Method and device for determining complex permittivity of irregular object, electronic equipment and storage medium
By designing a combination of fitted and irregular objects, and combining electromagnetic simulation and artificial intelligence model training, the problem of measuring the complex permittivity of irregular objects was solved, and high-precision calculation of the complex permittivity was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-14
- Publication Date
- 2026-05-15
AI Technical Summary
The complex permittivity of irregular objects is difficult to measure accurately, and existing technologies are not well adapted to their special shapes.
By designing combinations of fitting and irregular objects, measuring the S-parameters of the combined objects, and using electromagnetic simulation and artificial intelligence model training, the complex permittivity of the irregular objects is inverted.
It enables accurate calculation of the complex permittivity of irregular objects, improving measurement accuracy and efficiency.
Smart Images
Figure CN119574980B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of electromagnetic technology, and in particular to a method, apparatus, electronic device, and storage medium for determining the complex permittivity of an irregular object. Background Technology
[0002] The complex permittivity of microwaves is a key parameter in antenna and circuit design. Obtaining the complex permittivity value is essential for the microwave application of new materials. Various measurement schemes have been proposed in the industry for different material morphologies. Among these, several broadband measurement methods have been developed to address the design needs of modern broadband communication and broadband radar systems.
[0003] For large, difficult-to-manufacture objects, an open waveguide method has been proposed to achieve broadband measurement. For large, machinable objects, a transmission line method has been proposed. For small objects, methods such as transmission line radiation have been proposed. However, irregular objects are difficult to measure due to their unique shapes. Summary of the Invention
[0004] In view of this, the purpose of this disclosure is to provide a method, apparatus, electronic device and storage medium for determining the complex permittivity of irregular objects, which can specifically solve existing problems.
[0005] Based on the above objectives, in a first aspect, this disclosure proposes a method for determining the complex permittivity of an irregular object, comprising: if the irregular object and the fitting object have been fabricated, and the irregular object has been embedded in the fitting object to form a composite body, the S-parameters of the composite body are measured; electromagnetic simulation is performed based on the irregular object, the fitting object, and the set complex permittivity to simulate the S-parameters of the composite body; an artificial intelligence model is trained based on the set complex permittivity and the simulated S-parameters to obtain a trained artificial intelligence model; and the measured S-parameters are input into the trained artificial intelligence model to obtain the complex permittivity of the irregular object.
[0006] Secondly, a device for determining the complex permittivity of an irregular object is also provided, comprising: a measurement unit configured to measure the S-parameters of the irregular object if the irregular object and the fitting object have been fabricated, and the irregular object has been embedded in the fitting object to form a composite object; a simulation unit configured to perform electromagnetic simulation based on the irregular object, the fitting object, and a set complex permittivity to simulate the S-parameters of the composite object; a training unit configured to train an artificial intelligence model based on the set complex permittivity and the simulated S-parameters to obtain a trained artificial intelligence model; and a result unit configured to input the measured S-parameters into the trained artificial intelligence model to obtain the complex permittivity of the irregular object.
[0007] Thirdly, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor running the computer program to implement the method of the first aspect.
[0008] Fourthly, a computer-readable storage medium is also provided, on which a computer program is stored, the computer program being executed by a processor to implement the method described in any one of the first aspects.
[0009] Fifthly, a computer program product is also provided, comprising a computer program that is executed by a processor to implement the method described in any one of the first aspects.
[0010] In summary, this disclosure has at least the following beneficial effects: it utilizes simulation data to train an artificial intelligence model in real time, and uses the trained artificial intelligence model to calculate the complex permittivity of irregular objects. Attached Figure Description
[0011] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments disclosed in this disclosure and should not be construed as limiting the scope of this disclosure.
[0012] Figure 1 A flowchart is shown for a method for determining the complex permittivity of an irregular object according to an embodiment of the present disclosure;
[0013] Figure 2 A schematic diagram of the shape of an irregular object according to an embodiment of the present disclosure is shown;
[0014] Figure 3 A simulation process according to an embodiment of this disclosure is shown;
[0015] Figure 4 A schematic diagram of the architecture of an artificial intelligence model according to an embodiment of the present disclosure is shown;
[0016] Figure 5 Another schematic diagram of the architecture of an artificial intelligence model according to an embodiment of this disclosure is shown;
[0017] Figure 6 A schematic diagram of a device for determining the complex permittivity of an irregular object according to an embodiment of the present disclosure is shown;
[0018] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure is shown;
[0019] Figure 8 A schematic diagram of a storage medium provided according to an embodiment of the present disclosure is shown. Detailed Implementation
[0020] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.
