Phase-control multi-source microwave heating system temperature field optimization method based on numerical model prediction

Abstract

The invention relates to a phase-control multi-source microwave heating system temperature field optimization method based on numerical model prediction, and belongs to the field of microwave heating modeling simulation. The method comprises the following steps: constructing a microwave heating model simulation platform; microwave sources on the same side are combined, and a numerical model is established based on a physical heating model; the numerical model parameters are identified and corrected through the system to improve the prediction precision; the corrected numerical model parameters are substituted into a numerical model to predict electric field distribution under different relative phase differences; and carrying out relative phase difference selection according to the predicted electric field distribution so as to achieve the effect of optimizing the temperature uniformity. According to the method, complex electric field and temperature field coupling can be avoided, the calculation difficulty is reduced, and the microwave heating temperature uniformity can be improved.

Description

Technical Field

[0001] This invention relates to a method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction, and belongs to the field of microwave heating modeling and simulation. Background Technology

[0002] With the development of technology, the cost of clean energy development and utilization is constantly decreasing, and people's attention to clean energy is also increasing. Its applications are wide-ranging, covering food processing and chemical production, and it has become a research hotspot in recent years. However, the spatial non-uniformity of the electromagnetic field distribution within the heating cavity leads to differences in the intensity of polar molecular dipole response in different regions of the heated material, resulting in inconsistencies in energy conversion efficiency and the distribution of hot and cold regions. In areas where electromagnetic energy absorption is strong, hot spots may form, leading to localized overheating. This overheating phenomenon not only affects heating uniformity but also increases the risk of material thermal failure. In extreme cases, localized overheating may trigger thermal runaway, manifested as a rapid and uncontrolled rise in temperature, potentially leading to material decomposition, combustion, or even explosion. These consequences bring significant safety hazards, thus limiting the widespread application of microwave heating technology in high-precision or high-safety applications.

[0003] To address the aforementioned issues, recent research has introduced methods to improve heating uniformity and efficiency by adjusting the power, frequency, and phase of microwave sources. Compared to other methods, microwave parameter control (such as power, frequency, and phase) demonstrates significant advantages in improving heating uniformity, offering greater flexibility and adaptability. This method allows for high-precision real-time adjustment via electronic control, eliminating the need for complex mechanical equipment, significantly improving energy utilization efficiency and system stability, and making it more suitable for diverse and dynamic industrial applications. Numerical models hold significant value in microwave heating research. Through theoretical derivation, the electromagnetic field distribution within the cavity can be predicted, and the impact of microwave source parameters on heating performance can be systematically analyzed, providing a theoretical basis for optimizing system design. Simultaneously, it significantly reduces experimental workload and R&D costs, while rapidly validating design schemes, thereby greatly improving R&D efficiency.

[0004] While there has been extensive research on frequency and power regulation in multi-source systems, research on phase difference control is relatively limited. Furthermore, coupling the electromagnetic and temperature fields in a microwave heating system is a complex and cumbersome process. Summary of the Invention

[0005] The technical problem this invention aims to solve is as follows: This invention provides a method for optimizing the temperature field of a phase-controlled multi-source microwave heating system based on numerical model prediction. By establishing a numerical prediction model and using system identification to correct the model parameters, the electric field distribution of the heated material under different relative phases can be accurately predicted. The uniformity of the electric field of the heated material is used as an evaluation index to select the relative phase, aiming to improve the temperature uniformity of microwave heating.

[0006] The technical solution of this invention is: a method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction, the method comprising:

[0007] Step 1: Construct a microwave heating model simulation platform;

[0008] Step 2: Combine the microwave sources on the same side and establish a numerical model based on the physical heating model;

[0009] Step 3: Identify and correct numerical model parameters through the system to improve prediction accuracy;

[0010] Step 4: Substitute the corrected numerical model parameters into the numerical model to predict the electric field distribution under different relative phase differences;

[0011] Step 5: Select the relative phase difference based on the predicted electric field distribution to optimize temperature uniformity.

[0012] Furthermore, Step 1 includes:

[0013] Step 1.1: Construct a microwave heating model using the COMSOL Multiphysic simulation platform, including determining the geometry of the microwave heating system cavity, the configuration of the microwave source, and the physical properties of the heated material.

