APF-DTL beam dynamics optimization method and device, storage medium and terminal

Abstract

The present application provides a beam dynamics optimization method and device, storage medium and terminal of APF-DTL, the method comprising: obtaining the initial phase sequence of APF-DTL; randomly generating multiple groups of phase sequences within a set generation range based on the initial phase sequence to form a phase particle group; wherein each phase particle corresponds to a group of phase sequences; determining multiple optimization indicators of the beam dynamics of the APF-DTL to determine a normalized fitness function based on the multiple optimization indicators and constraints; searching for the optimal phase sequence in the iterative process of the phase particle group based on the normalized fitness function. The present application converts the multi-objective optimization of beam dynamics into a single-objective optimization, which can further narrow the search space of the optimal phase sequence, simplify the search process, reduce the search cost, and effectively optimize the beam dynamics of APF-DTL.

Description

Technical Field

[0001] The present application belongs to the field of dynamics technology, and in particular, relates to an APF-DTL beam dynamics optimization method and device, a storage medium and a terminal. Background Art

[0002] The alternating phase focusing drift tube linear accelerator (APF-DTL) is an accelerating structure that uses the DTL gap electric field to achieve beam transverse focusing and longitudinal bunching. In this type of accelerator, the beam is accelerated and focused in the gap electric field between a series of drift tubes, that is, the gap electric field between each drift tube and the next drift tube is the key to particle acceleration, so APF-DTL can accelerate the beam without a transverse focusing magnet.

[0003] Specifically, when APF-DTL is working, the synchronization phase of the gap is between -90° and +90°. When the synchronization phase is negative, the gap electric field has a longitudinal bunching effect on the beam and a transverse defocusing effect; when the synchronization phase is positive, the gap electric field has a transverse focusing effect on the beam and a longitudinal defocusing effect. Therefore, the phase arrangement of APF-DTL will have a great influence on the transmission effect of the beam. The appropriate positive and negative alternating phase sequence can effectively reduce the growth of the beam emittance and control the transverse and longitudinal envelopes, thereby effectively realizing the transverse focusing and longitudinal bunching of the beam, thereby improving the transmission efficiency of DTL. Therefore, how to effectively find the optimal phase sequence of APF-DTL to achieve the optimization of beam flow dynamics is a technical problem that needs to be solved urgently by technicians in this field. Summary of the invention

[0004] The purpose of the present application is to provide a method and device, storage medium and terminal for optimizing beam dynamics of APF-DTL, so as to solve the technical problem that it is difficult to effectively find the optimal phase sequence of APF-DTL to optimize beam dynamics in the prior art.

[0005] In a first aspect, the present application provides a method for optimizing beam dynamics of an APF-DTL, the method comprising:

[0006] Get the initial phase sequence of APF-DTL;

[0007] Based on the initial phase sequence, a plurality of phase sequences are randomly generated within a set generation range to form a phase particle group; wherein each phase particle corresponds to a phase sequence;

[0008] Determining a plurality of optimization indicators of the beam dynamics of the APF-DTL to determine a normalized fitness function based on the plurality of optimization indicators and constraints;

[0009] An optimal phase sequence is searched in the iterative process of the phase particle swarm based on the normalized fitness function.

[0010] In one embodiment of the present application, obtaining an initial phase sequence of APF-DTL includes:

[0011] Get multiple free parameters;

[0012] A plurality of initial phases are calculated based on the plurality of free parameters to obtain the initial phase sequence.

[0013] In one embodiment of the present application, determining a normalized fitness function based on the multiple optimization indicators and constraints includes:

[0014] Acquire corresponding optimization index membership based on the multiple optimization indexes;

[0015] Performing weighted calculation on the membership of multiple optimization indicators to obtain a basic fitness function;

[0016] The normalized fitness function is obtained based on the basic fitness function and constraints.

[0017] In one embodiment of the present application, the multiple optimization indicators include: the transmission efficiency indicator of the APF-DTL, the lateral envelope indicator of the beam, the x-x' phase space emittance growth indicator of the beam, the y-y' phase space emittance growth indicator of the beam, the PW phase space emittance growth indicator of the beam, and the export energy indicator.

[0018] In one embodiment of the present application, obtaining corresponding optimization index membership based on the multiple optimization indexes includes:

[0019] Obtaining a mapping parameter corresponding to each optimization indicator among the multiple optimization indicators;

[0020] The optimization indicator membership corresponding to each optimization indicator is obtained based on the mapping parameters.

[0021] In one embodiment of the present application, a plurality of the optimization index memberships are weightedly calculated to obtain a basic fitness function:

[0022] Constructing a judgment matrix, wherein the judgment matrix is ​​used to judge the importance of the multiple optimization indicators;

[0023] Performing a consistency check on the judgment matrix, and when the judgment matrix meets the consistency requirement, obtaining an eigenvector corresponding to the maximum eigenvalue of the judgment matrix;

[0024] Obtaining weights of a plurality of the optimization indicators based on the feature vector;

[0025] The memberships of the plurality of optimization indicators are weightedly calculated based on the weights of the plurality of optimization indicators to obtain a basic fitness function.

[0026] In one embodiment of the present application, the constraint condition includes: a constraint value of a phase difference between adjacent phases in the APF-DTL; wherein the constraint condition is restricted by a penalty function.

[0027] In a second aspect, the present application further provides a device for optimizing beam dynamics, the device being applied to an alternating phase focusing drift tube linear accelerator, the device comprising:

[0028] An acquisition module, used for acquiring an initial phase sequence of APF-DTL;

[0029] A generation module, used to randomly generate multiple groups of phase sequences within a set generation range based on the initial phase sequence to form a phase particle group; each phase sequence corresponds to a phase particle;

[0030] An indicator module, used to determine a plurality of optimization indicators of the beam dynamics of the APF-DTL, so as to determine a normalized fitness function based on the plurality of optimization indicators and constraint conditions;

[0031] A search module is used to search for an optimal phase sequence in the iterative process of the phase particle swarm based on a normalized fitness function.

[0032] In a third aspect, the present application also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the beam dynamics optimization method of the APF-DTL as described above.