[0021] It should be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.
[0022] Figure 1 A method for determining the complex permittivity of irregular objects according to this disclosure is illustrated. In embodiments of this disclosure, the method includes:
[0023] Step S101: If the irregular object and the fitting object have been fabricated, and the irregular object has been embedded in the fitting object to obtain a composite body, the S-parameters of the composite body are measured.
[0024] In this embodiment, the entity executing the method for determining the complex permittivity of an irregular object can measure the S-parameters, i.e., the scattering parameters, of the composite material. The complex permittivity can characterize material properties.
[0025] For example, any execution entity can scan the shape of an irregular object and model it using CAD software. Based on the shape of the irregular object, an object that can perfectly fit the irregular test object is designed, i.e., a fitting object. This fitting object can be designed as any regular shape, such as a rectangle, with the same length and width as the rectangular waveguide, and the same thickness as the thickest part of the irregular test object.
[0026] Then, the regular rectangular test object with the irregular test object embedded can be placed into a rectangular waveguide that fits it, and its S-parameters can be measured.
[0027] Optionally, the method further includes: designing the shape of a fitting object that perfectly fits and wraps around the irregular object according to the shape of the irregular object, and transmitting the designed shape to a 3D printing device so that the 3D printing device can manufacture the irregular object and the fitting object according to the shape of the irregular object and the shape of the fitting object, wherein the thickness of the fitting object and the irregular object is consistent.
[0028] Specifically, the designed shape of the fitting object can be exported, and it can be made using 3D printing technology. Irregular objects can be embedded into the 3D-printed fitting object so that they can fit together completely into a rectangular object with the same length and width as the rectangular waveguide.
[0029] Step S102: Perform electromagnetic simulation based on the irregular object, the fitting object, and the set complex permittivity to simulate the S-parameters of the composite object.
[0030] In this embodiment, the irregular object shape and the rectangular object shape are imported into electromagnetic simulation software for electromagnetic simulation to simulate the S-parameters of the combined object.
[0031] The complex permittivity set can be the complex permittivity itself, or other parameters used to determine the complex permittivity, such as the coefficients of uncertain parameters ε′ or ε″.
[0032] In some cases, the complex permittivity of the 3D printing material is a known value. Therefore, the material property of a rectangular object is the known complex permittivity of the 3D printing material, while the complex permittivity of an irregular object is an unknown value. In CST, the material property of the irregular object is set to ε. r =ε′-jε″, where ε′ and ε″ are uncertain parameters. Set the range and step of ε′ and ε″ in CST and perform simulation to obtain the corresponding S-parameters.
[0033] The complex permittivity of irregular objects needs to be set. Specifically, multiple complex permittivity values can be set to meet the sample size requirements during training.
[0034] Step S103: Based on the set complex permittivity and the S-parameters obtained from simulation, the artificial intelligence model is trained to obtain the trained artificial intelligence model.
[0035] In this embodiment, the aforementioned execution entity can train the artificial intelligence model using various methods based on the set complex permittivity and the simulated S-parameters. For example, training can be performed using input information including the complex permittivity and output information including the simulated S-parameters.
[0036] Step S104: Input the measured S-parameters into the trained artificial intelligence model to obtain the complex permittivity of the irregular object.
[0037] In this embodiment, the aforementioned execution entity can input the measured S-parameters into the trained artificial intelligence model, thereby inverting the complex permittivity of the actual irregular object.
[0038] Specifically, the model can output the complex permittivity itself or the other parameters mentioned above used to determine the complex permittivity.
[0039] This embodiment can use simulation data to train an artificial intelligence model in real time, and use the trained artificial intelligence model to calculate the complex permittivity of irregular objects.
[0040] In some optional implementations of any embodiment of this disclosure, training the artificial intelligence model based on the set complex permittivity and the simulated S-parameters includes: using a preset frequency of the electromagnetic wave and the simulated S-parameters as inputs to the artificial intelligence model, using the uncertain parameters in the complex permittivity as outputs to the artificial intelligence model, and training the artificial intelligence model.
