Magnetic field modeling method and device of electromagnetic navigation system, electronic equipment and storage medium
By combining three-dimensional Hall sensor array and residual convolutional neural network, the problem of poor reliability of traditional magnetic field modeling under high magnetic field strength is solved, and efficient magnetic field modeling of electromagnetic navigation systems is realized, which is suitable for multi-electromagnetic systems. Improves system performance and stability.
Patent Information
- Application Number
- CN202411095644.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-29
AI Technical Summary
It is difficult for traditional electromagnetic navigation systems to obtain reliable magnetic field models under high magnetic field strength, especially because the current and magnetic field have a complex nonlinear relationship, which limits the effectiveness of the magnetic field modeling method.
Using a method based on a three-dimensional Hall sensor array and a residual convolutional neural network, the electromagnetic field model is trained by collecting the magnetic field intensity and input current of multiple working points, and it is suitable for multi-electromagnet systems to achieve mapping of nonlinear relationships.
It improves the reliability and applicability of magnetic field modeling of electromagnetic navigation systems, can be effectively applied in different multi-electromagnetic systems, and enhances the performance and stability of magnetic navigation systems.
Smart Images

Figure CN119047417A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of medical device technology, and more specifically, to a magnetic field modeling method, device, electronic device and storage medium for an electromagnetic navigation system. Background Art
[0002] In recent years, magnetron technology has developed rapidly in the medical field. Magnetron technology has been used in magnetic navigation, targeted drug delivery, minimally invasive surgery, diagnosis and other fields. Its development has benefited from the increasing maturity of materials science and magnet manufacturing technology, as well as the inherent characteristics of the magnetic field, that is, unlike light and ultrasound, it can penetrate the human body without interference and will not cause damage to the human body, making it have great potential in the field of human vascular treatment.
[0003] Among them, the magnetic navigation technology in the magnetic control technology is to control medical equipment such as magnetic robots or magnetic guidewires to shuttle through small blood vessels by precisely manipulating the magnetic field in the working area, and can reach the human body positions that are difficult to reach with traditional methods, thereby greatly increasing the accuracy and stability of the operation. There are two common methods to generate the magnetic field required for magnetic navigation. Either the required magnetic field is generated by manipulating the rotation of the robotic arm or pushing the permanent magnet, or the magnetic field is generated by controlling the input current to control the multi-pole electromagnet. The magnetic navigation system using multi-pole electromagnets is called an electromagnetic navigation system.
[0004] The above two methods have their own advantages and disadvantages. The system based on permanent magnets is smaller in scale, has fewer magnets, has lower production costs, and the modeling and adjustment of the magnetic field are simpler and more intuitive. However, it requires precise control of the robotic arm to cooperate. The system based on electromagnets can generate a stronger magnetic field and can regulate the magnetic field gradient, giving the magnetic navigation system more degrees of freedom. However, the modeling and adjustment of the magnetic field will become more complicated.
[0005] The magnetic field modeling of the electromagnetic navigation system can be done by using the finite element analysis method, or by using the method of numerical interpolation of the measured spatial magnetic field, or by using a physics-based multipole model to fit the measured magnetic field. However, these methods are based on the assumption that the input current is linearly related to the magnetic field. When the magnetic field strength increases, the electromagnet core will saturate and magnetize, resulting in a complex nonlinear relationship between the current and the magnetic field, making the above traditional modeling methods difficult to apply and difficult to obtain a reliable electromagnetic field model. Summary of the invention
[0006] In view of this, the present application provides a magnetic field modeling method, device, electronic device and storage medium for an electromagnetic navigation system, which are used to model the magnetic field of an electromagnetic navigation system with multiple electromagnets to obtain an electromagnetic field model with better reliability so that it can be applied to different multi-electromagnet systems.
[0007] In order to achieve the above objectives, the proposed solution is as follows:
[0008] A magnetic field modeling method for a multi-electromagnet system is applied to electronic equipment, and the magnetic field modeling method comprises the following steps:
[0009] Based on a three-dimensional Hall sensor array, the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of an input current is collected to obtain a set of three-dimensional magnetic field information;
[0010] Model training is performed based on the input current and the set of three-dimensional magnetic field information to obtain an electromagnetic field model.
