drying temperature and garlic slice residence time.
2. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The standardization process includes: The acquired electrical signals were denoised using a standard moving average filtering algorithm; and physical quantities of different dimensions were mapped to a standard numerical space using a linear normalization method to obtain the processed raw drying data sequence.
3. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The dissipation coefficient satisfies the expression: ； In the formula, for Time dissipation coefficient; For time indexing; for Maintain constant drying temperature; for Real-time moisture content; For time variables Real-time moisture content; for Real-time moisture content gradient; for Constant flow of heating energy; This represents the characteristic value of water migration resistance. It is an exponential function with the natural constant as its base; For integration operations; For infinitesimal time; It is the absolute value symbol.
4. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 2, characterized in that, The method for obtaining the characteristic value of water migration resistance is as follows: The average thickness of garlic slices is obtained using an ultrasonic sensor; the characteristic value of water migration resistance is obtained based on the average thickness of garlic slices and a preset steady-state diffusion coefficient.
5. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The collaborative health status satisfies the expression: ； In the formula, For the first The collaborative health of each search particle; For population particle indexing; This represents the total number of samples taken over time. for Conveyor belt speed at all times; for Time dissipation coefficient; for Maintain constant drying temperature; for The first derivative of drying temperature with respect to time; The critical temperature for protecting allicin activity; This refers to the rate of quality degradation. It is an exponential function with the natural constant as its base; For summation operations; It is the absolute value symbol.
6. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 5, characterized in that, The quality decay rate is obtained as follows: Based on the color difference curve of standard garlic samples and the chemical reaction kinetic equation, the constant of color change with drying temperature is fitted, and the coefficient reflecting the thermal stability characteristics obtained by fitting is used as the quality decay rate.
7. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The update search speed satisfies the expression: ； In the formula, For the first Individual particles Search speed at any given moment; For population particle indexing; For time indexing; For the first Individual particles Search speed at any given moment; For the first The collaborative health of each search particle; This represents the highest health level in history. To prevent constants with a denominator of zero; For the first Individual particles The set of process parameters at any given time; for Time dissipation coefficient; This refers to partial derivative operations.
8. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The process of obtaining the optimal set of process parameters includes: A population consisting of multiple search particles is established, and the position and search speed of each search particle are updated in each iteration. The maximum value of the cooperative health of each search particle in the historical iteration process is recorded as the historical highest health. In response to the fact that the increment of cooperative health is less than the convergence threshold for a consecutive preset number of times, the position of the search particle corresponding to the historical highest health is determined as the optimal set of process parameters.
9. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The method for coordinating the drying temperature and the dwell time of the garlic slices includes: The optimal set of process parameters is sent to the dryer controller; the dryer controller compares the deviation between the real-time moisture content and the expected target in real time during each sampling period; based on the deviation between the real-time moisture content and the expected target, a heating energy flow adjustment signal is output to the dryer heating system, and a transmission motor frequency adjustment command is output to the conveyor belt drive motor.
10. The method for collaborative optimization of dehydrated garlic drying process parameters based on swarm intelligence according to claim 1, characterized in that, The set of process parameters includes drying temperature, air intake velocity, and conveyor belt speed; the raw drying data sequence includes drying temperature sequence, air intake velocity sequence, drive motor frequency sequence, and real-time moisture content sequence.