[0041] The S-parameters obtained from the simulation, which correspond one-to-one with the material properties of different complex permittivity of irregular objects, are organized. The S-parameters and the preset frequency are used as inputs, and the uncertain parameters ε′ and ε″ in the complex permittivity are used as outputs to train the artificial intelligence (AI) model, and the trained model is obtained.
[0042] In some optional implementations of any embodiment of this disclosure, designing the shape of the fitting object that perfectly fits and encloses the irregular object includes: scanning the shape of the irregular object, modeling it based on the scanning results and modeling software; and using the modeling results to design the shape of the fitting object that perfectly fits and encloses the irregular object.
[0043] The shape of the irregular object is scanned and modeled using the computer-aided drafting tool AutoCAD. Based on the shape of the irregular object, an object is designed that can perfectly fit the irregular test object. This object is designed as a rectangle, with its length and width being the same as the rectangular waveguide, and its thickness being the same as the thickness at the thickest point of the irregular test object.
[0044] In some optional implementations of any embodiment of this disclosure, the electromagnetic simulation based on the irregular object and the fitting object includes: importing the shape of the irregular object and the shape of the fitting object into electromagnetic simulation software; setting the complex permittivity and the frequency of the electromagnetic wave in the electromagnetic simulation software; and performing electromagnetic simulation based on the imported shape and the setting result.
[0045] Optionally, setting the complex permittivity in the electromagnetic simulation software includes: determining an expression for the complex permittivity; and setting a simulation range and step size for uncertain parameters in the expression.
[0046] In some optional implementations of any embodiment of this disclosure, the artificial intelligence model is a backpropagation neural network; training the artificial intelligence model includes: training with the minimum mean square error as a loss function to adjust the weights and biases in the artificial intelligence model, wherein the minimum mean square error is the average of the squared errors between the network output and the true value.
[0047] In some optional implementations of any embodiment of this disclosure, measuring the S-parameters of the composite material includes: placing the composite material in a waveguide that matches the shape of the composite material to measure the S-parameters of the composite material, wherein the simulated frequency range is the same as the dominant mode frequency of the waveguide.
[0048] Figure 2 A schematic diagram of the shape of an irregular object is shown, in which... Figure 2 (a) and Figure 2 (b) shows schematic diagrams of two irregular objects modeled in CAD. Figure 2 (c) A model that can be used in CAD Figure 2 (a) The irregular test object fits perfectly, and after fitting, it becomes a rectangular object with the same length and width as the waveguide, i.e., a fitted object. Figure 2 (d) is a model that can be used in CAD. Figure 2 (b) The irregular test object fits perfectly, and after fitting, it becomes a rectangular object with the same length and width as the waveguide, i.e., a fitted object. In this embodiment, a standard rectangular waveguide is selected, model EIA international standard WR-90, with a dominant mode frequency range of 8.2-12.4GHz, an inner cross-sectional width of 22.86 mm, and an inner cross-sectional height of 10.16 mm. Therefore... Figure 2 (c) and Figure 2 (d) illustrates a rectangular object that can perfectly fit the irregular test object, and after fitting, it has a cross-sectional width of 22.86 mm and a height of 10.16 mm.
[0049] Figure 3 The diagram schematically illustrates the process of obtaining AI model training data through simulation using the electromagnetic simulation software CST, as described in this disclosure. The waveguide selected is the EIA international standard WR-90, with a dominant mode frequency range of 8.2-12.4 GHz, an inner cross-section width of 22.86 mm, and an inner cross-section height of 10.16 mm—a standard rectangular waveguide. During the simulation, the components in CST include the irregular object under test, a 3D-printed rectangular object that fits the irregular object under test, and two ports.
[0050] Furthermore, considering that in the actual measurement, the object under test is placed close to port 2 of the rectangular waveguide, in the simulation settings, port 2 is set to be placed close to one side of the object under test, and the distance between port 1 and port 2 is the waveguide length of 9.78 mm in the actual measurement.