[0033] In a fourth aspect, the present application further provides a terminal, comprising a processor and a memory, wherein the memory is communicatively connected to the processor;

[0034] The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the beam dynamics optimization method of the APF-DTL as described above.

[0035] As described above, the APF-DTL beam dynamics optimization method and device, storage medium and terminal described in this application have the following beneficial effects:

[0036] This application introduces the particle swarm optimization algorithm into the APF-DTL beam dynamics optimization scenario, normalizes multiple optimization indicators of the beam into a fitness function, transforms the multi-objective optimization problem into a single-objective optimization problem, and simplifies the optimization process. At the same time, this application introduces a penalty function to restrict the constraints, transforms the constrained optimization into the unconstrained optimization, and finally proposes a single-objective optimization algorithm suitable for APF-DTL dynamics optimization, which can further narrow the search space, simplify the search process, and reduce the search cost, so as to effectively find the optimal phase sequence to achieve the beam dynamics optimization of APF-DTL. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic flow chart of the APF-DTL beam dynamics optimization method described in an embodiment of the present application is shown.

[0038] Figure 2 A schematic diagram of the initial phase sequence of the APF-DTL to be optimized according to an embodiment of the present application is shown.

[0039] Figure 3 A schematic flow chart of the APF-DTL beam dynamics optimization method described in an embodiment of the present application is shown.

[0040] Figure 4 A schematic flow chart of the APF-DTL beam dynamics optimization method described in an embodiment of the present application is shown.

[0041] Figure 5 The inlet phase diagram of the APF-DTL to be optimized described in the embodiment of the present application is shown.

[0042] Figure 6 A schematic diagram showing changes in the maximum value of the normalized fitness function during the iteration process described in an embodiment of the present application is shown.

[0043] Figure 7 A schematic diagram of the optimal phase sequence of the APF-DTL described in an embodiment of the present application is shown.

[0044] Figure 8 A schematic diagram of the drift tube length after APF-DTL optimization described in an embodiment of the present application is shown.

[0045] Fig. 9 The optimized outlet phase diagram of the APF-DTL described in the embodiment of the present application is shown.

[0046] Fig.10 A schematic diagram showing the growth of the root mean square emittance after APF-DTL optimization described in an embodiment of the present application is shown.

[0047] Fig.11A schematic diagram of the lateral 3 times root mean square envelope after APF-DTL optimization described in an embodiment of the present application is shown.

[0048] Figures 12a to 12d A schematic diagram showing the transverse and longitudinal beam density, phase density and energy density after APF-DTL optimization described in an embodiment of the present application is shown.

[0049] Fig.13 A schematic structural diagram of the APF-DTL beam dynamics optimization device described in an embodiment of the present application is shown.

[0050] Fig.14 A schematic diagram of the structure of the terminal described in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0051] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0052] The principles and implementation methods of the APF-DTL beam dynamics optimization method and device, storage medium and terminal of this embodiment will be explained in detail below, so that those skilled in the art can understand the beam dynamics optimization method, device, storage medium and terminal of this embodiment without creative work.

[0053] The traditional APF-DTL dynamics optimization adopts the method of manually optimizing the phase sequence. This method uses the dynamics calculation software to manually change the phase arrangement of the APF-DTL so that the dynamic results meet the requirements. However, the process of manually optimizing the phase sequence is extremely time-consuming, and the optimization range is limited, and it cannot guarantee the global optimal solution. In addition, when the number of APF-DTL gaps is too large, the workload of manually optimizing the phase sequence will also be extremely large, and the beam dynamics cannot be effectively optimized. In order to simplify the phase arrangement, Professor Iwata proposed a formula to calculate the phase arrangement, called the Iwata formula:

[0054] (1)

[0055] Where n is the number of gaps, , a, b, c and is a free parameter. By adjusting the five free parameters, the synchronization phase of n gaps can be generated. Formula (1) uses, , a, b, c and Any number of phase sequences can be obtained by constant calculation, thus simplifying the workload of phase sequence optimization in the case of multiple gaps. However, this method still has the problem that optimizing the phase sequence is time-consuming and labor-intensive, and cannot guarantee global search. In addition, the phase sequence obtained by this method is relatively single, which limits the exploration of other phase arrangements during the optimization process.

[0056] In addition, a multi-objective optimization method is also proposed in the prior art, which takes the export energy and transmission efficiency of APF-DTL as the optimization targets and searches for parameters that meet the requirements in the Pareto optimality. However, this method has complex optimization targets. When there are more than two targets, the Pareto front solution set contains many optimization results. How to find the optimal solution in the Pareto front solution set requires manpower. Moreover, this method does not include restrictions on constraints, resulting in the optimal solution found may not meet the actual situation of APF-DTL.

[0057] In order to solve the above-mentioned technical problems existing in the prior art, an embodiment of the present application provides a beam dynamics optimization method of APF-DTL. The application of the method can effectively search for the optimal phase sequence of APF-DTL, thereby effectively optimizing the beam dynamics of APF-DTL.

[0058] Figure 1 A schematic diagram of the process flow of the APF-DTL beam dynamics optimization method described in the embodiment of the present application is shown, as Figure 1 As shown, the APF-DTL beam dynamics optimization method provided in the embodiment of the present application includes steps S1 to S4.

[0059] S1. Obtain the initial phase sequence of APF-DTL.

[0060] Specifically, obtaining the initial phase sequence of the APF-DTL includes: obtaining a plurality of free parameters; and calculating a plurality of initial phases based on the plurality of free parameters to obtain the initial phase sequence.

[0061] In some embodiments, multiple initial phases can be calculated based on multiple free parameters using equation (1). The multiple free parameters are obtained as follows: =-70, a =0.0001, b =1.7, c =0.0072, =10, then the initial phase sequence obtained according to formula (1) is as follows Figure 2 shown.

[0062] S2. Based on the initial phase sequence, a plurality of phase sequences are randomly generated within a set generation range to form a phase particle group; wherein each phase particle corresponds to a phase sequence.

[0063] Specifically, based on the initial phase sequence, each initial phase in the initial phase sequence is randomly generated within a certain generation range to form multiple phase sequences. Each phase sequence can be regarded as a phase particle, and multiple phase sequences are represented by multiple phase particles, so that a phase particle group can be obtained.

