CN118131096B - Magnetic field homogenization treatment method - Google Patents
Magnetic field homogenization treatment method Download PDFInfo
- Publication number
- CN118131096B CN118131096B CN202410372327.5A CN202410372327A CN118131096B CN 118131096 B CN118131096 B CN 118131096B CN 202410372327 A CN202410372327 A CN 202410372327A CN 118131096 B CN118131096 B CN 118131096B
- Authority
- CN
- China
- Prior art keywords
- magnetic field
- local scene
- scene
- objective function
- local
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
- G01R33/28—Details of apparatus provided for in groups G01R33/44 - G01R33/64
- G01R33/38—Systems for generation, homogenisation or stabilisation of the main or gradient magnetic field
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The application discloses a magnetic field homogenization treatment method, which relates to the technical field of magnetic field homogenization treatment, wherein the complexity of a magnetic field environment is better considered through local scene division and in-scene magnetic field representation, and meanwhile, magnetic field adjustment parameters are adopted, so that adjustment is more flexible and universal, different magnetic field environments can be adapted through optimization parameters, the universality and flexibility of the magnetic field homogenization treatment method are improved, more comprehensive and global search is realized based on an artificial intelligence meta-heuristic algorithm, local optimal solution is effectively avoided, and the optimizing efficiency and robustness are improved.
Description
Technical Field
The invention relates to the technical field of magnetic field homogenization treatment, in particular to a magnetic field homogenization treatment method.
Background
The magnetic field homogenization treatment aims at making the magnetic field more uniform in a specific area, and has wide application in various fields at present, including magnetic resonance imaging, magnetic force sensing and magnetic navigation, generally speaking, the distribution of the magnetic field in space may be influenced by various factors, including the shape of an object, the magnetic properties of materials and the interference of an external magnetic field, while in some applications, good uniformity of the magnetic field is required to ensure the accuracy and performance of an instrument, such as magnetic resonance imaging MRI, in which high uniformity of the magnetic field in a certain area inside a human body is required to ensure the accuracy and definition of imaging and magnetic force sensing, in magnetic force sensors, the magnetic field is required to be uniform in a sensing area to obtain accurate magnetic measurement results and magnetic navigation, and in magnetic navigation applications, the magnetic field is required to be uniform in a navigation area to ensure the accuracy of navigation equipment, and it can be seen that the magnetic field homogenization treatment has important significance for ensuring the uniformity of the magnetic field in the specific area.
Halbach array (HalbachArray, halbachpermanentmagnet) is a magnet structure, which was found by american scholars KlausHalbach in 1979 to be a special permanent magnet structure and gradually perfected during electron acceleration experiments, and finally a Halbach magnet is formed, which is an approximately ideal structure in engineering, and the field intensity in the unit direction is enhanced by using a special magnet arrangement, so that the aim is to generate the strongest magnetic field by using the least amount of magnets.
For example, chinese patent discloses a homogenizing device for homogenizing a magnetic field, CN104515962B with a non-magnetic carrier and a balancing element made of a magnetic material, wherein the carrier has a carrier wall and the carrier wall encloses a carrier interior, wherein in the homogenizing device arranged in the magnetic field penetrates into the carrier interior through a first carrier region of the carrier wall and is extruded from the carrier interior through a second carrier region of the carrier wall, and each of the balancing elements arranged at the carrier contributes to the homogenization of the magnetic field at least in the carrier interior. In the homogenization device according to the invention, the handling is improved during homogenization, i.e. because recesses are provided in the holder wall and at least one of the equalization elements can be inserted and removed directly in each of the recesses.
Although the above solution has the advantages as described above, the conventional magnetic field homogenization treatment method adopts an overall adjustment manner, so that it is difficult to cope with the special requirements of different local scenes, which results in that when a complex magnetic field environment is processed, fine adjustment of local magnetic field differences cannot be performed, and flexibility is generally lacking, and once parameter setting is completed, it is difficult to adapt to the changes of different scenes or requirements, so that applicability of the method in complex practical application is limited, and therefore, a more flexible and universal adjustment magnetic field homogenization treatment method is needed to solve such problems.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a magnetic field homogenization treatment method, which solves the problems that the special requirements of different local scenes are difficult to deal with in the whole adjustment mode in the prior art, and the change of different scenes or requirements is difficult to adapt once parameter setting is completed.