[0051] Furthermore, in the simulation, the component material is set, wherein the material property of the 3D-printed rectangular object that fits the irregular test object is the complex permittivity ε of the 3D-printed material. r =3.0-j0.2, complex permeability μ r =1. In Parameter Sweep, set the material properties of the irregular object. Set the simulation range of ε′ for the complex permittivity of the irregular object to 2 to 10 with a step of 0.2, and the simulation range of ε″ to 0.02 to 0.4 with a step of 0.02. The simulation frequency range is 8.2-12.4 GHz, which is the same as the dominant mode frequency of the selected waveguide, with a frequency step of 0.01 GHz. The size of the simulation dataset is: size(database) = size(frequency) × size(ε′) × size(ε″), with a total of 345,220 sets of data.
[0052] Next, in this embodiment, 345,220 sets of data from CST will be input into the AI model for training.
[0053] Figure 4 and Figure 5 The schematic diagram illustrates the architecture of two AI models proposed in this application for measuring the complex permittivity of irregular objects in conjunction with 3D printing technology.
[0054] The AI model used is the back-propagation neural network (BPNN) from the Artificial Neural Network (ANN). Each ANN model includes an input layer, a hidden layer, and an output layer. The input layer receives input, the hidden layer processes the input, and the output layer generates the processing result. Each layer consists of multiple neurons.
[0055] Each ANN neural network has k hidden layers, and each layer has N hidden layers. K One neuron:
[0056]
[0057] Where i represents the i-th layer, i = 0 is the input layer, i = K+1 is the output layer, and except for the input layer, the data of each layer is a function representation of the data of the previous layer:
[0058] Y (i) =f (i) (W (i) Y (i-1) +b (i) (i = 1, 2, ..., K, K+1)
[0059] Where W is the weight matrix, b is the bias vector, and f is the activation function.
[0060] The training samples in the ANN training set include the input vector and the target output value. Using the least squares approach and the backpropagation algorithm for supervised learning, the network weights W are adjusted to minimize the error between the ANN's calculated output and the target output. In the regression fitting problem, the loss function is represented by the minimum mean square error (MSE).
[0061]
[0062] The main parameters of each ANN include:
[0063] The training objective, or minimum mean squared error (MSE), is a metric that measures the difference between the model's predicted values and the true values. In BPNNs, it is represented as the average of the squared errors between the network output and the true values. The training process minimizes this MSE by adjusting the network parameters (weights and biases), making the network's output closer to the desired output.
[0064] Activation function: In neural networks, the activation function is a non-linear function located inside each neuron (or node). Its role is to transform the input signal of the neuron into the output signal. The introduction of the activation function adds non-linearity to the neural network, enabling it to learn and represent more complex functional relationships, thus improving the network's expressive and approximation capabilities. This embodiment uses the sigmoid function, whose logical slope function is mathematically expressed as:
[0065]
[0066] Learning rate: The learning rate determines the step size of parameter updates along the gradient direction in each update. The learning rate controls the parameter update step size, thus controlling the stability of the training process, influencing the stability of gradient descent, and addressing local optima. A larger learning rate leads to faster model convergence but is prone to oscillations or excessively large parameter updates, even diverging. A smaller learning rate results in slower convergence, requiring more iterations or getting stuck in local optima. Adjusting the learning rate helps the model approach the optimal solution faster during training, improving learning efficiency and approximating the global optimum.
[0067] Training is performed using training data derived from simulation. Each data point in the aforementioned training dataset contains five data points and two output data points. When the S-parameters are complex, the five input data points are: frequency, S-parameter S... 11 The real part ReS of the parameter 11 S-parameters S 11 Imaginary part of the parameter ImS 11 S-parameters S 21 The real part ReS of the parameter 21 S-parameters S 21 Imaginary part of the parameter ImS 21 The two output data are ε′ and ε″ of the corresponding irregular objects. First, the input and output data are normalized to limit their range to a specified interval, thereby eliminating scale differences between different features and improving the training speed and performance of the neural network.
[0068] The BP network is then initialized, and its main parameters are set. In this embodiment, the training target is set to 1×10. -6 The activation function chosen is the sigmoid function, and the learning rate is set to 6 to balance convergence speed and training stability.
[0069] This embodiment employs two AI models: a conventional ANN model and four narrowband models (Sub-ANN models) that are divided according to frequency bands to save computing power and shorten training time. Specifically, the four frequency bands (GHz) are 8.2-9.0, 9.0-10, 10-11, and 11-12.4.
