Automatic control method of gradient driving system for ultralow field magnetic resonance

Abstract

The invention provides an automatic control method of a gradient driving system for ultralow-field magnetic resonance, and relates to the technical field of ultralow-field magnetic resonance imaging systems.The automatic control method comprises the steps that the target switching frequency of the gradient driving system is determined; determining a corresponding target sampling period and an initial control parameter of the gradient driving system according to the target switching frequency; determining a target control parameter of the gradient driving system according to a target current and the initial control parameter; and determining a target duty ratio signal of the gradient driving system according to the target control parameter, and realizing dynamic adjustment of the switching frequency of the gradient driving system, thereby reducing the switching loss.

Description

Technical Field

[0001] The present invention relates to the technical field of ultra-low field magnetic resonance imaging systems, and in particular to an automatic control method for a gradient drive system for ultra-low field magnetic resonance. Background Art

[0002] An ultra-low field magnetic resonance system includes a radio frequency transmitting and receiving system and a gradient drive system. During the operation of the gradient drive system, a large amount of heat is generated, and this part of the heat mainly comes from the switching loss of the switching tubes, which is proportional to the switching frequency. In order to achieve better dynamic performance and a more compact design (reducing the size of inductors and capacitors), it is often necessary to continuously increase the switching frequency, which makes the energy density of the switching tubes very large. The accumulated heat causes the junction temperature of the switching tubes to rise, thereby affecting the service life. The traditional gradient drive system controller design often focuses on the dynamic performance under large currents, and it does not specifically optimize for small currents or low-frequency currents, resulting in a very low energy efficiency conversion ratio under light loads. The traditional gradient drive system controller usually adopts a fixed switching frequency design, and even maintains a high switching frequency at small currents or low frequencies. Each switching operation generates losses (such as the on / off losses of the switching tubes, drive losses), and the switching loss is proportional to the frequency. Therefore, it is difficult for the traditional gradient drive system controller to simultaneously achieve high dynamic performance and low power consumption. At low fields, it is necessary to design a gradient drive system with both high integration and high performance, so it is particularly important to adaptively dynamically adjust the switching rate. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide an automatic control method for a gradient drive system for ultra-low field magnetic resonance, so as to adaptively dynamically adjust the switching frequency of the gradient drive system and automatically match the corresponding PID control signal, thereby reducing the switching loss.

[0004] An embodiment of the present invention provides an automatic control method for a gradient drive system for ultra-low field magnetic resonance.

[0005] Providing an automatic control method for a gradient drive system for ultra-low field magnetic resonance, including: Determining the target switching frequency of the gradient drive system; Determining the corresponding target sampling period and the initial control parameters of the gradient drive system according to the target switching frequency; Determining the target control parameters of the gradient drive system according to the target current and the initial control parameters; Determining the target duty cycle signal of the gradient drive system according to the target control parameters.

[0006] In some embodiments, determining the target switching frequency of the gradient drive system includes: Determine the target operating mode of the ultra-low field magnetic resonance among multiple operating modes, where the operating modes include: standby mode, conventional imaging mode, and fast diffusion weighted imaging mode; Each of the operating modes corresponds to a switching frequency; Determine the corresponding target switching frequency according to the target operating mode.

[0007] In some embodiments, determining the target switching frequency of the gradient drive system includes: Collect the actual output current of the gradient drive system in real time; Determine the current slope according to the output current at the current sampling moment and the output current at the historical sampling moment separated from the current sampling moment by a preset time; Obtain multiple preset current slope thresholds, and each of the preset current slope thresholds corresponds to a switching frequency; Determine the target switching frequency according to the current slope and the multiple preset current slope thresholds.

[0008] In some embodiments, determining the corresponding target sampling period and the initial control parameters of the gradient drive system according to the target switching frequency includes: The target sampling period and the target switching frequency are reciprocals of each other; The target switching frequency has a corresponding initial duty cycle signal of the gradient drive system, and the initial duty cycle signal is determined according to the initial control parameters, where the initial control parameters include an initial integral control parameter, an initial proportional control parameter, and an initial derivative control parameter.

[0009] In some embodiments, the initial duty cycle signal is determined according to the initial control parameters, including The duty cycle signal expression of the gradient drive system is: where represents the duty cycle signal at the k-th sampling moment, is the integral control parameter, is the proportional control parameter, is the derivative control parameter, represents the sampling period, represents the target current, represents the actual output current, represents the current error at the k-th sampling moment, represents the cumulative current error from the 0-th to the k-th moment; Determine the corresponding initial duty cycle signal according to the target switching frequency; The initial duty cycle signal is determined according to the initial integral control parameter, the initial proportional control parameter, and the initial derivative control parameter.

[0010] In some embodiments, determining the target control parameter of the gradient drive system according to the target current and the initial control parameter includes: Updating the initial control parameter to adjust the duty cycle signal and collecting the corresponding updated output current; The target current is a preset value of the gradient drive system; Determining the target control parameter of the gradient drive system according to the current error determined by the updated output current and the target current.