In order to achieve the above object, the present invention provides a magnetic field homogenization treatment method, including:
step 1, dividing a local scene and representing a magnetic field in the scene, discretizing a magnetic field region, and discretizing the whole magnetic field region into N sub-regions by defining grids or discrete points in the magnetic field region;
Step 2, defining an objective function, and taking the gradient difference of the magnetic field in the minimized local scene as an objective to minimize the gradient difference of the magnetic field in the local scene;
step 3, introducing magnetic field adjustment parameters, introducing the magnetic field adjustment parameters in the local scene, and adjusting the magnetic field in the local scene in the optimization process;
Step 4, optimizing problem modeling, namely modeling the problem as an optimization problem, namely minimizing a function of an objective function about adjustment parameters, and introducing constraint conditions to ensure that the adjusted local magnetic field meets application requirements and physical rationality;
and 5, applying an optimization algorithm, and adopting a meta-heuristic algorithm based on artificial intelligence to adjust the magnetic field homogenization of the local scene.
The invention is further arranged to: in the step1, the specific method of dividing the local scene is to perform discretization of the magnetic field region, discretize the whole magnetic field region into N sub-regions, namely the local scene, by defining a grid or a discrete point in the magnetic field region, and adopt coordinatesRepresenting the position of the jth discrete point in the ith sub-region, then the sub-regionWhereinRepresenting an i-th sub-region;
The invention is further arranged to: in the step 1, the magnetic field in the scene is expressed specifically in each sub-region In, use functionThe magnetic field is indicated as such,The function represents the local magnetic field, expressed in terms of the components of the vector field: Where r represents the position vector, Representing a magnetic field vector of the magnetic field at space r, the vector comprising three components，Indicating that the magnetic field is in a sub-regionEach component in the inner part;
Representing the whole magnetic field as the sum of the magnetic fields of all local scenes, i.e Here, whereThen the whole magnetic field is indicated;
the invention is further arranged to: in the step 2, the method for minimizing the gradient difference of the magnetic field in the local scene comprises the following steps:
At each local scene In the inner partDefining the gradient of the magnetic field as；
Gradient definition: Wherein Representing a local sceneMagnetic field at r position insideIs used for the dispersion of (a),The change rates of the magnetic field in the x, y and z directions at r are respectively shown;
Gradient differences were then calculated: Here, where Discrete operations representing gradients;
the invention is further arranged to: in the step 2, the method for minimizing gradient difference of the magnetic field in the local scene further includes:
taking the square of the gradient difference as an objective function, targeting the minimization of the gradient difference: ， wherein J represents an objective function for adjusting a magnetic field adjustment parameter J, minimizing gradient difference of magnetic field in local scene by adjusting parameters, thereby realizing magnetic field homogenization,Representing a sceneThe integration is performed and the integration is performed,Integrating volume elements within a local scene with the goal of finding a set of parametersCausing an objective functionMinimizing;
the invention is further arranged to: in the step 3, magnetic field adjustment parameters in the local scene are introduced, and the magnetic field in the local scene is adjusted in the optimization process, specifically:
Introducing local scenes Internal magnetic field adjustment parametersParameters are defined as scalar or vector quantities;
Defining a magnetic field adjusting function to adjust the magnetic field Adjusting parametersAssociating;
， representing the adjusted magnetic field;
The objective function is set For the magnetic field term of (a) the adjusted magnetic fieldInstead, the influence of the adjustment parameters is added,；
By adjusting parametersMinimizing the modified objective function;
the invention is further arranged to: in the step 4, the optimization problem modeling method specifically includes:
Formalize the problem as an optimized problem that can be solved mathematically, with the goal of adjusting Minimizing；
Performing parameter adjustment, including parameters of magnetic field intensity, direction and other relevant attributes;
Determining conditions that the adjusted magnetic field needs to meet, including the magnitude, direction and other physical limitations of the magnetic field;
integrating the information into a mathematical model, wherein the model comprises an objective function needing to be minimized and adjustment parameters meeting constraint conditions;
selecting an optimization algorithm to find a minimum value point in a parameter space;
Solving the problem and finding an optimal parameter combination that minimizes the objective function;
the invention is further arranged to: in the step 5, in the application of the optimization algorithm, a meta heuristic algorithm based on artificial intelligence is adopted to realize the magnetic field homogenization adjustment of the local scene, and the method is specifically as follows:
genetic algorithm, particle swarm optimization and simulated annealing are selected for searching an optimal solution;
Initializing a population in a parameter space by using the selected algorithm, wherein the parameters represent adjustment parameters in the local scene;
Calculating an objective function value, i.e. a measure of the magnetic field homogeneity, for each individual in the population;
in each generation, updating the population by selecting, crossing and mutating operations according to the design principle of the algorithm;
re-evaluating the objective function in the updated population;
The invention is further arranged to: in the step 5, in the application of the optimization algorithm, setting a termination condition, wherein the termination condition comprises the maximum iteration times, the convergence of the objective function, and the algorithm stops once the condition is met;
And extracting the individual with the minimum objective function value from the final population, namely, the adjustment parameter corresponding to the optimal solution.
