An adaptive time step control method for periodic impulse excitation simulation

Abstract

The present application relates to a kind of for periodic pulse excitation simulation adaptive time step control method, belong to computer aided engineering numerical simulation technical field, solve the existing calculation step number too much, simulation efficiency low Problem.It includes: according to the working frequency of pulse excitation source, duty ratio and period multiple factor determine pulse period and width;Pulse period is divided into pulse excitation zone, pulse post-transition zone, relaxation period middle section, pulse pre-transition zone in time domain;According to the minimum step number and width of pulse excitation zone determine fine step;According to the reference step number, period and width of relaxation period middle section determine coarse step;According to the step length growth parameter of pulse post-transition zone, from fine step start to calculate the step length of each step until coarse step, obtain the step length sequence of pulse post-transition zone;According to the step length attenuation parameter of pulse pre-transition zone, from coarse step start to calculate the step length of each step until fine step, obtain the step length sequence of pulse pre-transition zone.Total calculation step number is reduced, and simulation efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of computer-aided engineering (CAE) numerical simulation technology, and in particular to an adaptive time step control method for periodic pulse excitation simulation. Background Technology

[0002] In engineering simulations involving periodic pulse excitation, the excitation source exhibits typical pulse characteristics: extremely short pulse widths (typically on the order of milliseconds or even microseconds), high peak amplitudes, and relatively long pulse periods. Accurately capturing the rapid response during the pulse requires extremely small time steps, while during the relaxation phase between pulses, the response changes gradually, allowing for larger time steps to improve computational efficiency.

[0003] Commonly used time step control methods in the prior art include: 1. Fixed time step: Using an extremely small step size throughout the process results in a huge amount of computation and extremely low efficiency, which is unacceptable, especially for long-term simulations.

[0004] 2. Simple piecewise step size: Small step size is used in the impulse region and large step size in the relaxation region. However, abrupt changes in step size can easily cause numerical oscillations, leading to non-physical response steps. The mechanism is that a sudden change in the time step size is equivalent to introducing an artificial discontinuity on the time axis. This causes the local truncation error of the discrete scheme (such as finite difference, finite element, etc.) to increase instantaneously and may excite high-frequency numerical modes unrelated to the physical response. These non-physical oscillations can contaminate the simulation results and even cause computational divergence.

[0005] 3. Conventional Adaptive Transition: While variable step-size algorithms based on local truncation error estimation can automatically adjust the step size, the error estimator is often not sensitive enough to rapid changes when dealing with periodic pulses with extremely short pulse widths. This leads to excessively large step sizes during pulse intervals. When the pulse width is extremely short, the step size may directly span the entire pulse, resulting in ineffective resolution of the pulse excitation and loss of key physical processes. Furthermore, such methods are typically general-purpose designs and are not optimized for the repetitive characteristics of periodic pulses, leaving considerable room for improvement in computational efficiency.

[0006] Therefore, there is an urgent need for an adaptive time step strategy that can both ensure computational accuracy and minimize the number of computation steps. Summary of the Invention

[0007] Based on the above analysis, the embodiments of the present invention aim to provide an adaptive time step control method for periodic pulse excitation simulation, in order to solve the problems of excessively large total number of calculation steps and low simulation efficiency obtained by existing methods.

[0008] On one hand, embodiments of the present invention provide an adaptive time step control method for periodic pulse excitation simulation, comprising the following steps: The operating frequency, duty cycle, and period factor of the pulse excitation source used for simulation are obtained, and the pulse period and pulse width are determined based on the operating frequency, the duty cycle, and the period factor; wherein, a pulse period is divided in the time domain into a pulse excitation region, a post-pulse transition region, a relaxation period mid-segment, and a pre-pulse transition region. Obtain the minimum number of steps in the pulse excitation region, and determine the fine step size of the pulse excitation region based on the pulse width and the minimum number of steps; Obtain the reference number of the relaxation period mid-segment, and determine the coarse step size of the relaxation period mid-segment based on the pulse period, the pulse width, and the reference number of steps; Obtain the step size growth parameter of the post-pulse transition region, and starting from the fine step size, calculate the step size of each step according to the step size growth parameter until it grows to the coarse step size, thereby obtaining the step size sequence of the post-pulse transition region. Obtain the step size attenuation parameter of the pre-pulse transition region. Starting from the coarse step size, calculate the step size of each step according to the step size attenuation parameter until it attenuates to the fine step size, thereby obtaining the step size sequence of the pre-pulse transition region.

[0009] Based on a further improvement of the above method, the step size growth parameter is the step size growth rate; the step size decay parameter is the step size decay factor. Starting from the fine step size, each step size is calculated based on the step size growth parameters, including: The common ratio of step size growth in the exponential mode is determined based on the step size growth rate. Starting from the fine step size, each step size is calculated based on the common ratio of step size growth according to the exponential law. Starting from the coarse step size, the step size for each step is calculated based on the step size decay parameter, including: The step size decay factor is used as the common ratio of step size decay in the exponential mode, and the step size is calculated based on the common ratio of step size decay according to the exponential law.

[0010] Based on a further improvement of the above method, the step size growth rate is less than or equal to the reciprocal of the step size decay factor minus 1.

[0011] Based on a further improvement of the above method, the step size growth parameter is the expected number of growth steps in linear mode, and the step size decay parameter is the expected number of reduction steps in linear mode. Starting from the fine step size, each step size is calculated based on the step size growth parameters, including: The step size increment is determined based on the expected number of growth steps, the fine step size, and the coarse step size; Starting from the fine step size, calculate each step size based on the step size increment according to a linear law; Starting from the coarse step size, the step size for each step is calculated based on the step size decay parameter, including: The step size reduction is determined based on the desired reduction in steps, the fine step size, and the coarse step size; Starting from the coarse step size, each step size is calculated by decreasing the step size according to a linear law.

[0012] Further improvements to the above method also include: The total duration of the post-pulse transition region is determined based on the step size sequence of the post-pulse transition region. The total duration of the pre-pulse transition region is determined based on the step size sequence of the pre-pulse transition region. The total duration of the relaxation period is determined based on the pulse period, the pulse width, the total duration of the post-pulse transition region, and the total duration of the pre-pulse transition region. Determine whether the total duration of the relaxation period is an integer multiple of the coarse step size of the relaxation period. If it is not an integer multiple, determine the remaining time between the total duration of the relaxation period and an integer multiple of the coarse step size of the relaxation period. The step size sequence of the pre-pulse transition region is adjusted based on the remaining time.

[0013] A further improvement to the above method involves adjusting the step size sequence of the pre-pulse transition region based on the remaining time, including: The adjustment term is determined based on the first term of the step size sequence of the pre-pulse transition region and the remaining time; The adjustment term is added before the first term of the step size sequence in the pre-pulse transition region; Replace the first term of the step size sequence in the pre-pulse transition region with the aforementioned adjustment term; Replace the step size sequence of the pre-pulse transition region with the adjusted step size sequence described above.

[0014] A further improvement to the above method, the step of determining the adjustment term based on the first term of the step size sequence of the pre-pulse transition region and the remaining time, includes: The adjustment term is the sum of the first term of the step size sequence of the pre-pulse transition region and half of the remaining time.

[0015] A further improvement to the above method, the step size sequence of the pre-pulse transition region is adjusted according to the remaining time, including: The scaling factor is determined based on the total duration of the pre-pulse transition region and the remaining time. The adjusted step sequence is obtained by multiplying the scaling factor by the step sequence of the pre-pulse transition region.

[0016] A further improvement to the above method, the step of determining the scaling factor based on the total duration of the pre-pulse transition region and the remaining time, includes: The ratio of the sum of the total duration of the pre-pulse transition region and the remaining time to the total duration of the pre-pulse transition region is used as a scaling factor.

[0017] Based on a further improvement of the above method, when the growth exceeds the coarse step size, the coarse step size is taken in the last step; when the decay drops below the fine step size, the fine step size is taken in the last step.

[0018] On the other hand, embodiments of the present invention provide an adaptive time step control device for periodic pulse excitation simulation, comprising: The data acquisition module is used to obtain the operating frequency, duty cycle and period multiple factor of the pulse excitation source used for simulation, the minimum number of steps in the pulse excitation region, the reference number of steps in the relaxation period, the step size growth parameter of the post-pulse transition region, and the step size decay parameter of the pre-pulse transition region. Among them, a pulse cycle is divided into the pulse excitation region, the post-pulse transition region, the relaxation period, and the pre-pulse transition region in the time domain. The step size calculation module is used to determine the pulse period and pulse width based on the operating frequency, the duty cycle, and the period multiplier factor; determine the fine step size of the pulse excitation region based on the pulse width and the minimum number of steps; determine the coarse step size of the relaxation period based on the pulse period, the pulse width, and the reference number of steps; calculate each step size based on the step size growth parameter starting from the fine step size, until it grows to the coarse step size, to obtain the step size sequence of the post-pulse transition region; and calculate each step size based on the step size decay parameter starting from the coarse step size, until it decays to the fine step size, to obtain the step size sequence of the pre-pulse transition region.

[0019] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects: 1. Eliminating numerical oscillations: By setting a smooth transition region, the time step is continuously varied between the pulse region and the relaxation region, avoiding the surge in truncation error caused by abrupt changes in the step size and the excitation of non-physical high-frequency modes. This completely eliminates numerical oscillations caused by abrupt changes in the step size, ensuring the physical authenticity and numerical stability of the simulation solution.

[0020] 2. Significantly Reduced Computational Steps: By dividing the pulse period into a pulse region, a post-transition region, a mid-relaxation period, and a pre-transition region, and employing a smooth step-size transition strategy, this method significantly reduces the number of computational steps during the relaxation phase while maintaining computational accuracy throughout the pulse period. Compared to traditional methods that use fixed, fine step sizes throughout, this method typically reduces the total number of computational steps by more than an order of magnitude, and the smaller the ratio of pulse width to period (i.e., the smaller the duty cycle), the more significant the efficiency improvement. Significant efficiency gains can be achieved regardless of whether exponential or linear transitions are used, or whether the design is symmetric or asymmetric.

[0021] 3. Wide applicability: This method does not depend on specific physical fields or engineering backgrounds and can be applied to numerical simulations of various periodic pulse excitation problems in thermal, acoustic, and structural mechanics. Users only need to set basic parameters such as pulse frequency and duty cycle according to actual working conditions to integrate this method into their existing simulation workflow.

[0022] 4. Multiple implementation methods: It covers various transition laws such as exponential and linear, and supports symmetric / asymmetric configurations, making it easy to select the appropriate strategy according to the specific problem.

[0023] 5. Easy to integrate: It can be seamlessly integrated into mainstream commercial software (such as ANSYS Fluent, LS-DYNA, Abaqus, etc.) through user-defined functions without modifying the software kernel.

[0024] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0025] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0026] Figure 1 The flowchart (I) shows an adaptive time step control method for periodic pulse excitation simulation in Embodiment 1 of the present invention. Figure 2 This is a flowchart (II) of an adaptive time step control method for periodic pulse excitation simulation in Embodiment 1 of the present invention. Figure 3 This is a schematic diagram showing the comparison of time step size between the method of the present invention and the fixed fine step size method in two cycles in Embodiment 1 of the present invention; Figure 4This is a schematic diagram illustrating the step size variation of exponential transition (solid line) and linear transition (dashed line) in Embodiment 1 of the present invention. Detailed Implementation

[0027] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0028] Example 1, One specific embodiment of the present invention discloses an adaptive time step control method for periodic pulse excitation simulation. This method is a general adaptive time step control method that, considering the different rates of response change before and after the pulse in periodic pulse excitation simulation, eliminates numerical oscillations while significantly reducing the total number of computation steps and improving simulation efficiency by setting a smooth transition region. Figure 1 As shown, it includes the following steps: S11: Obtain the operating frequency, duty cycle, and period multiplier factor of the pulse excitation source used for simulation, and determine the pulse period and pulse width based on the operating frequency, the duty cycle, and the period multiplier factor; wherein, a pulse period is divided in the time domain into a pulse excitation region, a post-pulse transition region, a relaxation period mid-segment, and a pre-pulse transition region.

[0029] In practical implementation, an existing pulse cycle includes a pulse excitation period and a relaxation phase during the pulse interval. The pulse excitation region in this invention is the pulse excitation period. This invention divides the relaxation phase during the pulse interval into three stages: the post-pulse transition region, the middle of the relaxation period, and the pre-pulse transition region. The time corresponding to each of these three stages is calculated according to the algorithm described later.

[0030] S12: Obtain the minimum number of steps in the pulse excitation region, and determine the fine step length of the pulse excitation region based on the pulse width and the minimum number of steps.

[0031] S13: Obtain the reference number of the relaxation period mid-segment, and determine the coarse step size of the relaxation period mid-segment based on the pulse period, the pulse width, and the reference number of steps.

[0032] In practical implementation, this invention is achieved through user-defined functions (UDFs) in CFD / CAE software. First, parametric modeling is performed: based on the operating frequency of the pulse excitation source... (unit: Hz) and period factor ( A positive integer, representing the repetition period of the excitation pulse. It is the reciprocal of the operating frequency. of (Multiplied by 1). When there is a one-to-one correspondence between the excitation source and the excitation object (i.e., each pulse acts directly on the same object). However, in some multi-target time-division excitation or modulation signal problems, the actual pulse period experienced by a single object may be... For example, in a scenario where multiple targets on a rotating disk are alternately excited by pulses, at this time... Equal to the number of target components.

[0033] Therefore, pulse period Defined as:

[0034] Pulse width By duty cycle The duty cycle is defined as the ratio of pulse width to period.

[0035] Set the minimum number of steps within the pulse. To obtain the fine step size (fixed step size):

[0036] Set the baseline number of steps in the middle of the relaxation period This yields the coarse step size (fixed step size):

[0037] In this invention, the minimum number of steps is set based on computational efficiency considerations. and baseline steps .

[0038] The method of this invention is parameterized and universal: all key parameters are set through macro definitions or input files, and users only need to modify the operating frequency of the pulse source. and duty cycle It can automatically adapt to different working conditions. (Period multiplier factor) Step count requirement Step growth rate (To be mentioned later), step size decay factor (As will be mentioned later) etc. can all be adjusted as needed, exhibiting high flexibility and versatility.

[0039] S14: Obtain the step size growth parameter of the post-pulse transition region. Starting from the fine step size, calculate the step size of each step according to the step size growth parameter until it grows to the coarse step size, and obtain the step size sequence of the post-pulse transition region.

[0040] S15: Obtain the step size attenuation parameter of the pre-pulse transition region. Starting from the coarse step size, calculate the step size of each step according to the step size attenuation parameter until it is attenuated to the fine step size, and obtain the step size sequence of the pre-pulse transition region.

[0041] In some embodiments, the step size growth parameter is a step size growth rate; the step size decay parameter is a step size decay factor; Starting from the fine step size, each step size is calculated based on the step size growth parameters, including: The common ratio of step size growth in the exponential mode is determined based on the step size growth rate. Starting from the fine step size, each step size is calculated based on the common ratio of step size growth according to the exponential law. Starting from the coarse step size, the step size for each step is calculated based on the step size decay parameter, including: The step size decay factor is used as the common ratio of step size decay in the exponential mode, and the step size is calculated based on the common ratio of step size decay according to the exponential law.

[0042] In some embodiments, when the growth exceeds the coarse step size, the coarse step size is used in the last step; when the decay falls below the fine step size, the fine step size is used in the last step.

[0043] In practical implementation, this invention uses an exponential law as the method for step size transition. Within one cycle, the time axis is divided into four continuous regions: the pulse excitation region, the post-pulse transition region, the middle of the relaxation period, and the pre-pulse transition region. Among them, the step size in the two transition regions changes exponentially to ensure a smooth step size transition and fundamentally eliminate numerical oscillations caused by abrupt changes in step size.

[0044] Post-pulse transition region (growth): User-defined step size growth rate Common ratio step length growth .from Begin by multiplying by each step. (>1), until the step size reaches or exceeds The final step is to force the removal. This ensures that the step size accurately reaches the target value.

[0045] Pre-pulse transition region (attenuation): User-defined step-size attenuation factor ( The common ratio of step size decay is ), .from Begin by multiplying by each step. until the step size drops to The following is the final step: forced retrieval. This ensures that the step size accurately returns to the fine step size.

[0046] Among them, step size growth rate The step size decay factor can be adjusted within the range of 10% to 80%. It can be adjusted within the range of 0.1 to 0.9 to meet the smoothness requirements of different problems.

[0047] In some embodiments, the step size growth rate is less than or equal to the reciprocal of the step size decay factor minus 1.

[0048] In specific implementation, this invention does not limit the rate of change of the transition region before and after the pulse to be the same. Preferably, an asymmetric design can be adopted, that is, the step size growth rate of the growth phase is less than the reciprocal of the step size decay factor of the decay phase minus 1. This minimizes the number of steps while maintaining accuracy. Symmetrical transition ( (This may be included as a special example of the present invention.)

[0049] In some embodiments, the step size growth parameter is the expected number of growth steps in linear mode, and the step size decay parameter is the expected number of reduction steps in linear mode; Starting from the fine step size, each step size is calculated based on the step size growth parameters, including: The step size increment is determined based on the expected number of growth steps, the fine step size, and the coarse step size; Starting from the fine step size, calculate each step size based on the step size increment according to a linear law; Starting from the coarse step size, the step size for each step is calculated based on the step size decay parameter, including: The step size reduction is determined based on the desired reduction in steps, the fine step size, and the coarse step size; Starting from the coarse step size, each step size is calculated by decreasing the step size according to a linear law.

[0050] In practical implementation, besides the exponential law, this invention can also use a linear law to achieve the step size transition as an alternative to the exponential law. In the linear transition, the step size increases or decreases according to an arithmetic sequence, and the user needs to set the desired number of steps in the transition zone after the pulse. , ,Sure and The time difference (absolute value) divided by the expected number of steps , The value of is used to obtain the step size increment (i.e., tolerance) of the transition region after the pulse and the step size decrement (i.e., tolerance) of the transition region before the pulse. It should be noted that linear transition is inferior to exponential transition in terms of smoothness and parameter controllability, and therefore is not considered a preferred embodiment of the present invention, but is one of the extended embodiments of the present invention.

[0051] In some embodiments, such as Figure 2 As shown, after S15 is executed, the following steps are also included: S21: Determine the total duration of the post-pulse transition region based on the step size sequence of the post-pulse transition region; S22: Determine the total duration of the pre-pulse transition region based on the step size sequence of the pre-pulse transition region; S23: Determine the total duration of the relaxation period mid-segment based on the pulse period, the pulse width, the total duration of the post-pulse transition region, and the total duration of the pre-pulse transition region; S24: Determine whether the total duration of the relaxation period is an integer multiple of the coarse step size of the relaxation period. If it is not an integer multiple, determine the remaining time between the total duration of the relaxation period and an integer multiple of the coarse step size of the relaxation period. S25: Adjust the step size sequence of the pre-pulse transition region according to the remaining time.

[0052] In some embodiments, adjusting the step size sequence of the pre-pulse transition region based on the remaining time includes: The adjustment term is determined based on the first term of the step size sequence of the pre-pulse transition region and the remaining time; The adjustment term is added before the first term of the step size sequence in the pre-pulse transition region; Replace the first term of the step size sequence in the pre-pulse transition region with the aforementioned adjustment term; Replace the step size sequence of the pre-pulse transition region with the adjusted step size sequence described above.

[0053] In some embodiments, determining the adjustment term based on the first term of the step size sequence of the pre-pulse transition region and the remaining time includes: The adjustment term is the sum of the first term of the step size sequence of the pre-pulse transition region and half of the remaining time.

[0054] In practice, after pre-calculating the transition zone step size sequence, the total duration of the transition zone after the pulse is obtained. Total duration of the pre-pulse transition region The duration of the middle phase of the relaxation period is as follows: .

[0055] Since the step size sequence of the transition region is fixed, Maybe not Integer multiples of the number of steps in the middle of the relaxation period If the value is not an integer, the number of steps in the middle of the relaxation period needs to be rounded down. Rounding may result in the total duration of the relaxation period being shorter than the original total duration, thus creating residual time. , To absorb the residual time generated during the relaxation period due to step size rounding. This requires adjusting the original pre-pulse transition region step size sequence so that the total duration of the adjusted new pre-pulse transition region step size sequence equals the original total duration of the pre-pulse transition region plus the remaining time. To achieve a strictly closed period without disrupting the smoothness of the transition region, this invention proposes an original "one-step insertion method," the steps of which are as follows: Let the original pre-pulse transition region step size sequence be... ,in The common ratio is known. Let... Construct a new sequence The total number of steps becomes The total duration of this sequence increased compared to the original sequence. The step size sequence of the original pre-transition region is adjusted to precisely absorb the remaining time generated by rounding down the step number in the middle of the relaxation period. Furthermore, the step size of the new sequence exhibits a monotonically non-increasing trend.

[0056] The advantage of this method is that: (1) After the above calculation Therefore, the first step of the new sequence This ensures that the step length from the last step in the middle of the relaxation period to the first step in the pre-transition region does not increase, thus avoiding new mutations. (2) When At this point, the new sequence monotonically decreases overall, maintaining the smoothness of the original design. For commonly used decay factors... This condition is automatically met, therefore this method is universally applicable within the typical parameter range.

[0057] In some embodiments, adjusting the step size sequence of the pre-pulse transition region according to the remaining time includes: The scaling factor is determined based on the total duration of the pre-pulse transition region and the remaining time. The adjusted step sequence is obtained by multiplying the scaling factor by the step sequence of the pre-pulse transition region.

[0058] In some embodiments, determining the scaling factor based on the total duration of the pre-pulse transition region and the remaining time includes: The ratio of the sum of the total duration of the pre-pulse transition region and the remaining time to the total duration of the pre-pulse transition region is used as a scaling factor.

[0059] In practical implementation, when encountering extreme parameters (when... If the above conditions are not met, an alternative solution, the "proportional scaling method," can be used: maintaining the original step size ratio unchanged. Multiply the entire pre-transition sequence by a scaling factor A new sequence is obtained. This scheme strictly guarantees monotonicity and is suitable for scenarios where the first step is not sensitive to minor increases.

[0060] The following comparison uses specific data to illustrate that the method of this invention is superior to the method using a fixed step size.

[0061] For simplicity, this embodiment uses a rectangular pulse to represent the specific form of the pulse excitation source (the amplitude should be set according to the specific application scenario; here, an amplitude of 1 has no specific physical meaning). This embodiment considers the case of a one-to-one correspondence between the excitation source and the excited object, and takes a period multiple factor. For multi-objective time-sharing stimulus scenarios, simply set the parameters according to the actual number of objects. The value is the same, and the rest of the steps are exactly the same.

[0062] 1. Calculation of basic parameters With operating frequency Duty cycle For example, pulse period .

[0063] Pulse width:

[0064] Set the minimum number of steps within the pulse region , get the fine step length:

[0065] The middle of the relaxation period is used to determine the baseline number of steps for determining the coarse step size. , thus obtaining the rough step length:

[0066] 2. Post-pulse transition region (exponential growth) Take growth rate ,common ratio .from Begin with a geometric progression until the step size reaches [a certain value]. .ratio ,Depend on Solving (The first 14 steps are proportional, and the 15th step is a forced selection) Total time for the first 14 steps:

[0067] Add step 15 Total duration of the post-transition zone End time of this region:

[0068] 3. Pre-pulse transition region (exponential decay) Take the attenuation factor .from Start by multiplying by 0.3, until the step size is less than or equal to 0.3. The step size sequence is:

[0069] The fourth step is already less than Therefore, take Step 1, the first three steps are geometric, the fourth step is forced to be... Total time for the first three steps:

[0070] Add the fourth step The original total duration of the transition zone The start time of this region (before adjustment):

[0071] 4. Processing of the middle and remaining time of the relaxation period Theoretical duration of the middle phase of relaxation period:

[0072] With fixed step size Progression, theoretical number of steps:

[0073] Rounding up gives the actual number of steps in the middle section. The first 12 steps all use Total duration:

[0074] time left:

[0075] 5. One-step insertion method to adjust the transition zone before adjustment The original pre-transition step size sequence (exponential decay, 4 steps) is as follows:

[0076] Total duration of the original sequence According to the one-step insertion method, a step size is inserted at the beginning of the pre-transition zone and adjusted appropriately to absorb the remaining time. .

[0077] Calculate the insertion step size:

[0078] Constructing a new sequence: taking the original sequence Replace with two consecutive ones And retain the subsequent step size, that is, the new sequence is:

[0079] Substitute the values:

[0080] Verify the total duration of the new sequence:

[0081] This value is exactly equal to the total duration of the original transition zone plus the remaining time:

[0082] Step size variation analysis: From the last step in the middle section (step length) Step 1 to the new sequence The step size decreased slightly by 0.55%, without any increase. The second step is the same as the first step, with the same step size; From the second step to Step length from Down to The decrease was similar to that of the original sequence. arrive The decline was consistent; Subsequent step sizes continue to decrease according to the original decay factor, eventually smoothly transitioning to finer step sizes. .

[0083] As can be seen, the new sequence maintains the characteristic of monotonically decreasing step size, and the maximum decrease is the same as the original design, ensuring numerical stability and precise period closure.

[0084] 6. Total number of steps in a cycle Pulse zone: 20 steps; Post-pulse transition zone: 15 steps; Mid-relaxation period: 12 steps; Pre-pulse transition zone (after adjustment): 5 steps; Total: 52 steps.

[0085] 7. Efficiency Comparison Two cycles (0.004 s) using this method require Instead of using a fixed fine step size, a fixed fine step size is adopted. need This step improves computational efficiency by approximately 7.7 times, for example... Figure 3 As shown. Meanwhile, by setting a smooth exponential transition region and employing a one-step insertion method to precisely close the period, numerical oscillations caused by sudden changes in step size are completely avoided.

[0086] Figure 4 The step size variation patterns of exponential transition (solid line) and linear transition (dashed line) are illustrated. The exponential transition changes more slowly in the initial stage and more rapidly in the later stage, which is more conducive to suppressing numerical oscillations and is the preferred core implementation method of this invention; the linear transition can be used as an alternative, but its smoothness is slightly inferior to that of the exponential transition.

[0087] Compared with existing technologies, the adaptive time step control method for periodic pulse excitation simulation provided in this embodiment has the following significant advantages: 1. Eliminating numerical oscillations: By setting a smooth transition region, the time step is continuously varied between the pulse region and the relaxation region, avoiding the surge in truncation error caused by abrupt changes in the step size and the excitation of non-physical high-frequency modes. This completely eliminates numerical oscillations caused by abrupt changes in the step size, ensuring the physical authenticity and numerical stability of the simulation solution.

[0088] 2. Significantly Reduced Computational Steps: By dividing the pulse period into a pulse region, a post-transition region, a mid-relaxation period, and a pre-transition region, and employing a smooth step-size transition strategy, this method significantly reduces the number of computational steps during the relaxation phase while maintaining computational accuracy throughout the pulse period. Compared to traditional methods that use fixed, fine step sizes throughout, this method typically reduces the total number of computational steps by more than an order of magnitude, and the smaller the ratio of pulse width to period (i.e., the smaller the duty cycle), the more significant the efficiency improvement. Significant efficiency gains can be achieved regardless of whether exponential or linear transitions are used, or whether the design is symmetric or asymmetric.

[0089] 3. Wide applicability: This method does not depend on specific physical fields or engineering backgrounds and can be applied to numerical simulations of various periodic pulse excitation problems in thermal, acoustic, and structural mechanics. Users only need to set basic parameters such as pulse frequency and duty cycle according to actual working conditions to integrate this method into their existing simulation workflow.

[0090] 4. Multiple implementation methods: It covers various transition laws such as exponential and linear, and supports symmetric / asymmetric configurations, making it easy to select the appropriate strategy according to the specific problem.

[0091] 5. Easy to integrate: It can be seamlessly integrated into mainstream commercial software (such as ANSYS Fluent, LS-DYNA, Abaqus, etc.) through user-defined functions without modifying the software kernel.

[0092] Example 2, Another embodiment of the present invention discloses an adaptive time step control device for periodic pulse excitation simulation, thereby implementing the adaptive time step control method for periodic pulse excitation simulation in Embodiment 1. The specific implementation of each module is described in the corresponding description in Embodiment 1, including: The data acquisition module is used to obtain the operating frequency, duty cycle and period multiple factor of the pulse excitation source used for simulation, the minimum number of steps in the pulse excitation region, the reference number of steps in the relaxation period, the step size growth parameter of the post-pulse transition region, and the step size decay parameter of the pre-pulse transition region. Among them, a pulse cycle is divided into the pulse excitation region, the post-pulse transition region, the relaxation period, and the pre-pulse transition region in the time domain. The step size calculation module is used to determine the pulse period and pulse width based on the operating frequency, the duty cycle, and the period multiplier factor; determine the fine step size of the pulse excitation region based on the pulse width and the minimum number of steps; determine the coarse step size of the relaxation period based on the pulse period, the pulse width, and the reference number of steps; calculate each step size based on the step size growth parameter starting from the fine step size, until it grows to the coarse step size, to obtain the step size sequence of the post-pulse transition region; and calculate each step size based on the step size decay parameter starting from the coarse step size, until it decays to the fine step size, to obtain the step size sequence of the pre-pulse transition region.

[0093] In some embodiments, a sequence adjustment module is further included, for: The total duration of the post-pulse transition region is determined based on the step size sequence of the post-pulse transition region. The total duration of the pre-pulse transition region is determined based on the step size sequence of the pre-pulse transition region. The total duration of the relaxation period is determined based on the pulse period, the pulse width, the total duration of the post-pulse transition region, and the total duration of the pre-pulse transition region. Determine whether the total duration of the relaxation period is an integer multiple of the coarse step size of the relaxation period. If it is not an integer multiple, determine the remaining time between the total duration of the relaxation period and an integer multiple of the coarse step size of the relaxation period. The step size sequence of the pre-pulse transition region is adjusted based on the remaining time.

[0094] Compared with the prior art, the adaptive time step control device for periodic pulse excitation simulation provided in this embodiment has the following significant advantages: 1. Eliminating numerical oscillations: By setting a smooth transition region, the time step is continuously varied between the pulse region and the relaxation region, avoiding the surge in truncation error caused by abrupt changes in the step size and the excitation of non-physical high-frequency modes. This completely eliminates numerical oscillations caused by abrupt changes in the step size, ensuring the physical authenticity and numerical stability of the simulation solution.

[0095] 2. Significantly Reduced Computational Steps: By dividing the pulse period into a pulse region, a post-transition region, a mid-relaxation period, and a pre-transition region, and employing a smooth step-size transition strategy, this method significantly reduces the number of computational steps during the relaxation phase while maintaining computational accuracy throughout the pulse period. Compared to traditional methods that use fixed, fine step sizes throughout, this method typically reduces the total number of computational steps by more than an order of magnitude, and the smaller the ratio of pulse width to period (i.e., the smaller the duty cycle), the more significant the efficiency improvement. Significant efficiency gains can be achieved regardless of whether exponential or linear transitions are used, or whether the design is symmetric or asymmetric.

[0096] 3. Wide applicability: This method does not depend on specific physical fields or engineering backgrounds and can be applied to numerical simulations of various periodic pulse excitation problems in thermal, acoustic, and structural mechanics. Users only need to set basic parameters such as pulse frequency and duty cycle according to actual working conditions to integrate this method into their existing simulation workflow.

[0097] 4. Multiple implementation methods: It covers various transition laws such as exponential and linear, and supports symmetric / asymmetric configurations, making it easy to select the appropriate strategy according to the specific problem.

[0098] 5. Easy to integrate: It can be seamlessly integrated into mainstream commercial software (such as ANSYS Fluent, LS-DYNA, Abaqus, etc.) through user-defined functions without modifying the software kernel.

[0099] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the corresponding computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0100] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive time step control method for simulation of periodic pulse excitation, characterized in that, Includes the following steps: The operating frequency, duty cycle, and period factor of the pulse excitation source used for simulation are obtained, and the pulse period and pulse width are determined based on the operating frequency, the duty cycle, and the period factor; wherein, a pulse period is divided in the time domain into a pulse excitation region, a post-pulse transition region, a relaxation period mid-segment, and a pre-pulse transition region. Obtain the minimum number of steps in the pulse excitation region, and determine the fine step size of the pulse excitation region based on the pulse width and the minimum number of steps; Obtain the reference number of the relaxation period mid-segment, and determine the coarse step size of the relaxation period mid-segment based on the pulse period, the pulse width, and the reference number of steps; Obtain the step size growth parameter of the post-pulse transition region, and starting from the fine step size, calculate the step size of each step according to the step size growth parameter until it grows to the coarse step size, thereby obtaining the step size sequence of the post-pulse transition region. Obtain the step size attenuation parameter of the pre-pulse transition region. Starting from the coarse step size, calculate the step size of each step according to the step size attenuation parameter until it attenuates to the fine step size, thereby obtaining the step size sequence of the pre-pulse transition region.

2. The adaptive time step control method for periodic pulse excitation simulation according to claim 1, characterized in that, The step size growth parameter is the step size growth rate; The step size decay parameter is the step size decay factor; Starting from the fine step size, each step size is calculated based on the step size growth parameters, including: The common ratio of step size growth in the exponential mode is determined based on the step size growth rate. Starting from the fine step size, each step size is calculated based on the common ratio of step size growth according to the exponential law. Starting from the coarse step size, the step size for each step is calculated based on the step size decay parameter, including: The step size decay factor is used as the common ratio of step size decay in the exponential mode, and the step size is calculated based on the common ratio of step size decay according to the exponential law.

3. The adaptive time step control method for periodic pulse excitation simulation according to claim 2, characterized in that, The step size growth rate is less than or equal to the reciprocal of the step size decay factor minus 1.

4. The adaptive time step control method for periodic pulse excitation simulation according to claim 1, characterized in that, The step size growth parameter is the expected number of growth steps in linear mode, and the step size decay parameter is the expected number of reduction steps in linear mode; Starting from the fine step size, each step size is calculated based on the step size growth parameters, including: The step size increment is determined based on the expected number of growth steps, the fine step size, and the coarse step size; Starting from the fine step size, each step size is calculated based on the step size increment according to a linear law; Starting from the coarse step size, the step size for each step is calculated based on the step size decay parameter, including: The step size reduction is determined based on the desired reduction in steps, the fine step size, and the coarse step size; Starting from the coarse step size, each step size is calculated by decreasing the step size according to a linear law.

5. The adaptive time step control method for periodic pulse excitation simulation according to claim 1, characterized in that, Also includes: The total duration of the post-pulse transition region is determined based on the step size sequence of the post-pulse transition region. The total duration of the pre-pulse transition region is determined based on the step size sequence of the pre-pulse transition region. The total duration of the relaxation period is determined based on the pulse period, the pulse width, the total duration of the post-pulse transition region, and the total duration of the pre-pulse transition region. Determine whether the total duration of the relaxation period is an integer multiple of the coarse step size of the relaxation period. If it is not an integer multiple, determine the remaining time between the total duration of the relaxation period and an integer multiple of the coarse step size of the relaxation period. The step size sequence of the pre-pulse transition region is adjusted based on the remaining time.

6. The adaptive time step control method for periodic pulse excitation simulation according to claim 5, characterized in that, Adjusting the step size sequence of the pre-pulse transition region based on the remaining time includes: The adjustment term is determined based on the first term of the step size sequence of the pre-pulse transition region and the remaining time. The adjustment term is added before the first term of the step size sequence in the pre-pulse transition region; Replace the first term of the step size sequence in the pre-pulse transition region with the aforementioned adjustment term; Replace the step size sequence of the pre-pulse transition region with the adjusted step size sequence described above.

7. The adaptive time step control method for periodic pulse excitation simulation according to claim 6, characterized in that, The step of determining the adjustment term based on the first term of the step size sequence of the pre-pulse transition region and the remaining time includes: The adjustment term is the sum of the first term of the step size sequence of the pre-pulse transition region and half of the remaining time.

8. The adaptive time step control method for periodic pulse excitation simulation according to claim 5, characterized in that, The step size sequence of the pre-pulse transition region is adjusted according to the remaining time, including: The scaling factor is determined based on the total duration of the pre-pulse transition region and the remaining time. The adjusted step sequence is obtained by multiplying the scaling factor by the step sequence of the pre-pulse transition region.

9. The adaptive time step control method for periodic pulse excitation simulation according to claim 8, characterized in that, The step of determining the scaling factor based on the total duration of the pre-pulse transition region and the remaining time includes: The ratio of the sum of the total duration of the pre-pulse transition region and the remaining time to the total duration of the pre-pulse transition region is used as a scaling factor.

10. The adaptive time step control method for periodic pulse excitation simulation according to claim 1, characterized in that, When the growth exceeds the coarse step size, the final step takes the coarse step size; when the decay falls below the fine step size, the final step takes the fine step size.

Images