[0011] In some embodiments, determining the target control parameter of the gradient drive system according to the current error determined by the updated output current and the target current includes: Constructing a mean square current error function according to the LMS algorithm; Updating the initial control parameter using the gradient descent method; The control parameter corresponding to the minimum value of the mean square current error function is the target control parameter.

[0012] In some embodiments, the mean square current error function constructed according to the LMS algorithm is: .

[0013] In some embodiments, determining the target control parameter of the gradient drive system according to the target current and the initial control parameter includes: Inputting the initial control parameter into the trained neural network and outputting the target control parameter.

[0014] In some embodiments, the training process of the neural network includes: Taking multiple historical target control parameters as labels, and each initial control parameter corresponds to a historical target control parameter; Inputting multiple initial control parameters into the neural network to be trained, and determining the historical current error corresponding to each initial control parameter as a residual, where the historical current error is determined according to the target current and the updated output current; Accumulating all the residuals to obtain a loss function; Optimizing the neural network according to the loss function until a preset condition is met, and obtaining the optimized neural network.

[0015] The embodiments of the present invention bring the following beneficial effects: Dynamically adjusting the switching frequency of the gradient drive system, thereby reducing the switching loss.

[0016] Other features and advantages of the present disclosure will be set forth in the following description, or may be learned by inference from the description, or may be learned by implementing the above technologies of the present disclosure.

[0017] To make the above objects, features, and advantages of the present disclosure more apparent and understandable, the following specifically describes preferred embodiments in conjunction with the accompanying drawings as follows. Description of the Drawings

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 Flowchart of the automatic control method for the gradient drive system for ultra-low field magnetic resonance provided by an embodiment of the present invention; Figure 2 Flowchart of determining the target switching frequency provided by an embodiment of the present invention; Figure 3 Flowchart of determining the target control parameters of the gradient drive system based on the LMS algorithm provided by an embodiment of the present invention; Detailed Description of the Embodiments To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention fall within the scope of protection of the present invention.

[0020] To facilitate the understanding of this embodiment, first, a magnetic resonance image analysis system disclosed in an embodiment of the present invention will be introduced in detail.

[0021] Figure 1 Flowchart of the automatic control method for the gradient drive system for ultra-low field magnetic resonance provided by an embodiment of the present application.

[0022] First, determine the target switching frequency of the gradient drive system. The specific method can be based on Figure 2The provided method determines the target switching frequency. In an ultra-low field magnetic resonance device, the current collected in the present invention is the core physical quantity for driving the gradient coil to generate a spatially linear magnetic field gradient. The current slope is determined based on the output current at the current sampling moment and the output current at a historical sampling moment that is separated from the current sampling moment by a preset time. Specifically, the historical sampling moment can be the previous moment of the current sampling moment, which can ensure real-time calculation of the current slope and the real-time nature of control; it can also be separated from the current sampling moment by a preset time, such as half a period, N moments, where N is a positive integer greater than 1, etc., which can improve control efficiency and has a relatively high tolerance for the computing power of the hardware. In the present invention, multiple current slope thresholds are preset, and the relationship between the calculated current slope and the current slope thresholds is compared to determine the target switching frequency. Since different current slope thresholds correspond to different operating modes, the target switching frequency can also be determined through the target operating mode of the ultra-low field magnetic resonance. The present invention provides an embodiment of the operating mode, current slope threshold, and target switching frequency, as shown in Table 1: Table 1 When the operating mode is determined, the target switching frequency is determined. The operating mode is determined according to the imaging mode selected by the user, or can also be determined by calculating the current slope.

[0023] The method for determining the corresponding target sampling period and the initial control parameters of the gradient drive system according to the target switching frequency includes: when the target switching frequency is determined, the target sampling period is correspondingly determined, and the target sampling period and the target switching frequency are reciprocals of each other.

[0024] Each target switching frequency (which can also be understood as each operating mode) corresponds to an initial duty cycle signal. The expression of the duty cycle signal is: Among them, represents the duty cycle signal at the k sampling moment, is the integral control parameter, is the proportional control parameter, is the differential control parameter, represents the sampling period, represents the target current, represents the actual output current, represents the current error at the k sampling moment, represents the cumulative current error from 0 to the k moment.

[0025] Therefore, each target switching frequency corresponds to a set of initial control parameters ( , , ), when switching the working mode or detecting that the current slope falls to another current slope threshold, quickly switch to the corresponding initial control parameters. For example, in the low-power mode, relatively small and may be adopted to reduce the response speed and power consumption; in the high-performance mode, relatively large and relatively small may be adopted to improve the tracking accuracy and anti-interference ability. The initial control parameters can be preset in the system according to experience. The target current is the given value of the gradient current of the magnetic resonance system, that is, the current value expected by the system to drive the gradient coil.

[0026] As Figure 3 shown, construct the mean-square current error function according to the LMS algorithm. Exemplarily, the function can be . Use the gradient descent method to update the initial control parameters. When the initial control parameters are updated, the corresponding duty cycle signal is also updated. Synchronously, the output current of the system is also updated synchronously. Collect the updated output current to ensure that the updated output current of the system tracks the target current, that is, gets closer and closer to the target current. When the mean-square current error function is minimized, the determined control parameters are the target control parameters. The LMS algorithm ensures the continuous stability of the closed-loop performance at different switching frequencies by minimizing the mean-square error, so that the actual output current of the system is as close as possible to the expected target current. To this end, it continuously adjusts , , these three parameters. While switching the switching frequency, synchronously switch the initial control parameters to ensure that the switching action does not introduce additional interference. After the switching is completed, continuously update the parameters through LMS, so that the control performance becomes better and better during continuous learning.