The invention provides a magnetic field homogenization treatment method. The beneficial effects are as follows:
The magnetic field homogenization processing method provided by the application takes the partial scene division and the in-scene magnetic field as starting points, overcomes the dilemma that the traditional method cannot distinguish partial differences in the whole magnetic field area, achieves higher degree of customization and accurate control by dividing the whole magnetic field into a plurality of partial scenes and more finely adjusting the magnetic field in each scene, adds magnetic field adjustment parameters, adjusts the magnetic field in the partial scenes through the parameters, ensures that the magnetic field adjustment is more flexible, better adapts to the requirements of different scenes and applications on the premise of meeting the application requirements and physical rationality, and ensures that the processing method has stronger adaptability and universality and can be applied in various complex environments.
The existing magnetic rod is magnetized axially or radially, wherein the magnetic rod magnetized axially is an extrusion type magnetic field during assembly, the magnetic rod has obvious advantages and disadvantages, the magnetic field intensity at the extrusion position of the magnetic pole is highest, the magnetic field intensity at the middle position is extremely low, the magnetic field intensity is extremely uneven, and the homogenization control of the magnetic field is realized by the special structure of the magnetic rod and the magnetic characteristics of magnetic materials. The rod is typically composed of a plurality of magnetic materials having different magnetic properties that can produce different magnetic field strengths at different locations. Through reasonable design and manufacturing of the magnetic bars, the magnetic field intensity at different positions tends to be uniform, so that homogenization control of the magnetic field is realized.
Further, a meta-heuristic algorithm based on artificial intelligence is adopted, the parameter space is searched and optimized in a genetic algorithm and particle swarm optimization mode, the global search capability is stronger, a better solution can be found in a complex parameter space, and the meta-heuristic algorithm is selected, so that compared with a traditional local search method, the meta-heuristic algorithm shows higher efficiency and robustness on the aspect of large-scale and high-dimension problems.
In summary, the magnetic field homogenization processing method provided by the application better considers the complexity of the magnetic field environment through the local scene division and the magnetic field representation in the scene, solves the problem of the defect of the traditional method when processing the local difference, adopts the magnetic field adjustment parameters, so that the adjustment is more flexible and universal, can adapt to different magnetic field environments through the optimization parameters, improves the universality and the flexibility of the magnetic field homogenization processing method, realizes more comprehensive and global search based on the meta-heuristic algorithm of artificial intelligence, effectively avoids sinking into the local optimal solution, and improves the optimizing efficiency and the robustness.
The magnetic field homogenization treatment method provided by the application is not only a simple combination of a series of steps, but also a deep innovation of the traditional magnetic field homogenization method, the local magnetic field difference is comprehensively considered, the magnetic field uniformity is more finely and comprehensively adjusted by combining the introduced parameters and the meta-heuristic algorithm, and the magnetic field homogenization treatment method has a wide application prospect in the fields of magnetic resonance imaging, magnetic force sensing and the like.