[0011] Optionally, the three-dimensional Hall sensor array includes multiple layers of printed circuit boards arranged in parallel, and a planar Hall sensor array is arranged on each layer of the printed circuit board.
[0012] Optionally, the performing model training based on the input current and the set of three-dimensional magnetic field information comprises the steps of:
[0013] The electromagnetic field model is obtained by training the residual neural network by taking the input current as the input of the residual convolutional neural network and taking the set of three-dimensional magnetic field information as the output of the residual convolutional neural network.
[0014] Optionally, the residual convolutional neural network includes an input layer, a hidden layer and an output layer, wherein:
[0015] The input layer is a linear fully connected layer;
[0016] The hidden layer includes a large network block that is repeatedly connected, the large network block includes a small network block that is repeatedly connected, and the small network block includes a multi-layer convolution layer and a residual network connected to the multi-layer convolution layer;
[0017] The output layer has 3 convolution kernels, 1 step size, and 3 filters.
[0018] A magnetic field modeling device for a multi-electromagnet system, applied to electronic equipment, comprising:
[0019] A parameter acquisition module is configured to acquire the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current based on the three-dimensional Hall sensor array to obtain a set of three-dimensional magnetic field information;
[0020] The modeling execution module is configured to perform model training based on the input current and the set of three-dimensional magnetic field information to obtain an electromagnetic field model.
[0021] Optionally, the three-dimensional Hall sensor array includes multiple layers of printed circuit boards arranged in parallel, and a planar Hall sensor array is arranged on each layer of the printed circuit board.
[0022] Optionally, the modeling execution module is configured to train the residual neural network by taking the input current as the input of the residual convolutional neural network and taking the set of three-dimensional magnetic field information as the output of the residual convolutional neural network to obtain the electromagnetic field model.
[0023] Optionally, the residual convolutional neural network includes an input layer, a hidden layer and an output layer, wherein:
[0024] The input layer is a linear fully connected layer;
[0025] The hidden layer includes a large network block that is repeatedly connected, the large network block includes a small network block that is repeatedly connected, and the small network block includes a multi-layer convolution layer and a residual network connected to the multi-layer convolution layer;
[0026] The output layer has 3 convolution kernels, 1 step size, and 3 filters.
[0027] An electronic device, comprising at least one processor and a memory connected to the processor, wherein:
[0028] The memory is used to store computer programs or instructions;
[0029] The processor is used to execute the computer program or instruction so that the electronic device implements the magnetic field modeling method as described above.
[0030] A storage medium is applied to an electronic device, wherein the storage medium carries one or more computer programs, and the one or more computer programs can be executed by the electronic device, so that the electronic device can implement the magnetic field modeling method as described above.
[0031] It can be seen from the above technical scheme that the present application discloses a magnetic field modeling method, device, electronic device and storage medium for a multi-electromagnet system. The method and device are applied to electronic devices, specifically, based on a three-dimensional Hall sensor array, the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current is collected to obtain a set of three-dimensional magnetic field information; based on the input current and the three-dimensional magnetic field tensor B corresponding to a set of three-dimensional magnetic field information, the model is trained to obtain an electromagnetic field model. The present application fully considers the complex nonlinear relationship between current and magnetic field, and obtains an electromagnetic model by training the residual convolutional neural network based on the magnetic field strength and input current collected by the three-dimensional Hall sensor array, so that the obtained electromagnetic field model has strong reliability and can be applied to different multi-electromagnet systems. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0033] Figure 1 A flow chart of a magnetic field modeling method for a multi-electromagnet system according to an embodiment of the present application;
[0034] Figure 2 A schematic diagram of a three-dimensional Hall sensor array according to an embodiment of the present application;
[0035] Figure 3 A schematic diagram of a residual convolutional neural network according to an embodiment of the present application;
[0036] Figure 4 A block diagram of a magnetic field modeling device for a multi-electromagnet system according to an embodiment of the present application;
[0037] Figure 5 A block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0039] This application aims to solve the problem of poor reliability of traditional magnetic field modeling. This application proposes a fast magnetic field modeling method based on a three-dimensional array of 3D Hall effect sensors and a machine learning algorithm. The modeled magnetic field can obtain the magnetic field of the working area in real time according to the input current. Different from the traditional use of a three-dimensional Gaussmeter to measure the magnetic field size at a single point, this patent uses an m*m*m three-dimensional Hall effect sensor array, which can collect the magnetic field of m33 points in the area of interest in real time at one time, greatly improving the measurement efficiency. Combined with the nonlinear mapping capability of neural networks, this method can learn the nonlinear relationship between the input current and the magnetic field, and is suitable for linear and nonlinear scenarios, achieving fast and reliable magnetic field modeling.