[0014] The geometry of the microwave heating system cavity includes a cuboid structure with defined width, length, and height.

[0015] The configuration of the microwave sources includes the configuration of the number, type, and location of the microwave sources;

[0016] The physical properties of the heated material include dielectric constant, magnetic permeability, electrical conductivity, and heat capacity;

[0017] Step 1.2: Collect simulation experimental data, including electric field amplitude, initial phase of electric field, and electric field strength;

[0018] The phase parameter of one of the microwave sources is set as the optimization variable, and the microwave heating process is simulated to simulate the electric field distribution and temperature field distribution during the microwave heating process.

[0019] Further, Step 2 includes:

[0020] When two microwave sources on one side excite the cavity separately, a stable standing wave is formed, and the excited modes can be linearly superimposed. However, under simultaneous excitation, the two sources often couple to different channels of the same mode, forming a novel resonant structure dominated by coherent interference. The electric field distribution is no longer a simple superposition of the results of their respective excitations. Considering that the four sources operate simultaneously, combining sources on the same side during modeling helps to more accurately reflect the resonant field structure. Therefore, the microwave sources on the same side are combined, and a numerical model is established based on a physical heating model as follows:

[0021]

[0022] Among them, E ix E iy E iz and Let E represent the electric field amplitude and initial phase in three dimensions when only the microwave source i∈(n,m) is turned on. 3x E 3y E 3z and This represents the measurement parameters when microwave sources 3 and 4 are turned on simultaneously, δ m1 It is the Krone function, indicating the phase difference. It is added only when m=1; by measuring these parameters and substituting them into the above equation, the distribution of each electric field component in the microwave cavity can be predicted; substituting each electric field component into Predict the electric field distribution.

[0023] Furthermore, Step 3 includes:

[0024] To further improve prediction accuracy, the least squares method and gradient descent method are used to optimize the electric field amplitude and initial phase data, so that the electric field strength predicted by the theoretical model is as close as possible to the actual measured value. The error is minimized using weighted least squares, and the nonlinear optimization problem is solved using a quasi-Newton algorithm. To accelerate convergence and avoid getting trapped in local optima, the momentum method is introduced to optimize the gradient descent process, utilizing historical gradient information to improve convergence efficiency. For point i, the phase is... Actual electric field value The objective is to minimize the sum of weighted squared errors between the predicted and actual values:

[0025]

[0026] in, The electric field value is predicted by a numerical model; w ij As weight,

[0027] It is the objective function The gradient, representing the partial derivative of the objective function with respect to the magnitude and initial phase of the optimization variable, describes the trend of the objective function at the current optimization variable point. It is a key quantity used to guide the optimization direction in optimization problems, and is expressed as:

[0028]

[0029] Expanding it, we get:

[0030]

[0031] in, The rest of the developments are similar;

[0032] Finally, the amplitude and phase are updated using the momentum method; the momentum update is performed first:

[0033]

[0034] Among them, g E and These represent the gradients of the electric field amplitude and the initial phase, respectively. and The initial values ​​of all values ​​are 0; β is the momentum factor, which controls the degree of influence of past gradients. and These are the gradients of the objective function with respect to magnitude and phase, respectively; then, the updated gradients are used to adjust the magnitude and phase:

[0035]

[0036] In the above formula, E old and The initial values ​​are the initial electric field amplitude and phase obtained by measurement, and η is the learning rate, which controls the step size of each update. By repeating the momentum update multiple times, the electric field amplitude and initial phase can be updated, thereby improving the prediction accuracy.

[0037] Furthermore, Step 4 includes:

[0038] Substitute the model parameters identified and corrected in Step 3 into the numerical model parameters in Step 2 to predict the electric field distribution.