Drawings
FIG. 1 is a flow chart of a magnetic field homogenizing treatment method of the present invention;
FIG. 2 is a magnetic field diagram of the magnetic field homogenization treatment of the present invention;
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples
Referring to fig. 1-2, the present invention provides a magnetic field homogenizing method, which includes:
step 1, dividing a local scene and representing a magnetic field in the scene, discretizing a magnetic field region, and discretizing the whole magnetic field region into N sub-regions by defining grids or discrete points in the magnetic field region;
the specific method for dividing the local scene comprises the steps of discretizing a magnetic field area, discretizing the whole magnetic field area into N sub-areas, namely the local scene by defining grids or discrete points in the magnetic field area, and adopting coordinates Representing the position of the jth discrete point in the ith sub-region, then the sub-regionWhereinRepresenting an i-th sub-region;
the magnetic field within the scene is represented specifically as, in each sub-region In, use functionThe magnetic field is indicated as such,The function represents the local magnetic field, expressed in terms of the components of the vector field: Where r represents the position vector, Representing a magnetic field vector of the magnetic field at space r, the vector comprising three components，Indicating that the magnetic field is in a sub-regionEach component in the inner part;
Representing the whole magnetic field as the sum of the magnetic fields of all local scenes, i.e Here, whereThen the whole magnetic field is indicated;
the whole magnetic field is overlapped by the local magnetic field, and each local scene contribution is accumulated together;
Step 2, defining an objective function, and taking the gradient difference of the magnetic field in the minimized local scene as an objective to minimize the gradient difference of the magnetic field in the local scene;
In the step 2, the method for minimizing gradient difference of the magnetic field in the local scene comprises the following steps:
At each local scene In the inner partDefining the gradient of the magnetic field as；
Gradient definition: Wherein Representing a local sceneMagnetic field at r position insideIs used for the dispersion of (a),The change rates of the magnetic field in the x, y and z directions at r are respectively shown;
Gradient differences were then calculated: Here, where Discrete operations representing gradients;
in step 2, the method for minimizing gradient differences of the magnetic field in the local scene further comprises:
taking the square of the gradient difference as an objective function, targeting the minimization of the gradient difference: ， wherein J represents an objective function for adjusting a magnetic field adjustment parameter J, minimizing gradient difference of magnetic field in local scene by adjusting parameters, thereby realizing magnetic field homogenization,Representing a sceneThe integration is performed and the integration is performed,Integrating volume elements within a local scene with the goal of finding a set of parametersCausing an objective functionMinimizing;
step 3, introducing magnetic field adjustment parameters, introducing the magnetic field adjustment parameters in the local scene, and adjusting the magnetic field in the local scene in the optimization process;
in step 3, magnetic field adjustment parameters in the local scene are introduced, and the magnetic field in the local scene is adjusted in the optimization process, specifically:
Introducing local scenes Internal magnetic field adjustment parametersParameters are defined as scalar or vector quantities;
Defining a magnetic field adjusting function to adjust the magnetic field Adjusting parametersAssociating;
， representing the adjusted magnetic field;
The objective function is set For the magnetic field term of (a) the adjusted magnetic fieldInstead, the influence of the adjustment parameters is added,；
By adjusting parametersMinimizing the modified objective function;
Adjusting parameters The introduction of the magnetic field in the local scene is influenced by the magnetic field adjusting function, the correction of the objective function ensures that the influence of the adjusting parameters is considered in the optimization process, so that the uniformity of the magnetic field in the local scene is reflected more accurately, and in the optimization problem, the objective function is minimized by the adjusting parameters, so that the accurate adjustment of the magnetic field in the local scene is realized;
Step 4, optimizing problem modeling, namely modeling the problem as an optimization problem, namely minimizing a function of an objective function about adjustment parameters, and introducing constraint conditions to ensure that the adjusted local magnetic field meets application requirements and physical rationality;
In step 4, the optimization problem modeling method specifically includes:
Formalize the problem as an optimized problem that can be solved mathematically, with the goal of adjusting Minimizing；
Performing parameter adjustment, including parameters of magnetic field intensity, direction and other relevant attributes;
Determining conditions that the adjusted magnetic field needs to meet, including the magnitude, direction and other physical limitations of the magnetic field;
integrating the information into a mathematical model, wherein the model comprises an objective function needing to be minimized and adjustment parameters meeting constraint conditions;
selecting an optimization algorithm to find a minimum value point in a parameter space;
Solving the problem and finding an optimal parameter combination that minimizes the objective function;
through an optimization problem modeling step, the problem is converted into a form which can be solved by a mathematical method, so that the magnetic field in a local scene can be systematically adjusted to be more uniform, and the optimization is performed on the premise of meeting constraint conditions;
Step 5, optimizing algorithm application, namely homogenizing and adjusting the magnetic field of the local scene by adopting meta-heuristic algorithm based on artificial intelligence;
in step 5, in the application of the optimization algorithm, a meta-heuristic algorithm based on artificial intelligence is adopted to realize the magnetic field homogenization adjustment of the local scene, and the method is specifically as follows:
genetic algorithm, particle swarm optimization and simulated annealing are selected for searching an optimal solution;
Initializing a population in a parameter space by using the selected algorithm, wherein the parameters represent adjustment parameters in the local scene;
Calculating an objective function value, i.e. a measure of the magnetic field homogeneity, for each individual in the population;
in each generation, updating the population by selecting, crossing and mutating operations according to the design principle of the algorithm;
re-evaluating the objective function in the updated population;
In step 5, in the application of the optimization algorithm, setting a termination condition, wherein the termination condition comprises the maximum iteration times, the convergence of the objective function, and stopping the algorithm once the condition is met;
Extracting an individual with the minimum objective function value from the final population, namely, adjusting parameters corresponding to the optimal solution;
by adopting a meta-heuristic algorithm, an optimal solution of the local magnetic field uniformity is found in a high-dimensional parameter space.