[0040] This innovative modeling method not only improves the performance and reliability of the electromagnetic navigation system, but also provides important support for the further development of medical magnetic control technology.
[0041] Figure 1 This is a flow chart of a magnetic field modeling method for a multi-electromagnet system according to an embodiment of the present application.
[0042] like Figure 1 As shown, the magnetic field modeling method provided in the present application is applied to an electronic device for modeling the magnetic field of a multi-electromagnet system. The electronic device can be understood as a computer, server or cloud platform with data computing and information processing capabilities. The magnetic field modeling method includes the following steps:
[0043] S1. Based on the three-dimensional Hall sensor array, the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current is collected to obtain a set of three-dimensional magnetic field information.
[0044] In a multi-electromagnet system built with n electromagnets, an input current i is input into it. This i is a vector of n x1, and its components represent the driving current input to the coil of each electromagnet. The multi-electromagnet system will have a magnetic field strength b at position r in its working area. The expression of this magnetic field strength can be expressed as:
[0045] b(r)=f(r,i), (1)
[0046] If the current is linearly related to the magnetic field strength, the above formula can be used to capture Huawei:
[0047] b(r)=A(r)i, (2)
[0048] Where A(r) is a 3x n excitation matrix.
[0049] When collecting the magnetic field strength, the present application detects the working area in the multi-electromagnet system through a three-dimensional Hall sensor, thereby obtaining the set of field strength data.
[0050] The three-dimensional Hall sensor array in the present application includes multiple layers of parallel printed circuit boards PCB, which are fixed by pillars so that a preset distance is maintained between the printed circuit boards. A planar Hall sensor array is arranged on each printed circuit board. The so-called planar Hall sensor array means that Hall sensors 100 are arranged on each printed circuit board according to a fixed row spacing and column spacing. For the convenience of calculation, the distance between the printed circuit boards and the row spacing and column spacing of the Hall sensor array on each printed circuit board can be equal, and the number of printed circuit boards is the same as the number of Hall sensors on each row and the number of Hall sensors on each column, so that the three-dimensional Hall sensor array constitutes an array with m xmx m uniform arrangement, such as Figure 2 As shown in the figure, the array can collect the three-dimensional magnetic field information of m3 spatial points at one time, which can improve the data collection efficiency.
[0051] S2. Perform model training based on input current and a three-dimensional magnetic field tensor B corresponding to a set of three-dimensional magnetic field information to obtain an electromagnetic field model.
[0052] In order to obtain the mapping equation f between the current and the magnetic field, the present application constructs a residual convolutional neural network, whose input is the vector i and outputs the three-dimensional magnetic field information measured by the Hall sensor corresponding to the three-dimensional magnetic field tensor B of mxmxm x3. The input layer of the residual convolutional neural network of the present application is a linear fully connected layer, which maps and reshapes the input vector of mx 1 into where q is the number of large network blocks in the residual convolutional neural network and P is the number of filters in the convolutional layer.
[0053] Next, a hidden layer is connected. The hidden layer includes a large network grid that is repeatedly connected. The large network grid includes a small network grid that is repeatedly connected. The small network grid includes multiple convolutional layers and a residual network connected to the multiple convolutional layers. The convolutional layer has 3 convolution kernels, a step size of 1, and P filters. The convolutional layer outputs to the LReLu activation function to provide nonlinearity for the residual convolutional neural network. By repeatedly connecting the above convolutional layers and activation functions r times, and then using a residual to link the input of the first convolutional layer and the output of the last convolutional layer, the robustness of the residual convolutional neural network is increased and the gradient vanishing effect is reduced. The above multi-layer convolutional layers and residual connections constitute a small network grid SB. By repeating the small network grid t times, and then using the nearest neighbor interpolation method, the output of the small network grid is twice upsampled to form a large network grid BB. Repeat the large network grid q times and finally output it to the output layer.