[0070] The narrowband Sub-ANN model used in this embodiment is Figure 4 The diagram schematically illustrates that it includes an input layer consisting of 5 neurons; followed by a first hidden layer containing 50 nodes; then a second hidden layer containing 30 nodes; and finally an output layer containing two nodes. Overall, this ANN model consists of four sub-ANNs, and the training set settings, training time, and training results for each sub-ANN are shown in Table 1.
[0071] The specific structure of another ANN model designed in this embodiment is as follows: Figure 5 It consists of an input layer with 5 neurons; followed by a first hidden layer with 50 nodes; then a second hidden layer with 30 nodes; and finally an output layer with two nodes. The training set settings, training time, and training results are shown in Table 2.
[0072] Table 1 Training set settings, training time, and training results for irregular object 1
[0073]
[0074] In addition, frequency and ReS can be used. 11 Im S 11 ReS 21 ImS 21 The artificial intelligence model is tested using this as input data.
[0075] Table 2 Training set settings, training time, and training results for irregular object 2
[0076]
[0077] This disclosure provides an apparatus for determining the complex permittivity of an irregular object. This apparatus is used to execute the method for determining the complex permittivity of an irregular object as described in the above embodiments. Figure 6As shown, the device includes: a measurement unit 601, configured to measure the S-parameters of an irregular object and a fitting object that completely fits and encloses the irregular object, wherein the irregular object and the fitting object have been fabricated and the irregular object has been embedded in the fitting object to form a composite body; a simulation unit 602, configured to perform electromagnetic simulation based on the irregular object, the fitting object, and a set complex permittivity to simulate the S-parameters of the composite body; a training unit 603, configured to train an artificial intelligence model based on the set complex permittivity and the simulated S-parameters to obtain a trained artificial intelligence model; and a result unit 604, configured to input the measured S-parameters into the trained artificial intelligence model to obtain the complex permittivity of the irregular object.
[0078] The apparatus for determining the complex permittivity of irregular objects provided in the above embodiments of this disclosure and the method for determining the complex permittivity of irregular objects provided in the embodiments of this disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.
[0079] This disclosure also provides an electronic device corresponding to the method for determining the complex permittivity of irregular objects provided in the foregoing embodiments, for executing the aforementioned method for determining the complex permittivity of irregular objects. This disclosure is not limiting.
[0080] Please refer to Figure 7 This illustrates a schematic diagram of an electronic device provided by some embodiments of the present disclosure. For example... Figure 7 As shown, the electronic device 70 includes: a processor 700, a memory 701, a bus 702, and a communication interface 703. The processor 700, the communication interface 703, and the memory 701 are connected via the bus 702. The memory 701 stores a computer program that can run on the processor 700. When the processor 700 runs the computer program, it executes the method provided in any of the foregoing embodiments of this disclosure.
[0081] The memory 701 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 703 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.
[0082] Bus 702 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 701 is used to store programs. After receiving an execution instruction, the processor 700 executes the program. The method for determining the complex permittivity of the irregular object disclosed in any of the foregoing embodiments of this disclosure can be applied to the processor 700, or implemented by the processor 700.
[0083] The processor 700 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 700 or by instructions in software form. The processor 700 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this disclosure. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this disclosure can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 701. Processor 700 reads the information in memory 701 and, in conjunction with its hardware, completes the steps of the above method.
[0084] The electronic device provided in this disclosure and the method for determining the complex permittivity of irregular objects provided in this disclosure are based on the same inventive concept and have the same beneficial effects as the methods they employ, operate, or implement.
[0085] This disclosure also provides a computer-readable storage medium corresponding to the method for determining the complex permittivity of irregular objects provided in the foregoing embodiments. Please refer to... Figure 8 The computer-readable storage medium shown is an optical disc 80, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it executes the method for determining the complex permittivity of irregular objects provided in any of the foregoing embodiments.
[0086] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.
[0087] The computer-readable storage medium provided in the above embodiments of this disclosure and the method for determining the complex permittivity of irregular objects provided in the embodiments of this disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.
[0088] It should be noted that:
[0089] In the foregoing text, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in this disclosure is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
[0090] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this disclosure.