[0039] Furthermore, Step 5 includes:

[0040] The optimization objective was to improve the temperature uniformity of the sample after microwave heating. Temperature uniformity and electric field uniformity were measured using T... COV and E COV As an evaluation indicator, its calculation formula is as follows:

[0041]

[0042] Among them, T a T is the average temperature of the material, n is the number of points taken, and T is the average temperature of the material. i The temperature of the i-th point of the heating pattern;

[0043] The expression for the power loss density P of electromagnetic waves in a medium is:

[0044]

[0045] Where ω = 2πf is the angular frequency of the electromagnetic wave, f is the frequency; ε0 is the vacuum permittivity, E R 2 The electric field strength is the square of the electric field value; ε (ω) is the imaginary part of the dielectric constant, representing the material loss; power loss is converted into heat through the medium, causing a temperature rise, the rate of which is expressed by the following formula:

[0046]

[0047] Where ρ is the density of the material, in kg / m³. 3 c p The specific heat capacity of the material is expressed in J / (kg·K); from the above formula, the square of the electric field strength E is obtained. R 2 The power loss density is directly proportional to the power loss density P; the stronger the electric field, the greater the power loss density. Simultaneously, the rate of temperature change is also directly proportional to the power loss density P. This is because the electric field strength is affected by the phase difference. The phase difference between microwave sources can be adjusted to effectively control the temperature distribution inside the cavity, reduce the generation of hot and cold spots, and thus significantly improve the uniformity and efficiency of microwave heating.

[0048] This invention can predict the electric field distribution relatively accurately. However, the coupling relationship between the electric field and the temperature field is complex and the implementation process is cumbersome. Based on existing research data, this invention proposes an indirect optimization strategy: by selecting an appropriate relative phase difference through the predicted electric field distribution, the optimal control of the temperature field can be achieved.

[0049] The present invention also provides a temperature field optimization system for a phased-controlled multi-source microwave heating system based on numerical model prediction, the system comprising: a module for executing the temperature field optimization method for a phased-controlled multi-source microwave heating system based on numerical model prediction.

[0050] The beneficial effects of this invention are:

[0051] 1. This invention is based on a physical heating model and combines microwave sources on the same side to construct a finite element numerical model for predicting the electric field distribution under different relative phase differences;

[0052] 2. This invention combines system identification technology to correct numerical model parameters, thereby reducing prediction errors;

[0053] 3. This invention selects relative phase difference for heating based on the predicted electric field distribution to achieve the effect of optimizing temperature uniformity. Attached Figure Description

[0054] Figure 1 This is a flowchart from the present invention;

[0055] Figure 2 This is a cavity model of a multi-microwave source combined heating temperature field in an embodiment of the present invention;

[0056] Figure 3 This is a diagram showing the relationship between temperature uniformity and electric field uniformity in this invention.

[0057] Figure 4 This is a comparison chart of the temperature uniformity of the method proposed in this invention and the temperature uniformity without phase change. Detailed Implementation

[0058] Example 1: As Figures 1-4 As shown, this embodiment uses silicon carbide crystals as the heating material to verify a method for optimizing temperature uniformity; the method includes:

[0059] Step 1: Construct a microwave heating model simulation platform;

[0060] Furthermore, Step 1 includes:

[0061] Step 1.1: Construct a microwave heating model using the COMSOL Multiphysic simulation platform, such as... Figure 2 As shown, this includes determining the geometry of the microwave heating system cavity, the configuration of the microwave source, and the physical properties of the heated material;

[0062] The geometry of the microwave heating system cavity includes a cuboid structure with defined width, length, and height.

[0063] The configuration of the microwave sources includes the number of microwave sources (e.g., Figure 2 The configuration of microwave sources 1-4), their type, and their location;

[0064] The physical properties of the heated material include dielectric constant, magnetic permeability, electrical conductivity, and heat capacity;

[0065] Step 1.2: Collect simulation experimental data, including electric field amplitude, initial phase of electric field, and electric field strength;

[0066] The phase parameter of microwave source 1 is set as the optimization variable, and the microwave heating process is simulated to simulate the electric field distribution and temperature field distribution during the microwave heating process.