[0054] The output layer is a convolutional network with a convolution kernel of 3, a step size of 1, and a filter number of 3. The final output is a three-dimensional magnetic field tensor B of m xm xm x3. The complete residual neural network of this application is as follows Figure 3 After obtaining the tensor magnetic field of the prediction grid, the radial basis function interpolation method is used to perform three-dimensional interpolation, and the magnetic field information of any point in the working area can be obtained.
[0055] The loss function of the residual convolutional neural network in this application is:
[0056]
[0057] Among them, b m is the tensor magnetic field in the working area measured by the Hall sensor, b p is the tensor magnetic field predicted by the residual convolutional neural network. ||·||1 is the L1 norm, λ b and λ gThe two scalar parameters control the weights of the magnetic field L1 norm and the gradient magnetic field L1 norm. It can be seen that the loss function requires not only that the predicted magnetic field values are similar in value, but also that the magnetic field gradients are similar. The tensor magnetic field predicted by this model will have better continuity and accuracy.
[0058] It can be seen from the above technical scheme that this embodiment provides a magnetic field modeling method for a multi-electromagnet system, which is applied to electronic devices, specifically, based on a three-dimensional Hall sensor array, the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current is collected to obtain a set of three-dimensional magnetic field information; based on the input current and the three-dimensional magnetic field tensor B corresponding to a set of three-dimensional magnetic field information, the model is trained to obtain an electromagnetic field model. This application fully considers the complex nonlinear relationship between current and magnetic field, and obtains an electromagnetic model by training the residual convolutional neural network based on the magnetic field strength and input current collected by the three-dimensional Hall sensor array, so that the obtained electromagnetic field model has strong reliability and can be applied to different multi-electromagnet systems.
[0059] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0060] Although operations are depicted in a particular order, this should not be understood as requiring that the operations be performed in the particular order shown or in a sequential order.Multitasking and parallel processing may be advantageous under certain circumstances.
[0061] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.
[0062] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as C or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer.
[0063] Figure 4 This is a block diagram of a magnetic field modeling device for a multi-electromagnet system according to an embodiment of the present application.
[0064] like Figure 4 As shown, the magnetic field modeling device provided in the present application is applied to an electronic device for modeling the magnetic field of a multi-electromagnet system. The electronic device can be understood as a computer, server or cloud platform with data computing and information processing capabilities. The magnetic field modeling device includes a parameter acquisition module 10 and a modeling execution module 20.
[0065] The parameter acquisition module is used to acquire the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current based on the three-dimensional Hall sensor array, and obtain a set of three-dimensional magnetic field information.
[0066] In a multi-electromagnet system built with n electromagnets, an input current i is input into it. This i is a vector of n x1, and its components represent the driving current input to the coil of each electromagnet. The multi-electromagnet system will have a magnetic field strength b at position r in its working area. The expression of this magnetic field strength can be expressed as:
[0067] b(r)=f(r,i), (1)
[0068] If the current is linearly related to the magnetic field strength, the above formula can be used to capture Huawei:
[0069] b(r)=A(r)i, (2)
[0070] Where A(r) is a 3x n excitation matrix.
[0071] When collecting the magnetic field strength, the present application detects the working area in the multi-electromagnet system through a three-dimensional Hall sensor, thereby obtaining the set of field strength data.
[0072] The three-dimensional Hall sensor array in the present application includes multiple layers of parallel printed circuit boards PCB, which are fixed by pillars so that a preset distance is maintained between the printed circuit boards. A planar Hall sensor array is arranged on each printed circuit board. The so-called planar Hall sensor array means that Hall sensors 100 are arranged on each printed circuit board according to a fixed row spacing and column spacing. For the convenience of calculation, the distance between the printed circuit boards and the row spacing and column spacing of the Hall sensor array on each printed circuit board can be equal, and the number of printed circuit boards is the same as the number of Hall sensors on each row and the number of Hall sensors on each column, so that the three-dimensional Hall sensor array constitutes an array with m xmx m uniform arrangement, such as Figure 2 As shown in the figure, the array can collect the three-dimensional magnetic field information of m3 spatial points at one time, which can improve the data collection efficiency.