In the present application, the above is combined with the above matters:
The magnetic field homogenization processing method provided by the application takes the partial scene division and the in-scene magnetic field as starting points, overcomes the dilemma that the traditional method cannot distinguish partial differences in the whole magnetic field area, achieves higher degree of customization and accurate control by dividing the whole magnetic field into a plurality of partial scenes and more finely adjusting the magnetic field in each scene, adds magnetic field adjustment parameters, adjusts the magnetic field in the partial scenes through the parameters, ensures that the magnetic field adjustment is more flexible, better adapts to the requirements of different scenes and applications on the premise of meeting the application requirements and physical rationality, and ensures that the processing method has stronger adaptability and universality and can be applied in various complex environments.
Based on the graph of fig. 2, the existing magnetic rod is magnetized axially or radially, wherein the magnetic rod magnetized axially is an extrusion type magnetic field during assembly, the magnetic rod has obvious advantages and disadvantages, the magnetic field strength at the extrusion position of the magnetic pole is highest, the magnetic field strength at the middle position is extremely low, the magnetic field strength is extremely uneven, and the homogenization control of the magnetic field is realized by the special structure of the magnetic rod and the magnetic characteristics of magnetic materials. The rod is typically composed of a plurality of magnetic materials having different magnetic properties that can produce different magnetic field strengths at different locations. Through reasonable design and manufacturing of the magnetic bars, the magnetic field intensity at different positions tends to be uniform, so that homogenization control of the magnetic field is realized.
Further, a meta-heuristic algorithm based on artificial intelligence is adopted, the parameter space is searched and optimized in a genetic algorithm and particle swarm optimization mode, the global search capability is stronger, a better solution can be found in a complex parameter space, and the meta-heuristic algorithm is selected, so that compared with a traditional local search method, the meta-heuristic algorithm shows higher efficiency and robustness on the aspect of large-scale and high-dimension problems.
In summary, the magnetic field homogenization processing method provided by the application better considers the complexity of the magnetic field environment through the local scene division and the magnetic field representation in the scene, solves the problem of the defect of the traditional method when processing the local difference, adopts the magnetic field adjustment parameters, so that the adjustment is more flexible and universal, can adapt to different magnetic field environments through the optimization parameters, improves the universality and the flexibility of the magnetic field homogenization processing method, realizes more comprehensive and global search based on the meta-heuristic algorithm of artificial intelligence, effectively avoids sinking into the local optimal solution, and improves the optimizing efficiency and the robustness.
The magnetic field homogenization treatment method provided by the application is not only a simple combination of a series of steps, but also a deep innovation of the traditional magnetic field homogenization method, the local magnetic field difference is comprehensively considered, the magnetic field uniformity is more finely and comprehensively adjusted by combining the introduced parameters and the meta-heuristic algorithm, and the magnetic field homogenization treatment method has a wide application prospect in the fields of magnetic resonance imaging, magnetic force sensing and the like.