[0067] Step 2: Combine the microwave sources on the same side and establish a numerical model based on the physical heating model;

[0068] Further, Step 2 includes:

[0069] When two microwave sources on one side excite the cavity separately, a stable standing wave is formed, and the excited modes can be linearly superimposed. However, under simultaneous excitation, the two sources often couple to different channels of the same mode, forming a novel resonant structure dominated by coherent interference. The electric field distribution is no longer a simple superposition of the results of their respective excitations. Considering that the four sources operate simultaneously, combining sources on the same side during modeling helps to more accurately reflect the resonant field structure. Therefore, the microwave sources on the same side are combined, and a numerical model is established based on a physical heating model as follows:

[0070]

[0071] Among them, E ix E iy E iz and Let E represent the electric field amplitude and initial phase in three dimensions when only the microwave source i∈(n,m) is turned on. 3x E 3y E 3z and This represents the measurement parameters when microwave sources 3 and 4 are turned on simultaneously, δ m1 It is the Krone function, indicating the phase difference. It is added only when m=1; by measuring these parameters and substituting them into the above equation, the distribution of each electric field component in the microwave cavity can be predicted; substituting each electric field component into Predict the electric field distribution.

[0072] Step 3: Identify and correct numerical model parameters through the system to improve prediction accuracy;

[0073] Furthermore, Step 3 includes:

[0074] To further improve prediction accuracy, the least squares method and gradient descent method are used to optimize the electric field amplitude and initial phase data, so that the electric field strength predicted by the theoretical model is as close as possible to the actual measured value. The error is minimized using weighted least squares, and the nonlinear optimization problem is solved using a quasi-Newton algorithm. To accelerate convergence and avoid getting trapped in local optima, the momentum method is introduced to optimize the gradient descent process, utilizing historical gradient information to improve convergence efficiency. For point i, the phase is... Actual electric field value The objective is to minimize the sum of weighted squared errors between the predicted and actual values:

[0075]

[0076] in, The electric field value is predicted by a numerical model; w ij As weight,

[0077] It is the objective function The gradient, representing the partial derivative of the objective function with respect to the magnitude and initial phase of the optimization variable, describes the trend of the objective function at the current optimization variable point. It is a key quantity used to guide the optimization direction in optimization problems, and is expressed as:

[0078]

[0079] Expanding it, we get:

[0080]

[0081] in, The rest of the developments are similar;

[0082] Finally, the amplitude and phase are updated using the momentum method; the momentum update is performed first:

[0083]

[0084] Among them, g E and These represent the gradients of the electric field amplitude and the initial phase, respectively. and The initial values ​​of all values ​​are 0; β is the momentum factor, which controls the degree of influence of past gradients. and These are the gradients of the objective function with respect to magnitude and phase, respectively; then, the updated gradients are used to adjust the magnitude and phase:

[0085]

[0086] In the above formula, E old and The initial values ​​are the initial electric field amplitude and phase obtained by measurement, and η is the learning rate, which controls the step size of each update. By repeating the momentum update multiple times, the electric field amplitude and initial phase can be updated, thereby improving the prediction accuracy.

[0087] Step 4: Substitute the corrected numerical model parameters into the numerical model to predict the electric field distribution under different relative phase differences;

[0088] Furthermore, Step 4 includes:

[0089] Substitute the model parameters identified and corrected in Step 3 into the numerical model parameters in Step 2 to predict the electric field distribution.

[0090] Step 5: Select the relative phase difference based on the predicted electric field distribution to optimize temperature uniformity.

[0091] Furthermore, Step 5 includes:

[0092] The optimization objective was to improve the temperature uniformity of the sample after microwave heating. Temperature uniformity and electric field uniformity were measured using T... COV and E COV As an evaluation indicator, its calculation formula is as follows:

[0093]

[0094] Among them, T a T is the average temperature of the material, n is the number of points taken, and T is the average temperature of the material. i The temperature of the i-th point of the heating pattern;

[0095] The expression for the power loss density P of electromagnetic waves in a medium is:

[0096]

[0097] Where ω = 2πf is the angular frequency of the electromagnetic wave, f is the frequency; ε0 is the vacuum permittivity, E R 2 The electric field strength is the square of the electric field value; ε (ω) is the imaginary part of the dielectric constant, representing the material loss; power loss is converted into heat through the medium, causing a temperature rise, the rate of which is expressed by the following formula:

[0098]

[0099] Where ρ is the density of the material, in kg / m³. 3 c p The specific heat capacity of the material is expressed in J / (kg·K); from the above formula, the square of the electric field strength E is obtained. R 2 The power loss density is directly proportional to the power loss density P; the stronger the electric field, the greater the power loss density; simultaneously, the rate of temperature change is also directly proportional to the power loss density P. This is because the electric field strength is affected by the phase difference. The phase difference between microwave sources can be adjusted to effectively control the temperature distribution inside the cavity, reduce the generation of hot and cold spots, and thus significantly improve the uniformity and efficiency of microwave heating.