[0064] In some embodiments, each initial phase in the initial phase sequence can be made to randomly generate other phases within a generation range of ±20°. The generation range of ±20° ensures that the phase sequence (random phase particles) generated thereby is near the sinusoidal distribution and is not too dispersed, which can limit the search range and ensure a sufficient range.

[0065] Furthermore, the randomly generated phase should not exceed the boundary range of the phase in APF-DTL, that is, not exceed ±90°.

[0066] It should be noted that the number of particles can be set by oneself. That is, the number of groups of phase sequences randomly generated based on the initial phase sequence is determined manually, and the present application is not limited thereto.

[0067] Furthermore, after obtaining the phase particle swarm, the optimal phase sequence that meets the optimization goal can be iterated based on the particle swarm optimization algorithm. For the particle swarm optimization algorithm, during the iteration process of the phase particle swarm, the phase particles update their positions and velocities by tracking two “extreme values”. The first is the optimal solution found by the phase particle itself, which is called the individual optimal P best , and the other is the optimal solution found by the entire population, called the global optimal G best By continuously updating the speed and position of the phase particles, the optimal solution that meets the optimization goal is found.

[0068] Among them, in this embodiment, the individual optimal P best It refers to the best historical position of the same phase particle (the best solution of a set of individual phase sequences), the global optimal G best It refers to the optimal solution of the phase particle group (the optimal solution of multiple phase sequences, i.e. the optimal phase sequence). The speed of the phase particle is the change of the phase, and the position of the phase particle is the position of the phase.

[0069] Specifically, the velocity of the phase particle is updated by the following formula (2):

[0070] (2)

[0071] Among them, ω represents the inertia weight, r1 and r2 are acceleration constants, which represent the particle's self-cognition learning factor and social learning factor respectively; rand(0,1) is a random number on [0,1]; P t and G t Represents the local optimal (its own historical optimal) and global optimal positions of the particle.

[0072] In the three parts of formula (2), the first part actually represents the inertia of the particle, that is, the tendency to move according to the speed of its previous generation. If the inertia weight is large, the speed change between two adjacent generations of particles is small; the second part represents the reflection of the particle, that is, self-adjustment according to the historical optimal position; the third part represents the sociality of the particle, that is, adjusting its own speed according to the global optimal solution.

[0073] In the iterative process of the phase particle swarm, the particle inertia coefficient, self-learning factor, social factor and maximum speed limit will affect the particle swarm's global and local search capabilities. The maximum speed affects the search resolution. If it is too large, the program will easily skip the optimal solution directly. If it is too small, it will easily fall into the local optimum. The relevant parameters ω, r1 and r2 will affect the program's search ability. Therefore, in this embodiment, r1=0.729, r2=1.499445 are taken to ensure sufficient convergence ability and sufficient search range. In addition, the inertia weight coefficient ω It changes with the number of iterations. The inertia weight coefficient is small in the early stage, which can expand the search area of ​​the particles. The weight coefficient is large in the later stage, which speeds up the convergence of the optimal solution.

[0074] Specifically, the position of the phase particle is updated by the following formula (3):

[0075] (3)

[0076] The iterative process of the phase particle swarm in the above embodiment will be further described in detail below.

[0077] When obtaining the initial phase sequence (such as Figure 2As shown in the figure, a random number (rand(0,1) in formula (2)) is used to generate a phase within the generation range of ±20° of each initial phase. If the generated phase exceeds the phase boundary range, the phase is allowed to stay on the boundary, thereby generating a total of 20 sets of phase sequences, i.e., 20 phase particles, to form a phase particle group. Afterwards, the 20 phase particles are calculated simultaneously based on formulas (2) and (3) to expand the search range and shorten the convergence time. Among them, the initial velocity of the phase particle is the change in the phase in the first iteration. According to experience, the initial velocity can be set to +4, and the velocity of the phase particle in subsequent iterations is calculated by formula (2). The phase particles are continuously iterated through formulas (2) and (3) to approach the optimal solution.

[0078] It should be noted that the specific number of iterations can be set by yourself.

[0079] S3. Determine a plurality of optimization indices of the beam dynamics of the APF-DTL, and determine a normalized fitness function based on the plurality of optimization indices and constraint conditions.

[0080] In order to effectively improve the quality performance of APF-DTL beam dynamics in all aspects, multiple optimization indicators can be selected to comprehensively optimize the beam dynamics of APF-DTL.

[0081] In some embodiments, the multiple optimization indicators include: the transmission efficiency indicator of the APF-DTL, the lateral envelope indicator of the beam, the x-x' phase space emittance growth indicator of the beam, the y-y' phase space emittance growth indicator of the beam, the PW phase space emittance growth indicator of the beam, and the export energy indicator.

[0082] The transmission efficiency index of the alternating phase focusing drift tube linear accelerator refers to: the difference between the ideal transmission efficiency and the actual transmission efficiency of the alternating phase focusing drift tube linear accelerator (100% minus the actual transmission efficiency).

[0083] The transverse envelope index of the beam refers to the difference between the maximum transverse envelope of the beam and the ideal envelope. It should be noted that the transverse envelope of the beam includes the envelope in the x-direction and the envelope in the y-direction.

[0084] Among them, the export energy index refers to the absolute value of the difference between the actual export energy and the designed export energy.

[0085] It should be noted that, in other embodiments, the multiple optimization indicators may also include: the transmission efficiency indicator of the APF-DTL, the lateral envelope indicator of the beam, the x-x' phase space emittance growth indicator of the beam, the y-y' phase space emittance growth indicator of the beam, and the PW phase space emittance growth indicator of the beam. That is, the optimization indicators can actually be set according to actual optimization needs.

[0086] In the above-mentioned iteration process of the phase particle group, it is necessary to calculate the fitness of each phase particle to determine whether it is the optimal solution at this time. That is, in the phase optimization problem of APF-DTL, the fitness function is an indicator for evaluating the quality of the current phase sequence beam. The larger the fitness, the closer the phase particle is to the optimal solution. At this time, the phase sequence can bring higher beam quality and better optimization effect of beam dynamics. Specifically, in each iteration process, the historical best position (P best ). If in this iteration, the fitness of the phase particle is better than P best , then update P best At the same time, the optimal solution of the entire phase particle swarm (G best ), if the fitness of a phase particle is better than the current G best , then update G best In this way, when the phase particle swarm iteration is completed, G best is the optimal solution, that is, the optimal phase sequence. At this time, when APF-DTL adopts this optimal phase sequence, it can bring better beam quality.