[0073] The modeling execution module is used to perform model training based on the input current and a three-dimensional magnetic field tensor B corresponding to a set of three-dimensional magnetic field information to obtain an electromagnetic field model.
[0074] In order to obtain the mapping equation f between the current and the magnetic field, the present application constructs a residual convolutional neural network, whose input is the vector i and outputs the three-dimensional magnetic field information measured by the Hall sensor corresponding to the three-dimensional magnetic field tensor B of mxmxm x3. The input layer of the residual convolutional neural network of the present application is a linear fully connected layer, which maps and reshapes the input vector of mx 1 into where q is the number of large network blocks in the residual convolutional neural network and P is the number of filters in the convolutional layer.
[0075] Next, a hidden layer is connected. The hidden layer includes a large network grid that is repeatedly connected. The large network grid includes a small network grid that is repeatedly connected. The small network grid includes multiple convolutional layers and a residual network connected to the multiple convolutional layers. The convolutional layer has 3 convolution kernels, a step size of 1, and P filters. The convolutional layer outputs to the LReLu activation function to provide nonlinearity for the residual convolutional neural network. By repeatedly connecting the above convolutional layers and activation functions r times, and then using a residual to link the input of the first convolutional layer and the output of the last convolutional layer, the robustness of the residual convolutional neural network is increased and the gradient vanishing effect is reduced. The above multi-layer convolutional layers and residual connections constitute a small network grid SB. By repeating the small network grid t times, and then using the nearest neighbor interpolation method, the output of the small network grid is twice upsampled to form a large network grid BB. Repeat the large network grid q times and finally output it to the output layer.
[0076] The output layer is a convolutional network with a convolution kernel of 3, a step size of 1, and a filter number of 3. The final output is a three-dimensional magnetic field tensor B of m xm xm x3. The complete residual neural network of this application is as follows Figure 3 After obtaining the tensor magnetic field of the prediction grid, the radial basis function interpolation method is used to perform three-dimensional interpolation, and the magnetic field information of any point in the working area can be obtained.
[0077] The loss function of the residual convolutional neural network in this application is:
[0078]
[0079] Among them, b m is the tensor magnetic field in the working area measured by the Hall sensor, b p is the tensor magnetic field predicted by the residual convolutional neural network. ||·||1 is the L1 norm, λ b and λ g The two scalar parameters control the weights of the magnetic field L1 norm and the gradient magnetic field L1 norm. It can be seen that the loss function requires not only that the predicted magnetic field values are similar in value, but also that the magnetic field gradients are similar. The tensor magnetic field predicted by this model will have better continuity and accuracy.
[0080] It can be seen from the above technical scheme that this embodiment provides a magnetic field modeling device for a multi-electromagnet system, which is applied to electronic equipment, specifically, based on a three-dimensional Hall sensor array, the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current is collected to obtain a set of three-dimensional magnetic field information; based on the input current and the three-dimensional magnetic field tensor B corresponding to a set of three-dimensional magnetic field information, the model is trained to obtain an electromagnetic field model. This application fully considers the complex nonlinear relationship between current and magnetic field, and obtains an electromagnetic model by training the residual convolutional neural network based on the magnetic field strength and input current collected by the three-dimensional Hall sensor array, so that the obtained electromagnetic field model has strong reliability and can be applied to different multi-electromagnet systems.
[0081] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".
[0082] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0083] Figure 5A block diagram of an electronic device according to an embodiment of the present application.
[0084] Reference below Figure 5 , which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The electronic device is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0085] The electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory ROM 502 or a program loaded from an input device 506 into a random access memory RAM 503. In the RAM, various programs and data required for the operation of the electronic device are also stored. The processing device, ROM, and RAM are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0086] Typically, the following devices may be connected to the I / O interface: input devices including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 508 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 509. The communication device 509 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an electronic device with various devices, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have alternatively.
[0087] The present application also provides a computer-readable storage medium embodiment.
[0088] The computer-readable storage medium is applied to an electronic device and carries one or more computer programs. When the one or more computer programs are executed by the electronic device, the electronic device collects the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current based on the three-dimensional Hall sensor array to obtain a set of three-dimensional magnetic field information; the model is trained based on the input current and the three-dimensional magnetic field tensor B corresponding to the set of three-dimensional magnetic field information to obtain an electromagnetic field model. The present application fully considers the complex nonlinear relationship between current and magnetic field, and obtains an electromagnetic model by training the residual convolutional neural network based on the magnetic field strength and input current collected by the three-dimensional Hall sensor array, so that the obtained electromagnetic field model has strong reliability and can be applied to different multi-electromagnet systems.
[0089] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0090] In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.
[0091] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0092] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.
[0093] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.
[0094] The technical solution provided by the present invention is introduced in detail above. Specific examples are used in this article to illustrate the principle and implementation mode of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation mode and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A magnetic field modeling method for a multi-electromagnet system, applied to electronic equipment, It is characterized in that The magnetic field modeling method comprises the steps of: Based on a three-dimensional Hall sensor array, the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of an input current is collected to obtain a set of three-dimensional magnetic field information; Model training is performed based on the input current and the set of three-dimensional magnetic field information to obtain an electromagnetic field model.
2. The magnetic field modeling method according to claim 1, It is characterized in that The three-dimensional Hall sensor array comprises multiple layers of parallel and flat printed circuit boards, and a planar Hall sensor array is arranged on each layer of the printed circuit board.
3. The magnetic field modeling method according to claim 1, It is characterized in that The model training based on the input current and the set of three-dimensional magnetic field information comprises the steps of: The electromagnetic field model is obtained by training the residual neural network by taking the input current as the input of the residual convolutional neural network and taking the set of three-dimensional magnetic field information as the output of the residual convolutional neural network.
4. The magnetic field modeling method according to claim 3, It is characterized in that The residual convolutional neural network includes an input layer, a hidden layer and an output layer, wherein: The input layer is a linear fully connected layer; The hidden layer includes a large network block that is repeatedly connected, the large network block includes a small network block that is repeatedly connected, and the small network block includes a multi-layer convolution layer and a residual network connected to the multi-layer convolution layer; The output layer has 3 convolution kernels, 1 step size, and 3 filters.
5. A magnetic field modeling device for a multi-electromagnet system, applied to electronic equipment, It is characterized in that The magnetic field modeling device comprises: A parameter acquisition module is configured to acquire the magnetic field strength of multiple working points in the working area of the multi-electromagnet system under the excitation of the input current based on the three-dimensional Hall sensor array to obtain a set of three-dimensional magnetic field information; The modeling execution module is configured to perform model training based on the input current and the set of three-dimensional magnetic field information to obtain an electromagnetic field model.
6. The magnetic field modeling device according to claim 5, It is characterized in that The three-dimensional Hall sensor array comprises multiple layers of parallel and flat printed circuit boards, and a planar Hall sensor array is arranged on each layer of the printed circuit board.
7. The magnetic field modeling device according to claim 5, It is characterized in that The modeling execution module is configured to train the residual neural network by taking the input current as the input of the residual convolutional neural network and taking the set of three-dimensional magnetic field information as the output of the residual convolutional neural network to obtain the electromagnetic field model.
8. The magnetic field modeling device according to claim 7, It is characterized in that The residual convolutional neural network includes an input layer, a hidden layer and an output layer, wherein: The input layer is a linear fully connected layer; The hidden layer includes a large network block that is repeatedly connected, the large network block includes a small network block that is repeatedly connected, and the small network block includes a multi-layer convolution layer and a residual network connected to the multi-layer convolution layer; The output layer has 3 convolution kernels, 1 step size, and 3 filters.
9. An electronic device, It is characterized in that The electronic device comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs or instructions; The processor is used to execute the computer program or instruction so that the electronic device implements the magnetic field modeling method according to any one of claims 1 to 4.
10. A storage medium, applied to an electronic device, It is characterized in that The storage medium carries one or more computer programs, and the one or more computer programs can be executed by the electronic device, so that the electronic device can implement the magnetic field modeling method according to any one of claims 1 to 4.