[0100] This invention can predict the electric field distribution relatively accurately. However, the coupling relationship between the electric field and the temperature field is complex and the implementation process is cumbersome. Based on existing research data, this invention proposes an indirect optimization strategy: by selecting an appropriate relative phase difference through the predicted electric field distribution, the optimal control of the temperature field can be achieved.

[0101] The present invention also provides a temperature field optimization system for a phased-controlled multi-source microwave heating system based on numerical model prediction, the system comprising: a module for executing the temperature field optimization method for a phased-controlled multi-source microwave heating system based on numerical model prediction.

[0102] The relationship between electric field uniformity and temperature uniformity was analyzed using data. Simulation results show that the temperature field uniformity and electric field uniformity exhibit a peak-and-valley functional relationship as the relative phase difference changes. Notably, within a specific time interval (5s-7s) of microwave heating, when the temperature field uniformity reaches its optimum, the corresponding relative phase difference is usually located near the interval with the worst electric field uniformity, such as... Figure 3 As shown in the figure, this phenomenon can be explained by the temperature change rate formula: within a short heating time interval, the more uneven the electric field distribution (i.e., the more "chaotic"), the greater the difference in the temperature change rate corresponding to each sampling point, thus achieving a spatial compensation effect. This effect helps to balance the temperature growth in different regions, suppress local overheating or undercooling, thereby improving the uniformity of the overall temperature field and reducing the generation of cold and hot spots.

[0103] This invention predicted the electric field distribution at 5s, and the data are shown in Table 1. COV For temperature uniformity, T COV The smaller the value, the more uniform the temperature distribution. COV For the uniformity of the electric field, E ' COV To predict electric field uniformity, data shows that when the temperature field uniformity is optimal, the corresponding relative phase difference is often near the range of worst electric field uniformity. This invention uses a relative phase difference of 110°, which, although not optimal, only reduces uniformity by 0.02%, a limited impact.

[0104] Table 1 shows the selection of relative phase difference for microwave heating.

[0105]

[0106] Based on numerical results analysis, microwave heating can be performed by selecting the relative phase difference according to the predicted electric field distribution. The results show that the obtained relative phase difference selection effectively improves temperature uniformity. Figure 4As shown, the total heating time was 30 seconds. Experiment 1 was a traditional four-microwave source heating system with no relative phase difference, while Experiment 2 was the temperature field optimization method for a phase-controlled multi-source microwave heating system based on numerical model prediction proposed in this invention. Compared with traditional heating methods, the temperature uniformity of this invention was improved by 57.75%. The comparison shows that this invention is far superior to traditional heating methods in controlling multi-objective heating results.

[0107] The specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction, characterized in that: The method includes: Step 1: Construct a microwave heating model simulation platform; Step 2: Combine the microwave sources on the same side and establish a numerical model based on the physical heating model; Step 3: Identify and correct numerical model parameters through the system to improve prediction accuracy; Step 4: Substitute the corrected numerical model parameters into the numerical model to predict the electric field distribution under different relative phase differences; Step 5: Select the relative phase difference based on the predicted electric field distribution to optimize temperature uniformity.

2. The method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction as described in claim 1, characterized in that: Step 1 includes: Step 1.1: Construct a microwave heating model using the COMSOL Multiphysic simulation platform, including determining the geometry of the microwave heating system cavity, the configuration of the microwave source, and the physical properties of the heated material. The geometry of the microwave heating system cavity includes a cuboid structure with defined width, length, and height. The configuration of the microwave sources includes the configuration of the number, type, and location of the microwave sources; The physical properties of the heated material include dielectric constant, magnetic permeability, electrical conductivity, and heat capacity; Step 1.2: Collect simulation experimental data, including electric field amplitude, initial phase of electric field, and electric field strength; The phase parameter of one of the microwave sources is set as the optimization variable, and the microwave heating process is simulated to simulate the electric field distribution and temperature field distribution during the microwave heating process.