[0087] To this end, it is necessary to normalize the multiple optimization indicators determined above into a fitness function, so that the multiple optimization indicators of the APF-DTL beam can be converted into a single fitness parameter evaluation. That is, when the fitness of the phase particle calculated by the normalized fitness function is higher, it means that the phase particle performs well in multiple optimization indicators of the beam. In this way, the performance of APF-DTL in various aspects of beam quality can be comprehensively evaluated, and the key optimization of a certain beam indicator can be achieved by adjusting the corresponding weight of each indicator.

[0088] Therefore, after determining multiple optimization indicators, the embodiment of the present application determines a normalized fitness function based on the multiple optimization indicators and constraint conditions. Figure 3 As shown, in the embodiment of the present application, determining the normalized fitness function based on the multiple optimization indicators and constraints includes steps S31 to S33.

[0089] S31. Obtain corresponding optimization indicator membership based on the multiple optimization indicators.

[0090] Specifically, the normalized fitness function combines multiple optimization indicators into a fitness parameter for evaluating beam quality. At this time, since the dimensions and values ​​of each optimization indicator are different, it is necessary to first perform dimensionless processing on each optimization indicator, that is, The value of is mapped to [0,1]. The mapped value is the corresponding optimization index membership.

[0091] In some embodiments, a mapping relationship between each optimization indicator and the corresponding membership can be established through a Cauchy distribution membership function, and the mapping function of the Cauchy distribution is expressed as:

[0092] (4)

[0093] in, and They represent the mapping parameters of the optimization index X[k], and F(X[k]) is the optimization index membership. At this time, the undetermined coefficient method can be used to determine and These two mapping parameters establish a mapping relationship, thereby non-dimensionalizing the indicators of different dimensions and mapping them to [0,1], unifying the evaluation criteria of different dimensions. That is, obtaining the corresponding optimization indicator membership based on the multiple optimization indicators includes: obtaining the mapping parameters corresponding to each optimization indicator in the multiple optimization indicators; and obtaining the optimization indicator membership corresponding to each optimization indicator based on the mapping parameters.

[0094] The following will take the transmission efficiency index of the APF-DTL as an example to illustrate the specific process of establishing the mapping relationship through formula (4).

[0095] For the APF-DTL, the ideal transmission efficiency is 100%. When establishing the mapping relationship, the optimization index of transmission efficiency refers to the difference between the ideal transmission efficiency and the actual transmission efficiency (100% minus the actual transmission efficiency). The optimization index is mapped to [0,1]. The larger the difference, the smaller the membership of the optimization index. When the difference between the ideal transmission efficiency and the actual transmission efficiency is 100%, the membership of the transmission efficiency index is 1; when the difference is 2%, the membership is 0.8; when the difference is 5%, the membership is 0.5.

[0096] Substituting the two sets of numbers X[1]=2, F(X[1])=0.8; X[2]=5, F(X[2])=0.5 into formula (4), we can obtain:

[0097]

[0098] have to: , These two mapping parameters. Then the mapping function of the transmission efficiency index can be obtained as:

[0099] (5)

[0100] At this time, any transmission efficiency index value (unit %) is substituted into formula (5) to obtain the corresponding transmission efficiency index membership.

[0101] Furthermore, for the remaining optimization indicators, the mapping relationship between the indicator and the membership degree can also be established by referring to the above process, and this application will not go into details here. However, when solving the mapping parameters by the undetermined coefficient method, it is actually necessary to first determine two sets of mappings, that is, it is necessary to first determine the mapping relationship between the two sets of optimization indicator parameter values ​​and the membership degree. In this regard, the two sets of values ​​can actually be appropriately adjusted according to demand. The present application provides a mapping relationship between different optimization indicators and membership degrees, as shown in Table 1.

[0102] Table 1 Corresponding relationship between the optimization index parameter values ​​and membership degree

[0103] Indicator name Index value 1 Membership 1 Index value 2 Membership 2 Transmission efficiency index / % 2 0.8 5 0.5 Transverse envelope index / mm 1 0.8 2 0.5 x-x'emittance growth index / % 10 0.8 15 0.6 y-y' emissivity growth index / % 10 0.8 15 0.6 PW emittance growth index / % 15 0.8 25 0.6 Export energy index / MeV 0.02 0.8 0.06 0.5

[0104] As shown in Table 1, when the membership 1 and membership 2 remain unchanged, the smaller the index values ​​1 and 2 are, the stricter the restriction on the optimization index is. On the contrary, when the values ​​1 and 2 remain unchanged, the lower the membership, the stricter the restriction is. The membership value of a specific optimization index can be changed according to actual needs, so as to optimize the specific optimization index. In addition, when selecting the value, it is necessary to consider the actual range of the optimization index, and ensure that the membership corresponding to different index values ​​has sufficient gradient. Taking the transmission efficiency as an example, the ideal transmission efficiency should be greater than 99%. When the transmission efficiency is less than 95%, the beam loss is already very serious, and the response membership value is small. Therefore, when selecting the membership value, the transmission efficiency of 98% is 0.8, and the membership value is 0.5 when the transmission efficiency is 95%. In this way, the transmission efficiency [95%, 100%] is mapped to the range of [0.5, 1]. When the transmission efficiency is lower than 95%, the membership is lower than 0.5, which can motivate the optimization program to optimize the transmission efficiency to more than 95%. The values ​​of other optimization indicators are determined similarly, combining the beam dynamics characteristics of APF-DTL and the optimization objectives.