[0091] The embodiments of this disclosure have been described above with reference to the accompanying drawings. These are merely specific implementations of this disclosure, but this disclosure is not limited to the specific implementations described above. The specific implementations described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this disclosure without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this disclosure.
Claims
1. A method for determining the complex permittivity of an irregular object, characterized in that, include: If the irregular object and the fitting object have been fabricated, and the irregular object has been embedded in the fitting object to obtain a composite body, the S-parameters of the composite body are measured. Electromagnetic simulation is performed based on the irregular object, the fitting object, and the set complex permittivity to simulate the S-parameters of the composite object; Based on the complex permittivity set above and the S-parameters obtained from simulation, the artificial intelligence model is trained to obtain the trained artificial intelligence model. The measured S-parameters are input into the trained artificial intelligence model to obtain the complex permittivity of the irregular object; The training of the artificial intelligence model based on the set complex permittivity and the simulated S-parameters includes: The preset frequency of the electromagnetic wave and the S-parameters obtained from the simulation are used as inputs to the artificial intelligence model, and the uncertain parameters in the complex permittivity are used as outputs to train the artificial intelligence model. The electromagnetic simulation based on the irregular object and the fitting object includes: The shapes of the irregular object and the fitting object are imported into electromagnetic simulation software; the complex permittivity and the frequency of the electromagnetic wave are set in the electromagnetic simulation software; electromagnetic simulation is performed based on the imported shape and the setting results. Setting the complex permittivity in the electromagnetic simulation software includes: Determine the expression for the complex permittivity; set the simulation range and step size for the uncertain parameters in the expression.
2. The method according to claim 1, characterized in that, The method further includes: Based on the shape of the irregular object, design the shape of a fitting object that perfectly fits and wraps around the irregular object, and transmit the designed shape to the 3D printing equipment so that the 3D printing equipment can manufacture the irregular object and the fitting object according to the shape of the irregular object and the shape of the fitting object respectively, with the fitting object and the irregular object having the same thickness.
3. The method according to claim 2, characterized in that, Based on the shape of the irregular object, design the shape of a fitting object that perfectly surrounds the irregular object, including: The shape of irregular objects is scanned, and models are created based on the scan results and modeling software. Using the modeling results, design the shape of a fitting object that perfectly encloses the irregular object.
4. The method according to claim 1, characterized in that, The artificial intelligence model is a backpropagation neural network; The training of the artificial intelligence model includes: The minimum mean squared error is used as the loss function for training to adjust the weights and biases in the artificial intelligence model. The minimum mean squared error is the average of the squared errors between the network output and the true value.
5. The method according to claim 1, characterized in that, The measurement of the S-parameters of the composition includes: The composite material is placed in a waveguide whose shape matches that of the composite material to measure the S-parameters of the composite material. The simulated frequency range is the same as the dominant mode frequency of the waveguide.
6. A device for determining the complex permittivity of an irregular object, characterized in that, include: The measuring unit is configured to measure the S-parameters of the composite body if the irregular object and the fitting object have been fabricated, and the irregular object has been embedded in the fitting object to form a composite body, given the shape of the irregular object and the shape of the fitting object that completely fits the irregular object; The simulation unit is configured to perform electromagnetic simulation based on the irregular object, the fitting object, and the set complex permittivity, so as to simulate the S-parameters of the composite object; The training unit is configured to train the artificial intelligence model based on the complex permittivity and the simulated S-parameters, thereby obtaining the trained artificial intelligence model. The result unit is configured to input the measured S-parameters into a trained artificial intelligence model to obtain the complex permittivity of the irregular object; The training of the artificial intelligence model based on the set complex permittivity and the simulated S-parameters includes: The preset frequency of the electromagnetic wave and the S-parameters obtained from the simulation are used as inputs to the artificial intelligence model, and the uncertain parameters in the complex permittivity are used as outputs to train the artificial intelligence model. The electromagnetic simulation based on the irregular object and the fitting object includes: The shapes of the irregular object and the fitting object are imported into electromagnetic simulation software; the complex permittivity and the frequency of the electromagnetic wave are set in the electromagnetic simulation software; electromagnetic simulation is performed based on the imported shape and the setting results. Setting the complex permittivity in the electromagnetic simulation software includes: Determine the expression for the complex permittivity; set the simulation range and step size for the uncertain parameters in the expression.
7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method as described in any one of claims 1-5.