3. The method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction according to claim 1, characterized in that: Step 2 includes: The microwave sources on the same side are combined, and a numerical model is established based on the physical heating model as follows: Among them, E ix E iy E iz and Let E represent the electric field amplitude and initial phase in three dimensions when only the microwave source i∈(n,m) is turned on. 3x E 3y E 3z and This represents the measurement parameters when microwave sources 3 and 4 are turned on simultaneously, δ m1 It is the Krone function, indicating the phase difference. It is added only when m=1; by measuring these parameters and substituting them into the above equation, the distribution of each electric field component in the microwave cavity can be predicted; substituting each electric field component into Predict the electric field distribution.

4. The method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction according to claim 1, characterized in that: Step 3 includes: The least squares method and gradient descent method are used to optimize the electric field amplitude and initial phase data; the error is minimized by weighted least squares method, and the nonlinear optimization problem is solved by combining a quasi-Newton algorithm; the momentum method is introduced to optimize the gradient descent process, and historical gradient information is used to improve convergence efficiency; for the i-th point, the phase is... Actual electric field value The objective is to minimize the sum of weighted squared errors between the predicted and actual values: in, The electric field value is predicted by a numerical model; w ij As weight, It is the objective function The gradient, representing the partial derivative of the objective function with respect to the magnitude and initial phase of the optimization variable, describes the trend of the objective function at the current optimization variable point. It is a key quantity used to guide the optimization direction in optimization problems, and is expressed as: Expanding it, we get: in, The rest of the developments are similar; Finally, the amplitude and phase are updated using the momentum method; the momentum update is performed first: Among them, g E and These represent the gradients of the electric field amplitude and the initial phase, respectively. and The initial values ​​of all values ​​are 0; β is the momentum factor, which controls the degree of influence of past gradients. and These are the gradients of the objective function with respect to magnitude and phase, respectively; then, the updated gradients are used to adjust the magnitude and phase: In the above formula, E old and The initial values ​​are the initial electric field amplitude and phase obtained by measurement, and η is the learning rate, which controls the step size of each update. By repeating the momentum update multiple times, the electric field amplitude and initial phase can be updated, thereby improving the prediction accuracy.

5. The method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction according to claim 1, characterized in that: Step 4 includes: Substitute the model parameters identified and corrected in Step 3 into the numerical model parameters in Step 2 to predict the electric field distribution.

6. The method for optimizing the temperature field of a phased-controlled multi-source microwave heating system based on numerical model prediction according to claim 1, characterized in that: Step 5 includes: The optimization objective was to improve the temperature uniformity of the sample after microwave heating. Temperature uniformity and electric field uniformity were measured using T... COV and E COV As an evaluation indicator, its calculation formula is as follows: Among them, T a T is the average temperature of the material, n is the number of points taken, and T is the average temperature of the material. i The temperature of the i-th point of the heating pattern; The expression for the power loss density P of electromagnetic waves in a medium is: Where ω = 2πf is the angular frequency of the electromagnetic wave, f is the frequency; ε0 is the vacuum permittivity, E R 2 The electric field strength is the square of the electric field value; ε (ω) is the imaginary part of the dielectric constant, representing the material loss; power loss is converted into heat through the medium, causing a temperature rise, the rate of which is expressed by the following formula: Where ρ is the density of the material, in kg / m³. 3 c p The specific heat capacity of the material is expressed in J / (kg·K); from the above formula, the square of the electric field strength E is obtained. R 2 It is proportional to the power loss density P; at the same time, the rate of temperature change is also proportional to the power loss density P.

7. A temperature field optimization system for a phase-controlled multi-source microwave heating system based on numerical model prediction, characterized in that, The system includes a module for performing the temperature field optimization method for a phased multi-source microwave heating system based on numerical model prediction as described in any one of claims 1 to 6.

Images