[0105] For example, for the lateral envelope index, the TraceWin beam dynamics calculation software is used to calculate the 3 times rms (root mean square) envelope in the x-direction and y-direction, and then the maximum envelope point in the entire beam is found. At this time, the difference between the maximum envelope and the ideal envelope is calculated, and the difference is mapped to [0,1]. Combined with design experience, this embodiment sets the ideal envelope to 8mm. When the envelope is less than or equal to 8mm, the corresponding index membership is 1. When the maximum envelope is 9mm, the difference is 1, and the corresponding membership value is 0.8; when the maximum envelope is 10mm, the difference is 2, and the corresponding membership is 0.5.

[0106] For another example, for the x-x' phase space emittance growth index, y-y' phase space emittance growth index, and PW phase space emittance growth index of the beam, the TraceWin beam dynamics calculation software is used to calculate the emittance of each position of the accelerator in the x-x' phase space, y-y' phase space, and PW phase space, and find the position where the emittance increases the most relative to the entrance, and then define (maximum emittance / entrance emittance-1) as the maximum emittance growth. The maximum emittance growth of the beam in the three phase spaces can be used as an optimization indicator to evaluate the beam quality. Among them, when the emittance increases by 10% relative to the entrance, its membership is taken as 0.8, and when it increases by 15%, its membership is taken as 0.6.

[0107] Furthermore, in APF-DTL, the longitudinal emittance often increases more than the transverse emittance, so when determining the membership function, the transverse and longitudinal mapping functions can also have different values. For example, when the longitudinal maximum emittance increases by 15%, the membership is 0.8, and when it increases by 25%, the membership is 0.6. The distinction between transverse and longitudinal is more in line with the actual situation of APF-DTL dynamics.

[0108] It should be noted that the membership calculation of emittance growth needs to find the maximum value of emittance growth in the whole length of the beam, and cannot take the emittance at the exit. Because in the actual optimization process, it is found that when the exit emittance is used for calculation, if the emittance accounts for a large proportion in the fitness function, the program will actively discard some particles and sacrifice the transmission efficiency to preserve the emittance growth. The global optimal solution finally obtained has good emittance growth and lateral envelope, but the transmission efficiency is extremely low, which obviously cannot be used as the optimization result.

[0109] For another example, for the exit energy index, the TraceWin beam dynamics calculation software is used to calculate the actual exit energy, and the absolute value of the difference between the actual exit energy and the designed exit energy is mapped to [0,1]. When the energy difference is 0.02MeV, the membership is 0.8; when the energy difference is 0.06MeV, the membership is 0.6. In this way, the exit energy can be used as one of the optimization indicators, and the program can be used to optimize the exit energy to near the design value.

[0110] S32, performing weighted calculation on the membership of the plurality of optimization indicators to obtain a basic fitness function.

[0111] Since the normalized fitness function integrates multiple optimization indicators into a fitness parameter for evaluating beam quality, different optimization indicators can be designed with different weights, so that the key optimization of a certain optimization indicator can be achieved by adjusting the corresponding weight of each indicator. Figure 4 A schematic diagram of the process of optimizing the beam dynamics according to an embodiment of the present application is shown. Figure 4 As shown, in the embodiment of the present application, a plurality of optimization index memberships are weightedly calculated to obtain a basic fitness function, including steps S321 to S324.

[0112] S321: construct a judgment matrix, where the judgment matrix is ​​used to judge the importance of the multiple optimization indicators.

[0113] Specifically, after multiple optimization indicators are determined, the relative importance of multiple optimization indicators is judged using the judgment matrix C. Assuming there are n optimization indicators, numbered 1, 2, ..., n, the judgment matrix C is an n-order square matrix C n×n . The element C in the judgment matrix C ij Represents the importance of optimization index i relative to optimization index j, expressed by integers 1 to 9 and their reciprocals, that is, Among them, if C ij >1, it means that optimization index i is more important than optimization index j. The larger the number, the greater the difference in importance. On the contrary, if C ij <1, it indicates that the optimization index j is more important than the optimization index i. The smaller the number, the greater the difference in importance. ij =1, then the importance of the two optimization indicators is the same. The importance corresponding to the value of the specific judgment matrix C is shown in Table 2.

[0114] Table 2 The values ​​of the judgment matrix and their corresponding meanings

[0115] Cij assignment meaning illustrate 1 Equally important <![CDATA[Indicator C i and Indicator C j have the same importance]]> 3 Slightly important <![CDATA[Indicator C i Compared with Indicator C j is slightly more important]]> 5 Obviously important <![CDATA[Indicator C i than Indicator C j is significantly more important]]> 7 Very important <![CDATA[Indicator C i Compared with Indicator C j is very important]]> 9 Extremely important The index Ci is more important than the index Cj in calculation 2,4,6,8 Median The importance is between 1, 3, 5, 7, 9

[0116] S322: Perform a consistency check on the judgment matrix, and when the judgment matrix meets the consistency requirement, obtain an eigenvector corresponding to the maximum eigenvalue of the judgment matrix.

[0117] Specifically, in the process of constructing the judgment matrix, it is necessary to maintain consistency when judging and comparing the importance of each optimization index, that is, no contradictory importance judgments can be made, otherwise the weights of the obtained optimization indexes will be unreasonable.

[0118] According to matrix theory, when the matrix is ​​completely consistent, the matrix has the largest characteristic root λ1=λ max =n, the remaining eigenvalues ​​are zero; when the matrices are not completely consistent, the maximum eigenvalue λ max >n, therefore, we can use the above relationship to test the consistency of the judgment matrix, that is, introduce the consistency index CI, which is defined as:

[0119] (6)

[0120] It can be seen from formula (6) that when the judgment matrix is ​​completely consistent, CI=0. In order to quantitatively test the consistency of the judgment matrix, we introduce the random consistency ratio CR=CI / RI, and define that when CR<0.10, the judgment matrix has acceptable consistency. That is, CI is used to measure the consistency of the judgment matrix, and CR is used to quantitatively test whether the consistency of the matrix is ​​within an acceptable range. Among them, RI is called the average random consistency index, and its value is determined by the order of the judgment matrix. The specific values ​​are shown in Table 3.

[0121] Table 3 Judgment matrix order and corresponding average random consistency value

[0122] Matrix order 1 2 3 4 5 6 7 8 9 RI 0.00 0.00 0.58 0.90 1.12 1.24 1.32 1.41 1.45

[0123] Among them, if the consistency index does not meet the requirements, it means that the consistency of the judgment matrix is ​​unacceptable. At this time, the value of the judgment matrix needs to be adjusted until CR < 0.10.