[0027] The present invention also provides a method for determining the target control parameters through a neural network. The training process of the neural network includes: using multiple historical target control parameters as labels, and each initial control parameter corresponds to a historical target control parameter; inputting multiple initial control parameters into the neural network to be trained, and determining the historical current error corresponding to each initial control parameter as the residual, where the historical current error is determined according to the target current and the updated output current; accumulating all the residuals to obtain the loss function; optimizing the neural network according to the loss function until the preset conditions (such as minimizing the loss function) are met, and obtaining the optimized neural network. When using this neural network, input the initial control parameters into the trained neural network, and output the target control parameters.

[0028] Unless otherwise specifically stated, the relative steps, numerical expressions and values of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0029] In all of the examples shown and described herein, any specific values should be construed as merely exemplary and not as limitations. Thus, other examples of the exemplary embodiments may have different values.

[0030] Finally, it should be noted that: The above-described embodiments are only specific implementations of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: Any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments or easily conceive of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An automatic control method for a gradient drive system used in ultra-low field magnetic resonance, characterized in that, including: determining a target switching frequency of the gradient drive system; determining a corresponding target sampling period and initial control parameters of the gradient drive system according to the target switching frequency; determining target control parameters of the gradient drive system according to a target current and the initial control parameters; determining a target duty cycle signal of the gradient drive system according to the target control parameters.

2. The method according to claim 1, characterized in that The determining of the target switching frequency of the gradient drive system includes: determining a target operating mode of the ultra-low field magnetic resonance among a plurality of operating modes, where the operating modes include: standby mode, conventional imaging mode, and fast diffusion weighted imaging mode; each of the operating modes corresponds to a switching frequency; determining a corresponding target switching frequency according to the target operating mode.

3. The method according to claim 1, characterized in that The determining of the target switching frequency of the gradient drive system includes: real-time collecting an actual output current of the gradient drive system; determining a current slope according to the output current at the current sampling moment and the output current at a historical sampling moment separated from the current sampling moment by a preset time; acquiring a plurality of preset current slope thresholds, each of the preset current slope thresholds corresponding to a switching frequency; determining a target switching frequency according to the current slope and the plurality of preset current slope thresholds.

4. The method according to claim 1, wherein The determining of the corresponding target sampling period and the initial control parameters of the gradient drive system according to the target switching frequency includes: the target sampling period and the target switching frequency are reciprocals of each other; the target switching frequency has a corresponding initial duty cycle signal of the gradient drive system, and the initial duty cycle signal is determined according to the initial control parameters, and the initial control parameters include an initial integral control parameter, an initial proportional control parameter, and an initial derivative control parameter.

5. The method according to claim 4, wherein The determining of the initial duty cycle signal according to the initial control parameters includes the duty cycle signal expression of the gradient drive system is: Among them, represents the duty cycle signal at the k sampling moment, is the integral control parameter, is the proportional control parameter, is the derivative control parameter, represents the sampling period, represents the target current, represents the actual output current, represents the current error at the k sampling moment, represents the cumulative current error from 0 to the k moment; determining the corresponding initial duty cycle signal according to the target switching frequency; the initial duty cycle signal is determined according to the initial integral control parameter, the initial proportional control parameter, and the initial derivative control parameter.

6. The method according to claim 5, characterized in that, The determining of the target control parameters of the gradient drive system according to the target current and the initial control parameters includes: updating the initial control parameters to adjust the duty cycle signal and collecting a corresponding updated output current; determining the target control parameters of the gradient drive system according to a current error determined by the updated output current and the target current, where the target current is a preset value of the gradient drive system.

7. The method according to claim 6, characterized in that The determining of the target control parameters of the gradient drive system according to the current error determined by the updated output current and the target current includes: constructing a mean square current error function according to the LMS algorithm; updating the initial control parameters using the gradient descent method; when the value of the mean square current error function is the smallest, the corresponding control parameters are the target control parameters.

8. The method according to claim 7, wherein The mean square current error function constructed according to the LMS algorithm is .

9. The method according to claim 6, characterized in that The determining of the target control parameters of the gradient drive system according to the target current and the initial control parameters includes: inputting the initial control parameters into a trained neural network and outputting the target control parameters.

10. The method according to claim 9, characterized in that, The training process of the neural network includes: Using multiple historical target control parameters as labels, each of the initial control parameters corresponds to a historical target control parameter; Inputting the multiple initial control parameters into a neural network to be trained, determining the historical current error corresponding to each initial control parameter as a residual, where the historical current error is determined according to the target current and the updated output current; Accumulating all the residuals to obtain a loss function; Optimizing the neural network according to the loss function until a preset condition is met, to obtain the optimized neural network.

Images