It is to be understood that the above examples of the present invention are provided by way of illustration only and not by way of limitation of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
Claims (4)
1. A magnetic field homogenization treatment method, characterized by comprising:
step 1, dividing a local scene and representing a magnetic field in the scene, discretizing a magnetic field region, and discretizing the whole magnetic field region into N sub-regions by defining grids or discrete points in the magnetic field region;
Step 2, defining an objective function, and taking the gradient difference of the magnetic field in the minimized local scene as an objective to minimize the gradient difference of the magnetic field in the local scene;
step 3, introducing magnetic field adjustment parameters, introducing the magnetic field adjustment parameters in the local scene, and adjusting the magnetic field in the local scene in the optimization process;
Step 4, optimizing problem modeling, namely modeling the problem as an optimization problem, namely minimizing a function of an objective function about adjustment parameters, and introducing constraint conditions to ensure that the adjusted local magnetic field meets application requirements and physical rationality;
Step 5, optimizing algorithm application, namely homogenizing and adjusting the magnetic field of the local scene by adopting meta-heuristic algorithm based on artificial intelligence;
The specific method for dividing the local scene in the step 1 is that discretizing a magnetic field area is carried out, the whole magnetic field area is discretized into N sub-areas by defining grids or discrete points in the magnetic field area, namely, the local scene adopts a coordinate r ij to represent the position of the jth discrete point in the ith sub-area, and then the sub-area S i＝{ri1,ri2...,rij is divided, wherein S i represents the ith sub-area;
The magnetic field in the scene is specifically represented by using the component of the vector field in each sub-region S i: b i(r)＝(Bix(r),Biy(r),Biz (r)), where r represents the position vector, B i (r) represents the magnetic field vector of the magnetic field at space r, which vector contains three components B ix(r),Biy(r),Biz(r),Bix(r),Biy(r),Biz (r) representing the individual components of the magnetic field within the sub-region S i; representing the whole magnetic field as the sum of the magnetic fields of all local scenes, i.e Where B (r) then represents the entire magnetic field;
the method for minimizing gradient difference of the magnetic field in the local scene in the step 2 comprises the following steps: b i (r) within each local scene S i defines the gradient of the magnetic field as
Gradient definition: Wherein the method comprises the steps of Representing the divergence of magnetic field B i at the r-position within local scene S i,The change rates of the magnetic field in the x, y and z directions at r are respectively shown;
Gradient differences were then calculated: Here, the Discrete operations representing gradients;
The method for minimizing gradient differences of magnetic fields in a local scene further comprises:
taking the square of the gradient difference as an objective function, targeting the minimization of the gradient difference: J (C i) where J represents an objective function for adjusting the magnetic field adjustment parameters C i, J minimizes the gradient difference of the magnetic field within the local scene by adjusting the parameters to achieve magnetic field homogenization, +.s i represents integration of scene S i, dV i integrates volume elements within the local scene, and the goal is to find a set of parameters C i to minimize the objective function J (C i);
In the step 3, magnetic field adjustment parameters in the local scene are introduced, and the magnetic field in the local scene is adjusted in the optimization process, specifically:
A magnetic field adjustment parameter C i introduced into the local scene S i, the parameter being defined as a scalar or vector;
Defining a magnetic field adjustment function, and associating a magnetic field B i (r) with an adjustment parameter C i;
B' i(r,Ci)←(Bi(r),Ci),B'i(r,Ci) represents the adjusted magnetic field;
The magnetic field term in the objective function J (C i) is replaced by the adjusted magnetic field B' i(r,Ci), the influence of the adjustment parameters is added,
The modified objective function is minimized by adjusting parameter C i.
2. The method for homogenizing a magnetic field according to claim 1, wherein in the step 4, the optimization problem modeling method specifically comprises:
Formalize the problem as an optimization problem that can be solved mathematically, with the goal of minimizing J (C i) by adjusting C i;
Performing parameter adjustment, including parameters of magnetic field intensity, direction and other relevant attributes;
Determining conditions that the adjusted magnetic field needs to meet, including the magnitude, direction and other physical limitations of the magnetic field;
integrating the information into a mathematical model, wherein the model comprises an objective function needing to be minimized and adjustment parameters meeting constraint conditions;
selecting an optimization algorithm to find a minimum value point in a parameter space;
Solving the problem and finding the optimal combination of parameters that minimizes the objective function.
3. The method for homogenizing the magnetic field according to claim 2, wherein in the step 5, in the application of the optimization algorithm, a meta-heuristic algorithm based on artificial intelligence is adopted to realize the homogenization adjustment of the magnetic field of the local scene, and the method is specifically as follows:
genetic algorithm, particle swarm optimization and simulated annealing are selected for searching an optimal solution;
Initializing a population in a parameter space by using the selected algorithm, wherein the parameters represent adjustment parameters in the local scene;
Calculating an objective function value, i.e. a measure of the magnetic field homogeneity, for each individual in the population;
in each generation, updating the population by selecting, crossing and mutating operations according to the design principle of the algorithm;
the objective function is re-evaluated in the updated population.
