Numerical simulation method for heat-fluid-solid multi-field coupling response characteristics

Abstract

The invention discloses a numerical simulation method for a heat-fluid-solid multi-field coupling response characteristic, which relates to the technical field of mechanics and engineering simulation and comprises the following steps of: constructing an asynchronous sampling array, performing joint time-frequency acquisition on a periodic pressure pulsation signal of a fluid and a dynamic response signal of a solid structure, extracting a coupling phase fingerprint baseline, and calculating the coupling phase fingerprint baseline. As a unified reference for energy analysis; and performing coherent decomposition based on the coupling phase fingerprint baseline, calculating a phase-locked energy index, identifying an instantaneous resonance window in a dynamic evolution process, and determining a dangerous frequency band and an interface fragile area. According to the method, unified reference of heat-fluid-solid response data is realized by constructing a coupling phase fingerprint baseline; predicting a crack initiation threshold in combination with anti-fact playback and a risk cone; extracting a minimum disturbance set to construct an energy dissipation channel; time reversal micropulses are applied to cooperate with phonon band gaps and shadows for energy storage, phase lock release and vibration extinguishing control are completed, and closed-loop regulation and control of structure safety are achieved.

Description

Technical Field

[0001] This invention relates to the field of mechanics and engineering simulation technology, specifically to a numerical simulation method for the multi-field coupled response characteristics of thermo-fluid-solid systems. Background Technology

[0002] "Numerical simulation of the coupled response characteristics of thermal-fluid-solid systems" refers to the systematic calculation and prediction of the interaction process between heat transfer (thermal field), fluid motion (flow field), and solid structure deformation or stress distribution (solid field) through mathematical modeling and computer simulation. Since these three types of physical fields often coexist and influence each other—for example, fluid flow brings heat transfer effects, temperature changes cause thermal expansion and contraction of solids, and solid deformation, in turn, alters fluid channels and heat flow distribution—a coupled numerical method must be used to capture the dynamic response laws of these multi-physics fields. This method can reveal the evolutionary behavior of materials or structures under complex working conditions under the combined effects of thermal shock, fluid shock, and mechanical loads, providing a scientific basis for engineering design optimization, reliability assessment, and safety prediction in extreme environments.

[0003] The existing technology has the following shortcomings: In existing technologies, periodic pressure pulsations are easily generated when fluids operate under unstable conditions. When the instantaneous frequency of these pressure pulsations coincides with the natural vibration frequency of the solid structure, an abnormal superposition effect of energy occurs at the fluid-solid interface. This effect not only causes local stress to amplify exponentially in a very short time but also disrupts the stability of the interface, leading to instantaneous delamination of the interface layer and further causing crack propagation or even fracture failure in the overall structure. Such problems are difficult to predict and effectively suppress in existing technologies, and once they occur, they pose a serious threat to the safety and reliability of the structure.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a numerical simulation method for the multi-field coupling response characteristics of thermal fluid-solid systems, so as to solve the problems in the background art mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems, comprising the following steps: An asynchronous sampling array is constructed to jointly acquire the periodic pressure pulsation signal of the fluid and the dynamic response signal of the solid structure in time and frequency, and extract the coupled phase fingerprint baseline as a unified reference for energy analysis; Coherent decomposition is performed based on the coupled phase fingerprint baseline, the phase-locked energy index is calculated, the instantaneous resonance window in its dynamic evolution process is identified, and the dangerous frequency band and vulnerable interface region are determined. Based on the dangerous frequency bands and vulnerable areas of the interface, a counterfactual replay chain is constructed to deduce the crack initiation threshold of the key pulse sequence and generate a real-time risk cone; The boundary conditions are reconstructed online based on the real-time risk cone, and the fluid channel morphology is corrected by phase staggered peak probing to obtain the minimum perturbation set that interrupts the coupling process; Based on the minimum perturbation set, an energy dissipation channel containing time-varying virtual impedance and adjustable surface damping is constructed, and the energy suppression parameters and execution time window are output. Within the time window, time-reversal micropulses are applied according to the parameters to drive the programmable phonon bandgap structure and superimpose the shadow energy storage unit, thereby releasing the phase-locked state, completing the energy extinction, and realizing closed-loop control of the coupled response.

[0007] Preferably, the coupled phase fingerprint baseline extraction steps are as follows: Distributed acquisition channels are set up in the structural interface area to collect fluid pressure pulsation signals and dynamic response signals of solid structures; Based on the frequency domain characteristics of each signal, an asynchronous sampling frequency is set and a sliding window envelope analysis is performed to extract energy density, frequency dominant components and modulation ratio; A dynamic alignment strategy is used to rearrange the main axis of the signal window and construct a time mapping matrix to complete time-frequency alignment. The relative phase evolution trajectory in the coupled signal is extracted, the frequency coupling density and phase jump interval are calculated, the continuous path with the highest phase stability is selected, and the coupled phase fingerprint baseline is generated as a unified reference frame architecture.

[0008] Preferably, the process for determining the dangerous frequency band and the vulnerable area of ​​the interface is as follows: Within the target observation period, the fluid pressure signal and the solid structure response signal are segmented and processed, and a coupled propagation trajectory map is generated by phase alignment. Signal clusters are identified in the coupled propagation trajectory map and a phase-locked signal spectrum is constructed. The original sequence of the phase-locked energy index is calculated. Dynamic trend extraction is performed on the original sequence of phase-locked energy index to identify instantaneous resonance windows; The dominant frequency is extracted from the instantaneous resonance window and a dangerous frequency band is formed. The vulnerable area of ​​the structural interface is determined by combining the structural strain response.

[0009] Preferably, the real-time risk cone generation steps are as follows: Key pulse sequences were extracted from dangerous frequency bands and vulnerable regions of structural interfaces, and a pulse-response mapping table was constructed. The key pulse sequence is replaced with multiple perturbation versions, and phase interpolation is performed on the reconstructed signal to generate replacement paths and classify and construct path influence matrices. Based on the path influence matrix, crack initiation path characteristics are aggregated to form a crack initiation threshold and to deduce a crack propagation trend map. A real-time risk cone is generated based on the crack propagation trend map to characterize the distribution of future structural response risk.

[0010] Preferably, the steps for obtaining the minimum perturbation set are as follows: The boundary domain is extracted based on the characteristics of phase-locked energy amplification, frequency concentration density, and strain rate increase in the real-time risk cone. A multi-level perturbation sample set was constructed around the boundary response parameters, and perturbation response features were collected; Candidate parameters for decoupling are selected based on phase misalignment, frequency variation, and coupling delay phenomena in the disturbance response. The channel morphology shift was analyzed by combining particle trajectory reconstruction and strain-thermal coupling features were extracted. The intervention efficiency of all effective perturbation combinations is evaluated, and the minimum perturbation set is finally determined.

[0011] Preferably, the energy suppression parameters and execution time window output steps are as follows: Intervention conditions are extracted based on the minimum perturbation set and mapped to the interface region to construct a response network comprising an electrically induced flexible structure layer, a charge-sensing thin film, and a temperature-sensitive damping layer. The response elements were activated sequentially using the intervention parameters, and three feedback indicators were collected: energy absorption rate per unit area, structural instantaneous response frequency change curve, and fluid shear force disturbance attenuation rate. After the energy dissipation path is formed, the execution time window and intervention parameter table consisting of the response path length, energy dissipation delay and structural vibration recovery time are output.

[0012] Preferably, the intervention parameter table includes the voltage change curve of the electrically excited flexible structure layer, the temperature control gradient of the temperature-sensitive damping layer and the correction value of the structural flexibility, and each parameter corresponds one-to-one with the position of the corresponding response element, so as to realize the precise dynamic control of the energy dissipation channel.

[0013] Preferably, within the time window, a time-reversal micropulse is applied according to the parameters to drive the programmable phonon bandgap structure and superimpose a shadow energy storage unit. The steps to release the phase-locked state are as follows: Based on the combination of intervention parameters and the execution time window, time-reversed micropulses are applied at the thermal excitation hysteresis point, the shear stress convergence point, and the strain feedback delay node; After micropulse application, the programmable phonon bandgap structure is activated to adjust the lattice spacing, thereby forming frequency-selective reflection; Subsequently, the shadow energy storage unit is activated in the stress convergence area to absorb the residual energy, completing the phase lock release and energy extinguishing process.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention, by introducing a coupled phase fingerprint baseline, achieves for the first time a unified reference framework for the synchronous operation of thermal, flow, and solid-state response data. By constructing a counterfactual playback chain and a real-time risk cone, it enables threshold prediction and vulnerability assessment before crack initiation. Furthermore, by extracting the minimum disturbance set and constructing a variable impedance-damping coordinated energy dissipation channel, it endows the system with low-energy, high-efficiency local control capabilities. Finally, it utilizes time-reversal micropulses and phonon bandgap structures to achieve phase-lock release, and recovers energy through a shadow energy storage unit, completing rapid oscillation extinguishing and system steady-state recovery. This closed-loop optimization process has significant advantages such as fast response speed, high intervention accuracy, low energy loss, and adaptability to complex operating conditions, significantly improving the safety and reliability of the structure in unstable thermal-fluid-solid coupling environments. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0016] Figure 1 This is a flowchart of the numerical simulation method for the thermal-fluid-solid multi-field coupling response characteristics of the present invention. Detailed Implementation

[0017] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0018] This invention provides, for example Figure 1 The numerical simulation method for the thermo-fluid-solid multi-field coupled response characteristics shown includes the following steps: An asynchronous sampling array is constructed to jointly acquire time-frequency data of the periodic pressure pulsation signal generated by the fluid and the dynamic response signal of the solid structure. The coupled phase fingerprint baseline with characteristic correlation is extracted from the acquisition results to unify the reference frame architecture for subsequent energy analysis. To accurately capture the multi-source response characteristics during the thermo-fluid-structure interaction process, an asynchronous sampling array must first be constructed to jointly acquire time-frequency data of the fluid excitation and structural response signals, and then a representative coupling fingerprint baseline is extracted based on the phase relationship with high coupling degree. The specific implementation process is as follows: Multiple distributed acquisition channels are pre-set in the key interface areas of the structure under test to collect fluid pressure pulsation signals and dynamic response signals of the solid structure. Within the fluid channels, regions with significant velocity gradient changes, such as the inlet section of bends, the leading edge of throttling sections, and the trailing edge of diffusers, are selected for deployment of charge-output piezoelectric ceramic pressure sensors. These sensors have a frequency response bandwidth of at least 100 kHz and millibar-level pressure resolution. On the surface and inside the solid structure, distributed fiber optic grating arrays are embedded along structural connectors, edge cantilever areas, and interface composite material layers. These sensors measure strain changes in real time by demodulating the reflected wavelength drift, with a measurement bandwidth of at least 10 kHz and a spatial resolution better than 5 mm. All sensors are configured with different sampling frequencies and time bases according to their respective response speeds and expected frequency ranges, forming an asynchronous sampling acquisition scheme. Pressure acquisition operates at a 1 microsecond time resolution, while structural strain acquisition operates at a 10 microsecond time resolution, avoiding the overlap and weakening of characteristic information caused by response delay differences in conventional synchronous sampling.

[0019] Sliding window envelope analysis was performed on the acquired fluid and structural signals respectively. A non-overlapping Gaussian window function was used for initial segmentation, with window lengths set to 2 milliseconds and 4 milliseconds to accommodate the frequency characteristics of the fluid pulsation and structural vibration signals. Within each signal window, the short-time energy density, dominant frequency component, and modulation ratio variation amplitude of the waveform were calculated, generating fluid pressure feature vectors and structural response feature vector sets. Then, a dynamic alignment strategy was employed, rearranging the principal axes of the sliding window based on the position of the maximum peak of the waveform envelope to ensure consistent approximate time positioning of the same physical event in both signal sets. A dual-channel time mapping matrix was constructed by calculating the relative time deviations between energy transition points, frequency drift points, and phase abrupt change points within the sliding window. Based on this, time-frequency alignment between the two signal sources was completed according to the path that minimizes mismatch error in the mapping matrix, eliminating timing deviations caused by asynchronous sampling and achieving true synchronization under nonlinear physical processes.

[0020] Based on the time-frequency aligned signal sequence, energy-dominant segments are extracted from multiple consecutive action cycles, and stable causal relationships between frequency transition points in the fluid excitation signal and phase transition points in the structural response signal are identified. A joint phase curve spectrum is constructed between each pair of transition events, and the phase difference between each pair is standardized to obtain the relative phase evolution trajectory. Within multiple cycles, a set of trajectories with high overlap, periodic phase stability greater than 95%, and relative phase offset less than 3 degrees is selected. This set is sequence-calibrated, and the frequency coupling density and phase jump interval between adjacent signal segments are calculated, thereby constructing a coupling stability matrix. Furthermore, the continuous phase path with the highest stability score is selected from this matrix as a candidate coupling baseline. All candidate baseline segments are compared across cycles; if the phase difference change does not exceed a set tolerance range (e.g., ±2 degrees) within at least five cycles, the segment is confirmed as the final coupling phase fingerprint baseline.

[0021] Using this coupled phase fingerprint baseline as a unified reference frame architecture, dynamic mapping and comparison are performed with sampling data from all subsequent time periods. This extracts the relative phase drift, frequency deviation, and energy increment trends of new signal segments under the baseline coordinates, and establishes a three-dimensional coupled phase-frequency-energy relationship graph. This graph serves not only as the standard input for subsequent phase-locked loop energy index calculations but also as the fundamental scale for identifying resonance windows and constructing risk prediction models. By using this baseline as a reference, all newly sampled data will be accurately normalized to a reference system with physical consistency and temporal comparability, effectively avoiding feature mismatch problems caused by dynamic changes in the coupling state.

[0022] Coherent decomposition calculations are performed based on the coupled phase fingerprint baseline to obtain the phase-locked signal composition and corresponding energy distribution of the coupled system, calculate the phase-locked energy index, and identify the instantaneous resonance window during the dynamic change of the index in order to determine the dangerous frequency band and the vulnerable area of ​​the structural interface in the frequency concentration region. To identify anomalous energy accumulation behavior in thermo-fluid-structure interaction (TFS) structures, coherent decomposition and energy distribution analysis are performed based on the extracted coupled phase fingerprint baseline. This constructs a phase-locked energy index and identifies resonance-sensitive regions. The specific implementation process is as follows: The obtained coupled phase fingerprint baseline is used as a standard reference frame and applied to the joint signal sequence within the target observation time period. The joint signal includes real-time signals from pressure changes in the fluid channel and real-time signals from strain response in the solid structure. Each set of joint signals is segmented, with each segment being 5 milliseconds long and a sliding step of 1 millisecond to ensure that phase collapse does not occur within the resolution time. Within each time period, the main peak positions of the fluid signal and the structural signal are aligned based on the reference phase point recorded in the coupled phase fingerprint baseline, thereby obtaining a synchronization mapping based on the phase trajectory. Each synchronized mapping signal is constructed into a two-dimensional matrix, where the horizontal axis represents the actual time series and the vertical axis represents the relative phase offset angle. Each element in the matrix represents the relative phase relationship between the fluid excitation and the structural response at a certain moment. By visualizing the matrix data, a coupled propagation trajectory map is generated, clearly presenting the phase synchronization, phase jump, and phase hysteresis phenomena between signals, serving as the input data basis for coherent decomposition.

[0023] Based on the constructed coupled propagation trajectory map, multi-channel coherent decomposition was performed to identify signal clusters with fixed phase differences, high frequency stability, and concentrated energy density. First, signal sequences with phase shifts not exceeding 2 degrees and signal shape similarity exceeding 95% across multiple consecutive periods were selected as candidate phase-locked signal clusters. The total energy, dominant frequency, duration, and phase consistency of these candidate clusters were extracted, and based on these, multiple signal clusters were classified into three categories: frequency-guided, energy-burst, and phase-stabilized. Within each category, the clusters were sorted by energy proportion, and the phase-locked signal spectrum was reconstructed by superimposing the top-ranked clusters. This phase-locked signal spectrum reflects stable propagation paths and high-energy channels in the coupled signal, exhibiting extremely high time-frequency coupling characteristic performance. Further, the ratio of the total energy proportion of the phase-locked signal spectrum to the background signal energy was used as a normalization input to construct the original sequence of the phase-locked energy index for each time period. This original sequence reflects the degree of synergistic energy enhancement between structural response and fluid excitation in different time periods, providing an energy-level basis for subsequent risk area identification.

[0024] The original sequence of the constructed phase-locked energy index is smoothed and its dynamic trend is extracted to form a phase-locked energy trajectory curve that evolves over time. An evolution window is set at 100 milliseconds, and within each window, the phase-locked energy growth rate, the improvement in phase synchronization level, and the frequency concentration trend are calculated. If, within three consecutive evolution windows, the phase-locked energy growth rate is more than twice its past average, the phase synchronization index shows a gradient increasing trend, and the frequency concentration range is less than 30% of the previous average level, then the time range corresponding to that window is determined to be an instantaneous resonance window. The instantaneous resonance window is the critical stage where the structure is about to undergo a resonance response under thermo-fluid-structure interaction excitation. Its physical manifestation is that energy accumulates rapidly in time, the phase is highly locked in space, while the structural impedance has not yet completed adaptive adjustment, making it extremely prone to local stress concentration and failure.

[0025] Using the instantaneous resonance window as the core region, the dominant frequency components of all phase-locked signal clusters are extracted within this time period, and their concentration, drift rate, and energy dominance are statistically analyzed along the frequency dimension. Frequency values ​​with a drift rate less than 0.5 Hz per second, a duration exceeding 200 milliseconds, and an energy density accounting for more than 30% of the total energy are selected to construct a high-risk frequency list. Frequency ranges with an adjacent frequency difference of no more than 1 Hz are merged into a continuous frequency band, forming a danger zone. In the structural response data, the signal amplitude growth curves of each strain sensing point within the resonance window are extracted, and the consistency between the strain growth rate and frequency response is verified. If the strain growth rate of a measuring point during the danger zone exceeds twice the average rate of its approximate region, and its response frequency overlaps with the danger zone, then the area where that point is located is identified as a vulnerable area of ​​the structural interface. Finally, a corresponding mapping relationship between the danger zone and the vulnerable area of ​​the structural interface is constructed, providing two-dimensional and highly accurate target input conditions for subsequent counterfactual playback, dynamic risk cone generation, and energy intervention schemes.

[0026] Based on the dangerous frequency band and the vulnerable area of ​​the interface, a counterfactual replay chain is constructed. The key pulse sequence is deduced by replacement and reconstruction methods to identify the threshold conditions for crack initiation, and then a real-time risk cone describing the future structural risk evolution trend is generated. To achieve accurate prediction of future structural risks, a counterfactual replay chain is constructed based on identified hazardous frequency bands and vulnerable areas at structural interfaces. This chain replaces and reconstructs key pulse sequences, extrapolates crack initiation thresholds, and generates a real-time updated risk cone. The specific implementation is as follows: Multiple signal segments with continuous frequency energy accumulation were extracted within the identified hazardous frequency bands, and corresponding structural response signal sequences were selected. Each pair of signal sequences needed to cover at least one complete energy surge cycle, with a time span controlled between 200 and 300 milliseconds. For the fluid excitation signal, the peak position, pulse duration, frequency transition boundary, and amplitude change slope were extracted and compared point-by-point with the maximum strain position, rate of change, and phase lag region in the structural strain signal of the same time period. Signal combinations with peak phase alignment and corresponding structural responses exhibiting drastic strain changes within 3 milliseconds were selected as key pulse sequences, and a pulse-response mapping table was constructed. Subsequently, multiple perturbation versions were designed for these sequences. Each perturbation version changed only one excitation parameter, such as reducing the fluid excitation amplitude to 60% of its original value, delaying the peak time by 2 milliseconds, lengthening the rise edge to half of the original slope, or inserting additional frequency perturbations to disrupt the original rhythm, ensuring that the perturbation was individually controllable and physically feasible. All perturbation versions were combined into a counterfactual intervention candidate pool for the next simulation replacement.

[0027] The original key pulses are replaced one by one with counterfactual candidate pulse sequences, and the combined fluid excitation and structural response sequences after replacement are synchronously reconstructed. To ensure the continuity between the replaced signal and the original coupling trajectory, 20 milliseconds of original data are taken before and after the replacement point for phase interpolation transition processing to ensure no abrupt changes in phase connection. The constructed new signal path is mapped to the established coupling propagation trajectory map, and the retention of coupling characteristics after replacement is evaluated based on the phase consistency index. In paths that meet the phase matching requirements, the phase-locked energy value, structural strain growth rate, and phase stability within 100 milliseconds after the replacement point are extracted. If a replacement path shows a significant increase in phase-locked energy (more than 30% higher than the original path), a significant increase in structural strain rate (more than twice that of the original path), and accelerated convergence of phase perturbation during this time period, the path is marked as a crack-inducing path. Conversely, if the phase-locked energy decreases by more than 30% of the original path, the number of phase change points increases, and the structural response is stable, it is classified as a crack-prolonging path. All replacement paths are classified, and a path influence matrix is ​​constructed.

[0028] Based on the path influence matrix, threshold features are aggregated for all crack initiation paths to derive the boundary conditions for crack initiation. The aggregation targets include: the minimum phase-locked energy value triggering the crack, the location of the maximum frequency density, the critical range of phase stability, and the critical slope of the strain response. Specifically, the location where the structural response first jumps in each crack initiation path is selected, and its corresponding four indicators are statistically analyzed to form a set of critical state vectors. All vectors are clustered according to Euclidean distance, and the cluster centers are extracted as representative points for crack initiation thresholds. Subsequently, the response data of the original structural state in the future timeline is scanned point by point. A state vector is constructed every 50 milliseconds, and the distance to each critical point is calculated. The time and location where the distance is less than the set threshold are recorded, which are then identified as potential crack initiation points. These initiation points form an evolution trajectory, which, based on the growth direction of the coupling energy, the frequency convergence trend, and the response delay change, is expanded into a multi-path crack propagation trend map for subsequent risk prediction modeling.

[0029] A three-dimensional spatial risk cone is constructed based on the crack propagation trend map to describe the future response risk distribution of the structure under multi-field coupling in real time. The base of the cone represents the current time-frequency-spatial state of the structure, and the apex projects all possible crack propagation paths within the next 200 to 1000 milliseconds. Each propagation path is assigned a risk weight, and the risk level is calculated based on its phase-locked energy peak, propagation rate, path length, and degree of overlap with the vulnerable areas of the structural interface. The risk level is divided into five levels and represented by color segments: light blue indicates structural stability, yellow indicates that the frequency has entered the sensitive zone, orange indicates that the strain response has been activated but not yet fractured, red indicates that the structural boundary is unstable under critical conditions, and dark red indicates that there is a high probability of crack initiation. The risk cone is updated in real time every 50 milliseconds, synchronized with the coupling sampling, to achieve a complete closed loop between response prediction and risk assessment.

[0030] Based on the output of the real-time risk cone, the boundary conditions of the coupled system are reconstructed online, and the dynamic correction of the fluid channel geometry prediction is completed by staggering the phase response characteristics. The set of minimum disturbances that can interrupt the coupling process is extracted. To block the predicted coupling risk paths, boundary conditions should be reconstructed online based on real-time risk cone results. The fluid channel morphology should be precisely corrected through phase response staggered peak testing, and finally, the set of minimum perturbations that can interrupt the coupling process should be extracted. The specific process is as follows: Based on the structural regions corresponding to high-risk paths in the real-time risk cone, three core predictive indicators of the current structural state are extracted: the first is the time points and spatial distribution of phase-locked energy increases greater than 30%; the second is regions where the frequency concentration density is more than 50% higher than the average within the danger band; and the third is local response points with abnormal changes in structural strain rate. Combining these three indicators, the boundary action domain is located, and boundary response characteristics are extracted at three locations: the boundary stress contact surface, the channel inlet and outlet edges, and the structural transition section connection surface. These characteristics include unit heat flux, shear stress direction, and the rate of change of solid constraint stiffness. Within this spatial region, the time series of response parameters at each location is compared with historical coupling behavior data to select the most sensitive parameter direction, forming an initial set of intervention directions, which serves as the input foundation for subsequent online boundary reconstruction.

[0031] A sample set of slightly perturbable boundary conditions is constructed around the set of intervention directions. Within each intervention direction, at least five perturbation levels are set, with amplitudes controlled between 5% and 20% of the original boundary parameter values. Specifically, these include: boundary heat flux perturbation achieved through local heating or cooling, adjusted by constant power injection or thermal conductivity material bonding; boundary compliance perturbation achieved through micro-displacement loading at support points, adjusted by stepper motor-driven loading plate movement; and fluid inlet / outlet perturbation implemented using micro-flow orifice adjustment plates, adjusting the local channel cross-section through thickness gradient changes. After each perturbation version is applied, the instantaneous response at the corresponding boundary location needs to be recalculated, with the response duration being 200 milliseconds. For each perturbation before and after, three key indicators—heat flux density, strain curvature, and shear boundary distribution—are extracted to generate a perturbation response feature dataset, providing standardized input for subsequent peak-shifting experiments.

[0032] Peak shifting analysis was conducted using the phase change behavior in the perturbation response feature dataset. The instantaneous phase peaks of all sampling points were aligned on a unified time axis, with the reference phase fingerprint baseline under undisturbed conditions used as a control. The phase peak shift time, the duration of the phase synchronization region, and the average frequency difference before and after the phase jump were measured for each sampling point within 100 milliseconds after the perturbation. The comparison showed that under certain perturbation conditions, the phase peaks exhibited a delay exceeding 15 milliseconds, and the frequency change rate decreased by more than 20%, indicating that the phase response entered an unstable alternating state. These peak shifting response phenomena reflect the failure of the internal coupling paths of the structure to recover coordinated vibration in a timely manner after being disturbed. Perturbation conditions with the maximum shift time and no phase recovery were selected and included in the preliminary decoupling candidate set, and their corresponding physical perturbation parameters were recorded as potential suppression variables.

[0033] Based on the results of the phase shifting test, a geometric reconstruction analysis was performed on the actual flow behavior within the fluid channel to extract whether any morphological changes affecting the mainstream path had occurred. Particle tracking technology was used to simulate the flow trajectory inside the channel under perturbation conditions, comparing the overlap rate of the main channel path for fluid particles within 90 milliseconds before and after the perturbation. If the overlap rate decreased by more than 30%, and local velocity vectors exhibited directional reversal or entrainment phenomena, it was considered that the channel geometry had undergone a substantial shift. Simultaneously, the spatial correspondence between strain changes and heat flux density changes on the outer surface of the structure at this location was extracted. If the strain reduction was highly consistent with the fluid shift location, it indicated that the perturbation had substantially weakened the structure-fluid coupling link. All perturbation combinations that simultaneously satisfied the three indices of main channel shift, strain reduction, and phase shifting were compiled into a minimum perturbation combination candidate set.

[0034] Energy intervention efficiency analysis was performed on the candidate set of minimum perturbation combinations to quantify the actual performance of each perturbation parameter combination on the interruption capability of the coupling process. The analysis method was as follows: First, the change rate of phase-locked energy before and after the perturbation was calculated to determine whether the phase-locked energy decrease ratio exceeded 40% after the perturbation took effect; second, the decrease in total structural strain energy was calculated to determine whether it reached more than 15%; third, it was assessed whether the dominant frequency of structural vibration shifted out of the danger band by at least 3 Hz. If the above three indicators were met simultaneously, the perturbation combination was confirmed to be effective. Among all effective combinations, the one with the smallest perturbation intensity was selected as the minimum perturbation set. Finally, this set included three specific parameters: boundary heat flux change, local stiffness adjustment value, and channel area reduction ratio, each of which had a clear implementation path and physical controllability.

[0035] Based on the minimum disturbance set, an energy dissipation channel consisting of time-varying virtual impedance and adjustable surface damping is constructed to realize energy transfer and dissipation in the interface region, while outputting intervention parameters and execution time window for energy suppression control. Based on the extraction of the minimum perturbation set, to effectively reduce the energy at the thermal-fluid-structure interaction interface, an energy dissipation channel composed of time-varying virtual impedance and adjustable surface damping should be constructed, and intervention parameters and the optimal execution time window for subsequent energy suppression should be output. The specific implementation is as follows: Based on the extracted set of minimum perturbations, the minimum intervention conditions for inducing the uncoupling of the coupled state are determined and mapped onto the physical control quantities of the structure-fluid interface region. These intervention conditions are spatially projected onto three specific locations: first, the inlet contraction section in the fluid channel where the instantaneous pressure pulsation is strongest; second, the connection transition region where the stiffness of the structural material changes significantly; and third, the adhesion interface layer where the coupling strength between the fluid shear velocity and the structural strain gradient is strongest. Virtual response elements are arranged around these three types of regions, comprising an electro-induced tunable flexible structural layer, a charge-sensitive responsive film, and a temperature-sensitive damping coating. The electro-induced flexible structural layer changes its equivalent structural stiffness within the response time by a preset electric field, simulating a nonlinear impedance effect; the charge-sensitive film detects phase response synchronicity and triggers impedance changes; and the temperature-sensitive damping layer affects viscoelastic properties by adjusting surface temperature, thus forming a dynamic energy dissipation mechanism. The three types of elements are arranged in a spatially intersecting layout on the boundary, forming a perceptible, responsive, and dissipative structural response network.

[0036] Using specific disturbance parameters from the minimum disturbance set as control inputs, response elements are sequentially activated in the aforementioned structural regions, and three feedback indicators are collected in real time: energy absorption rate per unit area; the instantaneous response frequency change curve of the structure; and the fluid shear force disturbance attenuation rate. Under each disturbance activation condition, the evolution trend of these three indicators within 100 milliseconds before and after is recorded. If the energy absorption rate per unit area exceeds a set threshold of 0.3 joules per square centimeter, and the structural response dominant frequency decreases by more than 2 Hz, and the shear disturbance wave amplitude decreases by more than 25%, it indicates that the activation behavior has constructed an effective energy dissipation path in the interface region. Subsequently, the path is spatially reconstructed, and its cross-regional energy guiding and longitudinal dissipation efficiency are analyzed. If the average energy attenuation rate in the path from the inlet section to the transition section is higher than 50%, and more than 90% of the energy is finally released in the adhesion interface layer, it is considered that the channel has formed a complete and effective energy dissipation chain. By comparing the construction efficiency of the dissipation channel under multiple disturbance activation combinations, the response combination that can obtain the maximum dissipation efficiency under the minimum activation condition is selected as the channel layout scheme for final execution.

[0037] Based on the successful construction of the energy dissipation channel, control intervention parameters and execution time windows are output around three key dimensions: response path length, energy dissipation delay, and structural vibration recovery time. First, using the dissipation path length as a benchmark, the energy transmission time from the inlet point to the point of complete dissipation is calculated, and this time is used as the shortest response duration of the intervention window. Then, using the energy dissipation delay as a constraint, the time required from the activation signal to the unit energy absorption rate being stably maintained above 95% is identified as the midpoint of the time window. Finally, the structural vibration recovery time is used as the safety criterion to set the complete execution cycle. Under these three-dimensional judgments, the optimal execution time window range is output, for example, a continuous time period between 180 milliseconds and 280 milliseconds. Simultaneously, based on the response parameter changes of all virtual impedance elements and damping adjustment elements in the channel, a set of physical intervention parameter tables is output, including voltage change curves, temperature control gradients, and structural compliance correction values. Each parameter has directly applicable physical quantity units and corresponds one-to-one with specific locations and response elements. Finally, this parameter table, along with the time window, is submitted to the control signal transmission layer to achieve dynamic energy regulation based on the linkage control of time-varying virtual impedance and adjustable surface damping. In the overall control process, all inputs are pre-calculated based on predictive data, and all response behaviors achieve energy dissipation based on microscale disturbance behavior, truly achieving the goal of maximizing coupling reduction with minimal physical intervention.

[0038] Within the execution time window, micropulses in the form of time reversal are applied according to the intervention parameters to activate the programmable phonon bandgap structure to adjust the boundary response characteristics. By superimposing shadow energy storage units to absorb the return energy, the phase-locked state is released and the coupling system is quickly extinguished, thereby completing the closed-loop optimization process of identifying, predicting and dynamically controlling the multi-field coupling response of thermo-fluid-solid system. After the energy dissipation channel is constructed and the intervention parameters and optimal execution time window are obtained, micropulses based on the time reversal principle need to be applied within this time window. This, combined with a programmable phonon bandgap structure, adjusts the boundary response characteristics, and a shadow energy storage unit is superimposed to absorb residual energy, thereby achieving phase lock release and energy quenching. The specific implementation is as follows: Based on the determined combination of intervention parameters and the optimal execution time window, three types of response-sensitive locations were selected within the structure-fluid boundary region: thermal excitation hysteresis points, shear stress convergence points, and strain feedback delay nodes. A transient micropulse excitation device was implanted at each location. This device features nanosecond-level start-up time, a controllable excitation frequency range of 15 kHz to 30 kHz, and an excitation waveform length between 1 and 3 complete cycles. The micropulse construction principle is time inversion: using the frequency, amplitude, and phase of the phase-locked signal in the current structural response as a reference, a short pulse with reverse phase, equal amplitude mirror image, and matched frequency is generated and applied to the excitation point with a π phase difference, ensuring that the micropulse simultaneously satisfies the reverse cancellation condition in both the time and frequency domains. The micropulse excitation method employs a pre-set high-frequency piezoelectric ceramic sheet exciter, which drives the generation of transient stress waves at the structural interface through ultrashort electrical pulses. The specific stress waveform is matched to the original response through pre-inversion modeling, achieving a one-to-one reverse injection. During the application of this micropulse, the energy input must be controlled to not exceed 20% of the average unit strain energy of the structure to ensure that it does not induce structural mode reconstruction or out-of-spectral excitation. After the application is completed, immediately record three indicators at the target point: phase jump rate, frequency reconstruction delay, and energy density reduction magnitude, in order to judge the initial effect of the inversion perturbation.

[0039] Within the first response cycle after the micropulse is applied, before the structural frequency stabilizes, a programmable phonon bandgap structure array deployed on the surface of the structural boundary is synchronously activated. This array consists of periodically arranged multi-material thin film layers with strict density and elastic modulus ratio requirements between the materials to ensure Bragg reflection conditions are formed within a specific wavelength range, thereby achieving frequency-selective blocking. The phonon structure is based on linear guideways, and each unit can be finely positioned using a piezoelectric nanoparticle displacement actuator to create dynamically changing lattice spacing. During activation, the frequency range requiring the bandgap is calculated based on the measured value of the structural master frequency after the micropulse is released, and the structural units are adjusted to the corresponding arrangement. This causes energy propagation within the range to create a reflection superposition effect along the boundary, achieving physical isolation of local standing waves. To prevent standing wave backlash caused by reflection, asymmetric energy-absorbing layers are deployed at both ends of the phonon bandgap, using a damped gradient rubber-graphite composite film to absorb reflected energy, achieving synchronization of frequency constraint and energy dissipation. In this way, based on the inverse weakening of the phase-locked energy by micropulses, the boundary propagation path is further controlled by the spatial structure characteristics, so that the residual resonant energy cannot form a closed reflection circuit inside the structure, thus completely blocking the continuous amplification chain of the phase-locked signal.

[0040] After the energy propagation path of the structure is disrupted, residual energy remains within the system, primarily in the form of residual strain, residual thermal pressure, or microscale shear vortices. To achieve closed-loop energy management, a shadow energy storage unit must be introduced to absorb this residual energy. The shadow energy storage unit is deployed in the stress-concentration zone of the structure, and consists of three functional layers: the first layer is a shape memory alloy coating that absorbs energy and undergoes microstructural phase transitions under heat or pressure, converting it into internal energy storage; the second layer is a microcapacitor energy storage layer that generates and stores charge under dynamic load using piezoelectric materials; and the third layer is a graphene microporous heat-absorbing mesh that locks in high-frequency thermal excitation energy through a nanocapillary structure. Within 10 milliseconds after micropulse excitation and phonon bandgap structure adjustment, if the energy density decreases by less than 70% in the structural response, the shadow energy storage unit will be immediately activated, with the strain rate as the trigger threshold. In passively activated mode, the energy storage unit can complete energy absorption and remain stable within 20 milliseconds, without reflecting back to other areas of the structure. After energy storage is completed, the system phase-locked energy index drops below the background disturbance, the phase synchronization rate is less than 30%, the structural response curve recovers the random disturbance distribution, the frequency concentration area disappears, marking the release of phase-locking and termination of coupling response.

[0041] This invention, by introducing a coupled phase fingerprint baseline, achieves for the first time a unified reference framework for the synchronous operation of thermal, flow, and solid-state response data. By constructing a counterfactual playback chain and a real-time risk cone, it enables threshold prediction and vulnerability assessment before crack initiation. Furthermore, by extracting the minimum disturbance set and constructing a variable impedance-damping coordinated energy dissipation channel, it endows the system with low-energy, high-efficiency local control capabilities. Finally, it utilizes time-reversal micropulses and phonon bandgap structures to achieve phase-lock release, and recovers energy through a shadow energy storage unit, completing rapid oscillation extinguishing and system steady-state recovery. This closed-loop optimization process has significant advantages such as fast response speed, high intervention accuracy, low energy loss, and adaptability to complex operating conditions, significantly improving the safety and reliability of the structure in unstable thermal-fluid-solid coupling environments.

[0042] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems, characterized in that, Includes the following steps: An asynchronous sampling array is constructed to jointly acquire the periodic pressure pulsation signal of the fluid and the dynamic response signal of the solid structure in time and frequency, and extract the coupled phase fingerprint baseline as a unified reference for energy analysis; Coherent decomposition is performed based on the coupled phase fingerprint baseline, the phase-locked energy index is calculated, the instantaneous resonance window in its dynamic evolution process is identified, and the dangerous frequency band and vulnerable interface region are determined. Based on the dangerous frequency bands and vulnerable areas of the interface, a counterfactual replay chain is constructed to deduce the crack initiation threshold of the key pulse sequence and generate a real-time risk cone; The boundary conditions are reconstructed online based on the real-time risk cone, and the fluid channel morphology is corrected by phase staggered peak probing to obtain the minimum perturbation set that interrupts the coupling process; Based on the minimum perturbation set, an energy dissipation channel containing time-varying virtual impedance and adjustable surface damping is constructed, and the energy suppression parameters and execution time window are output. Within the time window, time-reversal micropulses are applied according to the parameters to drive the programmable phonon bandgap structure and superimpose the shadow energy storage unit, thereby releasing the phase-locked state and completing the energy extinction.

2. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems according to claim 1, characterized in that, The steps for extracting the baseline of the coupled phase fingerprint are as follows: Distributed acquisition channels are set up in the structural interface area to collect fluid pressure pulsation signals and dynamic response signals of solid structures; Based on the frequency domain characteristics of each signal, an asynchronous sampling frequency is set and a sliding window envelope analysis is performed to extract energy density, frequency dominant components and modulation ratio; A dynamic alignment strategy is used to rearrange the main axis of the signal window and construct a time mapping matrix to complete time-frequency alignment. The relative phase evolution trajectory in the coupled signal is extracted, the frequency coupling density and phase jump interval are calculated, the continuous path with the highest phase stability is selected, and the coupled phase fingerprint baseline is generated as a unified reference frame architecture.

3. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems according to claim 2, characterized in that, The process for determining dangerous frequency bands and vulnerable interface regions is as follows: Within the target observation period, the fluid pressure signal and the solid structure response signal are segmented and processed, and a coupled propagation trajectory map is generated by phase alignment. Signal clusters are identified in the coupled propagation trajectory map and a phase-locked signal spectrum is constructed. The original sequence of the phase-locked energy index is calculated. Dynamic trend extraction is performed on the original sequence of phase-locked energy index to identify instantaneous resonance windows; The dominant frequency is extracted from the instantaneous resonance window and a dangerous frequency band is formed. The vulnerable area of ​​the structural interface is determined by combining the structural strain response.

4. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems according to claim 3, characterized in that, The steps for generating a real-time risk cone are as follows: Key pulse sequences were extracted from dangerous frequency bands and vulnerable regions of structural interfaces, and a pulse-response mapping table was constructed. The key pulse sequence is replaced with multiple perturbation versions, and phase interpolation is performed on the reconstructed signal to generate replacement paths and classify and construct path influence matrices. Based on the path influence matrix, crack initiation path characteristics are aggregated to form a crack initiation threshold and to deduce a crack propagation trend map. A real-time risk cone is generated based on the crack propagation trend map to characterize the distribution of future structural response risk.

5. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems according to claim 4, characterized in that, The steps to obtain the minimum perturbation set are as follows: The boundary domain is extracted based on the characteristics of phase-locked energy amplification, frequency concentration density, and strain rate increase in the real-time risk cone. A multi-level perturbation sample set was constructed around the boundary response parameters, and perturbation response features were collected; Candidate parameters for decoupling are selected based on phase misalignment, frequency variation, and coupling delay phenomena in the disturbance response. The channel morphology shift was analyzed by combining particle trajectory reconstruction and strain-thermal coupling features were extracted. The intervention efficiency of all effective perturbation combinations is evaluated, and the minimum perturbation set is finally determined.

6. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid system according to claim 5, characterized in that, The steps for outputting energy suppression parameters and execution time windows are as follows: Intervention conditions are extracted based on the minimum perturbation set and mapped to the interface region to construct a response network comprising an electrically induced flexible structure layer, a charge-sensing thin film, and a temperature-sensitive damping layer. The response elements were activated sequentially using the intervention parameters, and three feedback indicators were collected: energy absorption rate per unit area, structural instantaneous response frequency change curve, and fluid shear force disturbance attenuation rate. After the energy dissipation path is formed, the execution time window and intervention parameter table consisting of the response path length, energy dissipation delay and structural vibration recovery time are output.

7. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid system according to claim 6, characterized in that, The intervention parameter table includes the voltage change curve of the electrically induced flexible structure layer, the temperature control gradient of the temperature-sensitive damping layer, and the correction value of the structural flexibility, and each parameter corresponds one-to-one with the position of the corresponding response element.

8. The numerical simulation method for the multi-field coupled response characteristics of thermal-fluid-solid systems according to claim 6, characterized in that, Within the time window, time-reversal micropulses are applied according to the parameters to drive the programmable phonon bandgap structure and superimpose a shadow energy storage unit. The steps to release the phase-locked state are as follows: Based on the combination of intervention parameters and the execution time window, time-reversed micropulses are applied at the thermal excitation hysteresis point, the shear stress convergence point, and the strain feedback delay node; After micropulse application, the programmable phonon bandgap structure is activated to adjust the lattice spacing, thereby forming frequency-selective reflection; Subsequently, the shadow energy storage unit is activated in the stress convergence area to absorb the residual energy, completing the phase lock release and energy extinguishing process.

Images