[0124] After obtaining the judgment matrix that meets the consistency requirement, the eigenvector corresponding to the maximum eigenvalue of the judgment matrix is ​​obtained. In some embodiments, the maximum eigenvalue λ of the judgment matrix C is assumed to be max The corresponding eigenvector is:

[0125]

[0126] Among them, I max It is actually a column vector, representing the relative weights in the judgment matrix. That is, each component a1 to a n That is, it represents the relative weights of n optimization indicators.

[0127] S323: Obtain weights of the plurality of optimization indicators based on the feature vector.

[0128] As shown above, the weights of the multiple optimization indicators can be obtained by normalizing the feature vector. That is, the weight of the i-th indicator is It is expressed as:

[0129] (7)

[0130] S324: Perform weighted calculation on the membership of the plurality of optimization indicators based on the weights of the plurality of optimization indicators to obtain a basic fitness function.

[0131] Specifically, after the weights of multiple optimization indicators are obtained, the memberships of the corresponding optimization indicators are weighted added according to the weights to obtain the basic fitness function.

[0132] In some embodiments, let the membership degree of the i-th indicator be , the weight is , then the basic fitness function of n indicators is:

[0133] (8)

[0134] Obviously, formula (8) is the basic fitness function without constraints, with a value range of [0,1], and the result is optimal when the fitness function is 1, and the closer it is to 0, the worse the result. Through formula (8), multiple optimization indicators can be unified into one fitness evaluation parameter to evaluate the beam quality affected by the phase sequence in APF-DTL, that is, the multi-objective optimization in APF-DTL is converted into a single-objective optimization.

[0135] S33. Acquire the normalized fitness function based on the basic fitness function and constraint conditions.

[0136] As mentioned above, in an alternating phase focusing drift tube linear accelerator (APF-DTL), the length of the drift tube is related to the synchronous phase difference between two adjacent gaps. If the distance between two adjacent gaps is too small, it will cause difficulties in processing and placing the drift tube. Therefore, when considering the phase sequence optimization, in addition to considering the influence of the phase sequence on beam transmission, it is also necessary to constrain the phase difference between adjacent phases to not exceed a certain value. That is, the constraint condition includes the constraint value of the phase difference between adjacent phases in the alternating phase focusing drift tube linear accelerator.

[0137] In some embodiments, the constraint condition is restricted by a penalty function. In combination with the APF-DTL phase distribution characteristics, a static penalty function is used to restrict the constraint condition, and the final normalized fitness function is obtained:

[0138] (9).

[0139] in, This is the basic fitness function of formula (8); is the penalty function, a is a parameter with a value of (0,1) and is used to adjust the size of the penalty term; m is the number of gaps; As constraints, is the Heaviside function, expressed as:

[0140] (10).

[0141] From formula (10), we can see that if , then the constraint is violated and a penalty term needs to be imposed. , the constraint is satisfied and no penalty is imposed.

[0142] S4. Searching for an optimal phase sequence in the iterative process of the phase particle swarm based on a normalized fitness function.

[0143] In summary, formula (9) is the normalized fitness function proposed in this application, which uses a static penalty function on the basis of the basic fitness function to transform a constrained optimization problem into an unconstrained optimization problem. That is, in the iterative process of the phase particle group, the fitness of each phase particle is calculated by formula (9), so that the optimal phase particle, that is, the optimal phase sequence, can be obtained. Therefore, the normalized fitness function proposed in the above embodiment of the present application can convert the multi-objective optimization in the APF-DTL dynamic optimization process into a single-objective and unconstrained optimization, so as to achieve a fast and refined search of the phase sequence.

[0144] The following is an example to illustrate the specific process and technical effects of the APF-DTL beam dynamics optimization method provided in this application.

[0145] In this embodiment, the inlet kinetic parameters of the APF-DTL to be optimized are shown in Table 4. In addition, the inlet phase diagram of the APF-DTL to be optimized is shown in Table 4. Figure 5 As shown. Figure 5 In the figure, the red ellipse is the 99.9% boundary.

[0146] Table 4 APF-DTL inlet dynamic parameters

[0147] Entry parameters Parameter Value Entry energy / MeV 0.6915 Peak current / mA 14.53 Operating frequency / MHz 200 Duty cycle / % 5 Normalized rms emittance at inlet xx / mm∙mrad 0.175 Normalized rms emittance at inlet yy' / mm∙mrad 0.207 Entrance longitudinal rms emittance / deg∙MeV 0.059

[0148] For the optimized APF-DTL, the fixed phase is -10°, and the initial phases of the remaining 14 gaps are calculated using the Iwata formula. , a =0.0001, b =1.7, c =0.0072, According to formula (1), the initial phase sequence is calculated as follows: Figure 2 As shown. Afterwards, a random number is used to generate the initial phase within the range of ±20° of each initial phase. If the generated random number exceeds the phase boundary, the initial phase is allowed to stay on the boundary. Each phase group contains 14 gaps, with a total of 20 groups, i.e., 20 phase particles, constituting a phase particle group. Among them, the initial velocity of the phase particle is the change in the phase in the first iteration. Based on experience, the initial velocity is set to +4. The velocity of the phase particle in the subsequent iteration process is calculated by formula (2).

[0149] In order to evaluate the fitness of each phase particle through the normalized fitness function in the iterative process of the phase particle group, the weight of each optimization index should be determined first. Taking the optimization indexes including the transmission efficiency index, the beam transverse envelope index (x direction and y direction), the beam x-x' phase space emittance growth index, the beam y-y' phase space emittance growth index, and the beam PW phase space emittance growth index as an example, the judgment matrix can be constructed as shown in Table 5.

[0150] Table 5 Judgment matrix

[0151] Trans_eff Envelope_x Envelope_y Emit_x Emit_y Emit_P Trans_eff 1 7 7 6 5 5 Envelope_x 1 / 7 1 1 1 / 4 1 / 2 1 / 2 Envelope_y 1 / 7 1 1 1 / 4 1 / 2 1 / 2 Emit_x 1 / 6 4 4 1 2 2 Emit_y 1 / 5 2 2 1 / 2 1 1 Emit_P 1 / 5 2 2 1 / 2 1 1

[0152] The judgment matrix is ​​judged for consistency, and the calculated CR value is 0.023468, which meets the consistency requirement. Then the eigenvector corresponding to the maximum eigenvalue of the matrix is ​​calculated and normalized to obtain the weight vectors of the six optimization indicators (0.53, 0.05, 0.05, 0.17, 0.095, 0.095). It should be noted that the export energy indicator can also be added to form 7 optimization indicators.

[0153] Furthermore, when optimizing the APF-DTL, it is necessary to ensure that the drift tube is larger than 10 mm. , which can meet the requirements of the drift tube. As mentioned above, the value range of formula (8) is [0,1]. According to the minimum penalty function principle, the penalty function term should not be too large, so the phase difference needs to be normalized. And the penalty function should not be too small to ensure that the optimization result meets the constraint conditions. Therefore, considering the phase distribution of APF-DTL, the constraint condition can be Defined as:

[0154] (11)

[0155] Then substitute formula (11) into formula (9). When formula (11) is greater than 0, the constraint condition is satisfied and the penalty function takes the value of 0. When formula (11) is less than 0, the constraint condition is violated and H(-g i (x)) takes the value of 1, and the penalty function is calculated as , where m is the number of gaps. Then the final normalized fitness function can be obtained based on the penalty function and the basic fitness function.

[0156] Specifically, after the weights and penalty functions of the optimization indicators are determined, the fitness of the phase particles of the phase particle swarm can be calculated based on the normalized fitness function, so as to search for the optimal phase sequence in the iterative process of the phase particle swarm.

[0157] As mentioned above, when the APF-DTL is optimized, a total of 20 phase particles are set, each particle is a set of independent phase sequences, and the 20 groups of particles are calculated in parallel, which can expand the search range and shorten the convergence time. Twenty particles generate twenty groups of initial phases, and the phase boundaries are set to -80° and 80°. The maximum speed is 8. When the calculated particle speed exceeds the maximum value, . Set 200 iterations, and the maximum value of the fitness function changes with the number of iterations during the iteration process as shown in Figure 6 As shown. Figure 6 It can be seen that in the early stage of the iteration process, the particle search range is large, so the fitness changes greatly, and gradually converges to the optimal value in the later stage. The fitness function only increases slightly, and the fitness function reaches 0.513 at the end of the iteration.

[0158] Figure 7 and Figure 8 The optimal phase sequence and drift tube length after APF-DTL optimization are shown. Figure 7 As shown in , since the first gap synchronization phase is fixed at -10°, and in order to improve the longitudinal acceptance at the entrance, the synchronization phase of the first few gaps should be close to the minimum value, so the synchronization phase of the first few gaps is gradually reduced to -80°. Figure 8 As shown, the drift tube is also shorter, and the shortest drift tube length is 10.14 mm (the third drift tube), which meets the engineering requirements. It can be considered that the optimization result meets the constraints.

[0159] Fig. 9 The optimized APF-DTL outlet phase diagram is shown in Figure 1, where Fig. 9 The red ellipse in the figure is the 99.9% boundary. Fig.10 shows the RMS emittance growth after the APF-DTL optimization, Fig.11The lateral 3-fold RMS envelope after APF-DTL optimization is shown. Figures 12a to 12d The transverse and longitudinal beam density diagrams, phase density diagrams, and energy density diagrams of the APF-DTL after optimization are shown. It can be seen that the optimization results of the beam dynamics perform well in the transverse direction, the transverse normalized RMS emittance increases slightly, and the transverse 3-fold RMS envelope is less than 8mm, achieving the optimization goal. However, the longitudinal divergence of the beam is large, and the RMS emittance increases from 0.28 to 1.8, which is a common problem of APF-type DTL. From the longitudinal exit phase diagram, it can be seen that the exit particle energy is dispersed within 0.25MeV, and no excessive particle loss is caused, and the transmission efficiency is higher than 99%.

[0160] Furthermore, the optimized kinetic parameters of the APF-DTL are shown in Table 6.

[0161] Table 6 Kinetic parameters of APF-DTL optimization results

[0162] APF-DTL kinetic parameters Parameter Value Entry energy / MeV 0.69 Export energy / MeV 2.50 Number of gaps 15 Transmission efficiency / % 99.0 Acceleration length / cm 59.5 Export x-x' normalized rms emittance / mm·mrad 0.3524 Export y-y' normalized rms emittance / mm·mrad 0.3191 Export PW rms emittance / deg·MeV 0.3665

[0163] It can be seen that the optimization method of beam dynamics provided in the embodiment of the present application can achieve a good optimization effect. And from the perspective of dynamics, the optimized dynamic parameters fully meet the requirements of practical applications.

[0164] Therefore, the optimization method of beam dynamics provided in the embodiment of the present application introduces the particle swarm optimization algorithm into the beam dynamics optimization scenario of APF-DTL, normalizes the optimization indicators of multiple beams into a fitness function, transforms the multi-objective optimization problem into a single-objective optimization problem, simplifies the optimization process, and reduces the optimization cost. At the same time, the present application introduces a penalty function to restrict the constraints, transforms the constrained optimization into the unconstrained optimization, and finally proposes a single-objective optimization algorithm suitable for APF-DTL dynamics optimization, thereby effectively realizing the dynamics optimization of APF-DTL.

[0165] The protection scope of the APF-DTL beam dynamics optimization method of the embodiment of the present application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present application are included in the protection scope of the present application.

[0166] An embodiment of the present application also provides an APF-DTL beam dynamics optimization device, which can implement the APF-DTL beam dynamics optimization method described in the present application. However, the implementation device of the APF-DTL beam dynamics optimization device described in the present application includes but is not limited to the structure of the APF-DTL beam dynamics optimization device listed in the present embodiment. All structural deformations and replacements of the prior art made according to the principles of the present application are included in the protection scope of the present application.

[0167] Fig.13 The schematic diagram of the structure of the APF-DTL beam dynamics optimization device described in the embodiment of the present application is shown as follows: Fig.13 As shown, the beam dynamics optimization device includes:

[0168] An acquisition module, used to acquire the initial phase sequence of APF-DTL;

[0169] A generation module, used to randomly generate multiple groups of phase sequences within a set generation range based on the initial phase sequence to form a phase particle group; each phase sequence corresponds to a phase particle;

[0170] An indicator module, used to determine a plurality of optimization indicators of the beam dynamics of the APF-DTL, so as to determine a normalized fitness function based on the plurality of optimization indicators and constraint conditions;

[0171] A search module is used to search for an optimal phase sequence in the iterative process of the phase particle swarm based on the normalized fitness function.

[0172] It should be noted that the structures and principles of the acquisition module, the generation module, the indicator module and the search module, as well as the beneficial effects achieved by the device are the same as those in the above embodiment, and will not be described in detail here.

[0173] The embodiment of the present application also provides a storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, all steps of the beam dynamics optimization method of the APF-DTL embodiment are implemented.

[0174] Among them, the specific steps of the APF-DTL beam dynamics optimization method and the beneficial effects obtained by applying the readable storage medium provided in the embodiment of the present application are the same as those in the above embodiment and will not be described in detail here.

[0175] A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above embodiment can be completed by instructing a processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state disk (SSD)), etc.

[0176] The embodiment of the present application also provides a terminal. Fig.14 A schematic diagram of the structure of the terminal described in the embodiment of the present application is shown. Fig.14 As shown, the terminal of the embodiment of the present application includes a processor and a memory, and the memory and the processor are communicatively connected; the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory, so that the terminal executes all steps of the beam dynamics optimization method of the APF-DTL in the above-mentioned embodiment.

[0177] The specific steps of the beam dynamics optimization method of APF-DTL and the beneficial effects obtained by applying the terminal provided in the embodiment of the present application are the same as those in the above embodiment and will not be described in detail here.

[0178] It should be noted that the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. Similarly, the processor may also be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0179] Although the embodiments disclosed in this application are as above, the contents described are only embodiments adopted for facilitating the understanding of this application and are not intended to limit this application. Any technician in the technical field to which this application belongs can make any modifications and changes in the form and details of implementation without departing from the spirit and scope disclosed in this application, but the scope of protection of this application shall still be based on the scope defined in the attached claims.

Claims

1. A beam dynamics optimization method for APF-DTL, characterized in that: The method comprises: Get the initial phase sequence of APF-DTL; Based on the initial phase sequence, a plurality of phase sequences are randomly generated within a set generation range to form a phase particle group; wherein each phase particle corresponds to a phase sequence; Determining a plurality of optimization indicators of the beam dynamics of the APF-DTL to determine a normalized fitness function based on the plurality of optimization indicators and constraints; Searching for an optimal phase sequence in an iterative process of the phase particle swarm based on the normalized fitness function; Wherein, determining the normalized fitness function based on the multiple optimization indicators and constraint conditions includes: Acquire corresponding optimization index membership based on the multiple optimization indexes; Performing weighted calculation on the membership of multiple optimization indicators to obtain a basic fitness function; The normalized fitness function is obtained based on the basic fitness function and constraints.

2. The APF-DTL beam dynamics optimization method according to claim 1, characterized in that: The initial phase sequence for obtaining APF-DTL includes: Get multiple free parameters; A plurality of initial phases are calculated based on the plurality of free parameters to obtain the initial phase sequence.

3. The APF-DTL beam dynamics optimization method according to claim 1, characterized in that: The multiple optimization indicators include: the transmission efficiency indicator of the APF-DTL, the lateral envelope indicator of the beam, the x-x' phase space emittance growth indicator of the beam, the y-y' phase space emittance growth indicator of the beam, the PW phase space emittance growth indicator of the beam, and the export energy indicator.

4. The APF-DTL beam dynamics optimization method according to claim 1, characterized in that: Acquiring corresponding optimization index membership based on the multiple optimization indexes includes: Obtaining a mapping parameter corresponding to each optimization indicator among the multiple optimization indicators; The optimization indicator membership corresponding to each optimization indicator is obtained based on the mapping parameters.

5. The APF-DTL beam dynamics optimization method according to claim 1, characterized in that: The membership of multiple optimization indicators is weighted and calculated to obtain the basic fitness function: Constructing a judgment matrix, wherein the judgment matrix is ​​used to judge the importance of the multiple optimization indicators; Performing a consistency check on the judgment matrix, and when the judgment matrix meets the consistency requirement, obtaining an eigenvector corresponding to the maximum eigenvalue of the judgment matrix; Obtaining weights of a plurality of the optimization indicators based on the feature vector; The memberships of the plurality of optimization indicators are weightedly calculated based on the weights of the plurality of optimization indicators to obtain a basic fitness function.

6. The APF-DTL beam dynamics optimization method according to claim 1, characterized in that: The constraint condition includes: a constraint value of a phase difference between adjacent phases in the APF-DTL; wherein the constraint condition is restricted by a penalty function.

7. A beam dynamics optimization device for APF-DTL, characterized in that: The device comprises: An acquisition module, used for acquiring an initial phase sequence of APF-DTL; A generation module, used to randomly generate multiple groups of phase sequences within a set generation range based on the initial phase sequence to form a phase particle group; each phase sequence corresponds to a phase particle; An indicator module, used to determine a plurality of optimization indicators of the beam dynamics of the APF-DTL, so as to determine a normalized fitness function based on the plurality of optimization indicators and constraint conditions; A search module, used for searching an optimal phase sequence in an iterative process of the phase particle swarm based on the normalized fitness function; Wherein, determining the normalized fitness function based on the multiple optimization indicators and constraint conditions includes: Acquire corresponding optimization index membership based on the multiple optimization indexes; Performing weighted calculation on the membership of multiple optimization indicators to obtain a basic fitness function; The normalized fitness function is obtained based on the basic fitness function and constraints.

8. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the beam dynamics optimization method of the APF-DTL described in any one of claims 1 to 6 is implemented.

9. A terminal, characterized in that: It includes a processor and a memory, and the memory is communicatively connected to the processor; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the APF-DTL beam dynamics optimization method as described in any one of claims 1 to 6.

Images