4. A magnetic field homogenizing treatment method according to claim 3, wherein in step 5, in optimizing algorithm application, a termination condition is set, including reaching a maximum number of iterations, objective function convergence, and once the condition is satisfied, the algorithm is stopped;
And extracting the individual with the minimum objective function value from the final population, namely, the adjustment parameter corresponding to the optimal solution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410372327.5A CN118131096B (en) | 2024-03-29 | 2024-03-29 | Magnetic field homogenization treatment method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410372327.5A CN118131096B (en) | 2024-03-29 | 2024-03-29 | Magnetic field homogenization treatment method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118131096A CN118131096A (en) | 2024-06-04 |
| CN118131096B true CN118131096B (en) | 2024-07-26 |
Family
ID=91235423
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410372327.5A Active CN118131096B (en) | 2024-03-29 | 2024-03-29 | Magnetic field homogenization treatment method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118131096B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101533077A (en) * | 2009-04-17 | 2009-09-16 | 中国科学院电工研究所 | Optimal design method of superconducting magnet used for magnetic resonance imaging (MRI) device |
| CN104515962A (en) * | 2013-08-20 | 2015-04-15 | 克洛纳有限公司 | Homogenization device for homogenization of a magnetic field |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102011089445B4 (en) * | 2011-12-21 | 2015-11-05 | Siemens Aktiengesellschaft | Method and gradient system for reducing mechanical vibrations in a magnetic resonance imaging system |
| CN110361683B (en) * | 2019-07-17 | 2021-06-18 | 哈尔滨工程大学 | Magnetometer calibration method based on particle swarm optimization with dual objective functions |
-
2024
- 2024-03-29 CN CN202410372327.5A patent/CN118131096B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101533077A (en) * | 2009-04-17 | 2009-09-16 | 中国科学院电工研究所 | Optimal design method of superconducting magnet used for magnetic resonance imaging (MRI) device |
| CN104515962A (en) * | 2013-08-20 | 2015-04-15 | 克洛纳有限公司 | Homogenization device for homogenization of a magnetic field |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118131096A (en) | 2024-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Molokovsky et al. | Intense electron and ion beams | |
| Alotto et al. | Stochastic algorithms in electromagnetic optimization | |
| CN102150222B (en) | Magnetic assembly and method for defining a magnetic field for an imaging volume | |
| CN106556813B (en) | Linear mixed optimization method of active shimming coil in magnetic resonance system | |
| CN109765510B (en) | Method for designing radial superconducting shimming coil with fillet | |
| CN101533077A (en) | Optimal design method of superconducting magnet used for magnetic resonance imaging (MRI) device | |
| CN108416107B (en) | A Finite Element Algorithm for Promoting Particle Motion Applied in PIC | |
| CN110278651B (en) | Reinforced learning method for calibrating beam offset of accelerator | |
| Gal’chenko et al. | Pareto-optimal parametric synthesis of axisymmetric magnetic systems with allowance for nonlinear properties of the ferromagnet | |
| CN118131096B (en) | Magnetic field homogenization treatment method | |
| CN113971349B (en) | A method for obtaining the number and initial position of coils of a solenoid-shaped superconducting magnet | |
| US6377148B1 (en) | Asymmetric zonal shim coils for magnetic resonance | |
| CN114896729B (en) | Method and system for designing magnet by using induced current | |
| CN108460188B (en) | Charge distribution finite element FEM solving algorithm applied to PIC electrostatic model | |
| Ziolkowski et al. | Genetic algorithm coupled with Bézier curves applied to the magnetic field on a solenoid axis synthesis | |
| Cingoski et al. | Inverse shape optimization using dynamically adjustable genetic algorithms [electric machine design] | |
| CN117592346A (en) | High-power microwave source device optimization design method and system | |
| CN112231880A (en) | A kind of superconducting magnet coil design method | |
| CN116933676A (en) | Flow numerical simulation method, device, equipment and medium | |
| Bouchekara | Electromagnetic device optimization based on electromagnetism-like mechanism | |
| CN108446429B (en) | A Finite Element Solution Algorithm for Particle Force Applied to PIC Electrostatic Model | |
| CN119347817B (en) | A method for optimizing the concentration of magnetic particles in a conical magnetic continuous soft robot | |
| Ma et al. | The arithmetic optimization algorithm based-on particle energy and particle diffusion | |
| Wu et al. | Planar coil optimization for the Eos stellarator using sparse regression | |
| CN118446077B (en) | An optimization method for a low-power three-axis Helmholtz coil model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |