Adaptive configuration method for optimal working point of ultrasonic thermal excitation system for concrete hidden crack

Abstract

The application discloses a kind of concrete hidden crack ultrasonic thermal excitation system optimal working point self-adapting configuration method, comprising: by constructing the parameter space including initial coupling pressure and working frequency, acquisition system is under different parameter combination Multiple physical field response data;Based on transducer active power response or thermal image temperature difference, construct excitation efficiency evaluation index, lock initial optimal working point by identifying the extreme characteristics of index response function;Using Hessian matrix to the neighborhood of optimal point carries out second-order sensitivity analysis, constructs tolerance ellipse model to quantify parameter control boundary;While based on the power-phase decoupling mechanism of voltage and current Real-time monitoring frequency and pressure drift.The application can be matched with optimal working parameters adaptively for different stiffness loading device, scientifically set engineering tolerance, improve the energy conversion efficiency and detection robustness of ultrasonic thermal excitation system.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic nondestructive testing technology, and in particular to an adaptive configuration method for the optimal operating point of an ultrasonic thermal excitation system for hidden cracks in concrete. Background Technology

[0002] Under long-term service loads, concrete structures inevitably develop cracks, among which latent cracks are a key early indicator of macroscopic crack formation and structural performance degradation. These hidden cracks often trigger a chain reaction of steel corrosion and load-bearing capacity reduction, easily leading to sudden structural hazards. Due to their small size, typically on the order of 0.01 mm in width, and concealed distribution, the initial morphology and evolution process of latent cracks are difficult to reliably capture using conventional non-destructive testing techniques.

[0003] Ultrasonic infrared thermography is a novel non-destructive testing method that integrates ultrasonic excitation and infrared thermography. It utilizes high-energy ultrasonic waves to excite material defects (such as cracks and delamination), causing energy dissipation and generating localized heat at the defect location, which accumulates and forms a visible temperature difference distribution on the material surface. Due to its non-contact nature, rapid response, and high sensitivity to hidden cracks in concrete, this technology has significant application value in the health monitoring and early damage identification of concrete structures.

[0004] Existing ultrasonic infrared thermal imaging detection systems typically consist of an ultrasonic generator, a piezoelectric transducer, and a pressurization device. In practice, operating parameters are mostly set based on manual experience or simple frequency sweep tests. This involves manually adjusting the initial coupling pressure of the pressurization device and the operating frequency of the ultrasonic generator-transducer system before detection, and observing the signal strength to estimate and determine the operating point. For example, operators often fix the frequency near the transducer's nominal resonant frequency and apply an estimated preload force by manually turning a knob.

[0005] However, the above empirical settings cannot guarantee that the system is at its optimal operating point. Under ideal conditions, the externally excited topology ultrasonic generator and the sandwich piezoelectric transducer can achieve impedance conjugate matching, making the reactance of the ultrasonic system close to zero when operating at the transducer's nominal frequency, thus achieving maximum electro-acoustic power transmission. However, in actual ultrasonic systems, due to the differences in transducer structural shapes at different frequencies, and the change in the equivalent impedance of the sandwich transducer under coupling pressure with mechanical load, the ultrasonic circuit inevitably has a certain reactance when operating at the transducer's design frequency. Furthermore, it is difficult to match the generator's internal resistance with the transducer's equivalent resistance, leading to a corresponding drift in the power peak point. This causes the system's electrical state to deviate from the optimal operating point, resulting in poor ultrasonic thermal excitation effect on hidden cracks in concrete. Therefore, the nominal frequency of the sandwich piezoelectric transducer does not necessarily correspond to the maximum electrical power output state of the ultrasonic system. There is a system-level matching relationship between the type of loading device, the initial coupling pressure and the operating frequency; when the three achieve synergistic optimization matching, the ultrasonic thermal excitation system is at its optimal operating point, enabling the transducer to output higher acoustic energy and improving the detectability of hidden cracks. Summary of the Invention

[0006] The purpose of this invention is to provide an adaptive configuration method for the optimal operating point of an ultrasonic thermal excitation system for hidden cracks in concrete, in order to solve the aforementioned problems existing in the prior art.

[0007] According to one aspect of this application, a method for adaptively configuring the optimal operating point of an ultrasonic thermal excitation system for hidden cracks in concrete includes:

[0008] Obtain the excitation control parameter space of the ultrasonic thermal excitation system to be searched. The excitation control parameter space shall include at least the initial coupling pressure dimension and the operating frequency dimension.

[0009] The ultrasonic thermal excitation system is driven to operate within the excitation control parameter space to be searched, and multi-physics response data of the system under different parameter combinations are collected.

[0010] Based on multiphysics response data, an excitation effectiveness evaluation index is constructed. The parameter points that make the excitation effectiveness evaluation index meet the extreme value conditions are searched, and these (parameter points that meet the extreme value conditions) are determined as the initial optimal operating points.

[0011] Within the neighborhood of the initial optimal operating point, the sensitivity characteristics of the excitation performance evaluation index relative to the excitation control parameters are calculated. Based on the sensitivity characteristics, the parameter tolerance boundary is determined, resulting in the optimal operating point configuration that includes the parameter center value and the parameter tolerance range.

[0012] Optionally, based on the optimal operating point configuration, the system operating status is monitored in real time during the detection process, and the parameter drift type is identified.

[0013] According to another aspect of this application, the multiphysics response data is electrical power data, which includes voltage data, current data and active power data under different excitation control parameters;

[0014] The excitation efficiency evaluation index is constructed based on multi-physics response data, including: under fixed initial coupling pressure parameters, calculating active power data corresponding to different frequencies based on voltage and current data obtained by frequency sweep, constructing a response function of active power as a function of frequency, and obtaining active power response; extracting the peak value or its statistical characteristics of the active power response, and using it (the peak value or its statistical characteristics of the active power response) as an excitation efficiency evaluation index characterizing the system's ability to effectively convert electrical energy.

[0015] Optionally, the excitation effectiveness evaluation index constructed based on multiphysics response data can also be:

[0016] Under multiple discrete pressure levels in the initial coupling pressure dimension, active power data corresponding to different frequencies are calculated based on voltage and current data obtained by frequency sweep, and a response function of active power as a function of frequency is constructed for each pressure level.

[0017] The peak value or statistical characteristic of the active power response under each pressure level is extracted and used as an incentive performance evaluation index to characterize the system's ability to effectively convert electrical energy.

[0018] According to another aspect of this application, the search for parameter points that cause the incentive effectiveness evaluation index to satisfy the extreme value condition includes:

[0019] Call the active power response function that varies with the initial coupling pressure parameters;

[0020] Identify the stagnation point where the first-order difference sign of the active power response function changes from positive to negative with respect to the initial coupling pressure, and determine the pressure corresponding to the stagnation point as the optimal initial coupling pressure;

[0021] Under optimal initial coupling pressure, the active power response function as a function of operating frequency is analyzed, and the frequency at which the maximum value is reached is determined as the optimal operating frequency.

[0022] The optimal initial coupling pressure and the optimal operating frequency are combined to form the initial optimal operating point.

[0023] According to another aspect of this application, the sensitivity characteristics of incentive effectiveness evaluation indicators relative to incentive control parameters are calculated, including:

[0024] Centered on the initial optimal operating point, a second-order performance surface model of the incentive effectiveness evaluation index is established with respect to the dimensions of initial coupling pressure and operating frequency.

[0025] Calculate the Hessian matrix of the second-order performance surface model at the initial optimal operating point;

[0026] The Hessian matrix is ​​subjected to eigenvalue decomposition, and the normalized curvature coefficients are extracted as sensitivity features representing the sensitivity of the parameter control.

[0027] According to another aspect of this application, the sensitivity characteristics of incentive effectiveness evaluation indicators relative to incentive control parameters are calculated, including:

[0028] Centered on the initial optimal operating point, a second-order performance surface model of the incentive effectiveness evaluation index is established with respect to the dimensions of initial coupling pressure and operating frequency.

[0029] Calculate the Hessian matrix of the second-order performance surface model at the initial optimal operating point;

[0030] Eigenvalue decomposition is performed on the Hessian matrix, and normalized curvature coefficients are extracted as sensitivity features characterizing the sensitivity of the parameter control. These normalized curvature coefficients include κ... _P With κ _f And satisfy:

[0031] κ _P =-(P 2 _opt / W _max )×H _PP ;

[0032] κ _f =-(f 2 _opt / W _max )×H _ff ;

[0033] Among them, P _opt with f _opt These represent the initial coupling stress and operating frequency corresponding to the initial optimal operating point, respectively, W. _max To maximize the incentive performance evaluation index at the initial optimal operating point, H _PP H represents the main diagonal element of the Hessian matrix in the pressure dimension. _ff represents the main diagonal element of the Hessian matrix in the frequency dimension.

[0034] According to another aspect of this application, determining parameter tolerance boundaries based on sensitivity characteristics includes:

[0035] Based on the eigenvalues ​​and eigenvectors of the Hessian matrix, a tolerance ellipse model centered on the initial optimal operating point is constructed.

[0036] Calculate the projection intercepts of the tolerance ellipse model on the initial coupling pressure dimension axis and the operating frequency dimension axis;

[0037] The projection intercept is determined as the allowable tolerance range of the initial coupling pressure parameter and the allowable tolerance range of the operating frequency parameter, respectively. The dimension axis with a larger eigenvalue has a smaller allowable tolerance range.

[0038] According to another aspect of this application, the operational status characteristics of the system are monitored in real time during the detection process, including:

[0039] The system acquires the current voltage and current data in real time, calculates the corresponding active power and voltage-current phase difference, and compares them with the active power and phase difference at the initial optimal operating point to obtain the change in active power and phase change.

[0040] When the change in active power exceeds a preset power threshold, but the change in phase does not exceed a preset phase threshold, the system is determined to have experienced coupling pressure drift.

[0041] When the phase change exceeds the preset phase threshold, the system is determined to have experienced operating frequency drift, regardless of whether the active power change exceeds the preset power threshold.

[0042] According to another aspect of this application, the operational status characteristics of the system are monitored in real time during the detection process, including:

[0043] Real-time acquisition of the system's current voltage and current data, and calculation of the corresponding active power W _curr and voltage-current phase difference θ _curr And respectively compared with the active power W at the initial optimal operating point. _opt and phase difference θ _opt By comparison, the changes in active power ΔW and phase Δθ are obtained, where:

[0044] ΔW=W _curr -W _opt ; Δθ=θ _curr -θ _opt ;

[0045] When abs(ΔW) exceeds the preset power threshold W _th And abs(Δθ) does not exceed the preset phase threshold θ _th When this occurs, it is determined that the system has experienced coupling pressure drift;

[0046] When abs(Δθ) exceeds the preset phase threshold θ _th At that time, regardless of whether abs(ΔW) exceeds the preset power threshold W _th All of these results indicate that the system has experienced operating frequency drift.

[0047] According to another aspect of this application, the multiphysics response data is infrared thermal imaging temperature data, which includes a sequence of thermal images obtained by infrared imaging of the surface of the object under test under different excitation control parameters.

[0048] Excitation effectiveness evaluation indicators are constructed based on multiphysics response data, including:

[0049] Identify the heat-generating regions before and after the excitation of hidden cracks in the thermal image sequence;

[0050] The average temperature difference between the heat-generating region after excitation of the hidden crack and the background heat-generating region before excitation is calculated, and this (average temperature difference) is used as an evaluation index of excitation efficiency characterizing the thermal excitation conversion efficiency of the system.

[0051] According to another aspect of this application, the search for parameter points that cause the incentive effectiveness evaluation index to satisfy the extreme value condition includes:

[0052] Under discrete initial coupling pressure levels, frequency sweep excitation is performed within a preset frequency range.

[0053] Compare the average temperature difference under different combinations of pressure levels and frequencies;

[0054] The combination of pressure level and frequency that produces the maximum average temperature difference is determined as the initial optimal operating point.

[0055] According to another aspect of this application, the ultrasonic thermal excitation system includes an externally excited topological ultrasonic generator and a sandwich piezoelectric transducer, and the loading device is selected from an aluminum alloy loading device or a glass fiber reinforced nylon loading device.

[0056] According to another aspect of this application, when using an aluminum alloy loading device, the initial optimal operating point determined by the method is: an initial coupling pressure of 2900 N and an operating frequency range of 38.1 kHz to 38.2 kHz.

[0057] According to another aspect of this application, when using a glass fiber reinforced nylon loading device, the initial optimal operating point determined by the method is: an initial coupling pressure of 3100N and an operating frequency range of 38.3kHz to 38.4kHz.

[0058] According to another aspect of this application, after identifying the stationary point where the first-order difference sign of the active power response function changes from positive to negative with respect to the initial coupling stress, the method further includes:

[0059] Count the number of stationary points identified within the excitation control parameter space to be searched;

[0060] If the number of stagnation points is greater than 1, calculate the active power value corresponding to each stagnation point and select the stagnation point with the largest value as the global optimal initial coupling pressure point.

[0061] If the number of stagnation points is equal to 1, then the current stagnation point is directly determined as the global optimal initial coupling pressure point, and the uniqueness of the optimal working point is confirmed.

[0062] According to another aspect of this application, the sensitivity feature also includes a parameter coupling index, which is calculated based on the ratio of the off-diagonal elements to the diagonal elements of the Hessian matrix.

[0063] Optionally, the parameter coupling index is calculated based on the ratio of the absolute value of the off-diagonal elements to the geometric mean of the diagonal elements of the Hessian matrix.

[0064] The method also includes: when the parameter coupling index exceeds a preset coupling threshold, generating a coordinated adjustment strategy for the frequency parameter and the initial coupling pressure parameter; when the parameter coupling index is lower than the preset coupling threshold, generating an independent adjustment strategy for the frequency parameter and the initial coupling pressure parameter.

[0065] According to another aspect of this application, the sensitivity feature also includes a parameter coupling index ρ, which satisfies: ρ = abs(H _Pf ) / sqrt(H _PP ×H _ff );

[0066] Among them, H _Pf H represents the off-diagonal elements of the Hessian matrix. _PP With H _ff These are the diagonal elements of the Hessian matrix.

[0067] The method also includes: when the parameter coupling index ρ exceeds the preset coupling threshold ρ _th When the frequency parameter and the initial coupling pressure parameter are adjusted in a coordinated manner, a strategy is adopted; when the parameter coupling degree index ρ is lower than the preset coupling threshold ρ _th At that time, an independent adjustment strategy is generated for the frequency parameters and the initial coupling pressure parameters.

[0068] Beneficial effects: This invention can adaptively match optimal operating parameters for loading devices with different stiffnesses, set engineering tolerances, and improve the energy conversion efficiency and detection robustness of the ultrasonic thermal excitation system. Attached Figure Description

[0069] Figure 1 The flowchart illustrates an adaptive configuration method for the optimal operating point of an ultrasonic thermal excitation system for hidden cracks in concrete, as provided in this application embodiment.

[0070] Figure 2 A flowchart illustrating the search for parameter points that enable the incentive performance evaluation index to meet extreme value conditions, as provided in the embodiments of this application.

[0071] Figure 3 A flowchart illustrating the sensitivity characteristics of the incentive performance evaluation index relative to incentive control parameters, as provided in the embodiments of this application.

[0072] Figure 4 This is a flowchart illustrating the determination of parameter tolerance boundaries based on sensitivity features, provided in an embodiment of this application.

[0073] Figure 5 A flowchart illustrating the search for parameter points that enable the incentive performance evaluation index to meet extreme value conditions, as provided in the embodiments of this application. Detailed Implementation

[0074] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0075] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0076] To address the aforementioned issues, the applicant conducted in-depth searches and analyses, and discovered:

[0077] Ultrasonic thermal excitation is a multi-physics field strong coupling process. The optimal operating point is not fixed, but dynamically drifts with the stiffness of the loading device, the material properties of the component under test, and the contact state of the coupling interface. This causes the fixed nominal parameters to deviate from the actual optimal point, resulting in a significant decrease in energy transmission efficiency.

[0078] Existing parameter adjustments lack quantitative tolerance design basis, making it difficult for operators to determine whether the current frequency or pressure deviation is within the linear range allowed by the system. This can easily lead to a precipitous drop in excitation efficiency due to small parameter fluctuations, and it is impossible to distinguish whether the performance degradation is caused by frequency drift or pressure relaxation, making it difficult to guarantee the repeatability and reliability of the detection results.

[0079] To solve these problems, combined with Figures 1 to 5 The present invention will be specifically described through the following embodiments.

[0080] On the one hand, an exemplary scheme is provided for an adaptive configuration method of the optimal operating point of an ultrasonic thermal excitation system for hidden cracks in concrete, describing the basic architecture construction of the ultrasonic thermal excitation system, the selection strategy of the loading device, and the complete process from parameter space scanning to the generation of the optimal configuration.

[0081] Step 101: Obtain the excitation control parameter space of the ultrasonic thermal excitation system to be searched. The excitation control parameter space shall include at least the initial coupling pressure dimension and the operating frequency dimension.

[0082] Alternatively, based on the equipment characteristics and engineering experience of the ultrasonic thermal excitation system, the excitation control parameter space to be searched is determined. The excitation control parameter space includes at least the initial coupling pressure dimension and the operating frequency dimension.

[0083] In this embodiment, the ultrasonic thermal excitation system mainly consists of a self-excited topological ultrasonic generator, a sandwich piezoelectric transducer, and a transducer limiting loading device. Specifically, the generator can be a KMD-M3 type, which supports wideband adjustment from 17kHz to 135kHz and has a maximum output power of 300W. The transducer is specifically a KMD 40kHz 50W PZT-4 type piezoelectric transducer, which has high sensitivity to hidden cracks in concrete.

[0084] Regarding the transducer limiting loading device, this embodiment emphasizes the influence of the loading device stiffness on the system amplitude stability. To meet the needs of different detection scenarios, the loading device can be specifically selected from aluminum alloy loading devices or glass fiber reinforced nylon loading devices. Among them, aluminum alloy loading devices have higher overall stiffness and better structural strength; while glass fiber reinforced nylon loading devices have lower overall stiffness, which can reduce additional force disturbances and improve excitation effect, making them suitable for scenarios with high detection requirements.

[0085] Based on this, the space of excitation control parameters to be searched is determined. This is a two-dimensional set of parameters, encompassing the initial coupling pressure dimension P and the operating frequency dimension f. For example, based on engineering experience and equipment characteristics, the search range for the initial coupling pressure can be set to 1500N to 4000N, with a step size of 20N to 200N; the search range for the operating frequency can be set to 35kHz to 45kHz, with a step size of 0.01kHz to 0.1kHz. This space is set to cover the potential optimal operating point near the transducer design frequency.

[0086] Step 102: Drive the ultrasonic thermal excitation system within the excitation control parameter space to be searched, and collect multi-physics response data of the system under different parameter combinations.

[0087] Specifically, the control system traverses each parameter in the parameter space according to a preset step size (P). _i f _j ) combination, P _i Corresponding to the initial coupling pressure level, f _j The corresponding operating frequency. Under each parameter combination, the system generator outputs stably in constant power control mode at the preset power setting value, and simultaneously acquires the system response. Multiphysics response data can include electrical response data, such as voltage and current, to calculate active power; it can also include thermal response data, such as the surface temperature field distribution of the measured object acquired by an infrared thermal imager. The above data reflects the efficiency and state of the system in converting electrical energy into acoustic energy and then into thermal energy under the current control parameters.

[0088] Step 103: Based on the constructed incentive effectiveness evaluation index, search for parameter points that make the incentive effectiveness evaluation index meet the extreme value conditions, and determine them (parameter points that meet the extreme value conditions) as the initial optimal working points.

[0089] In this embodiment, the excitation efficiency evaluation index is a physical quantity used to quantify the system's energy conversion efficiency. If electrical response data is used, this index can be active power, because in electrical theory, the maximum active power means the transducer converts electrical energy into mechanical vibration (overcoming mechanical losses) with the highest efficiency, achieving maximum mechanical amplitude output. If thermal response data is used, this index can be the average temperature difference before and after excitation in the crack region, directly reflecting the heat generation efficiency at the hidden crack.

[0090] The search process involves finding the extreme points of the index in the parameter space. For example, finding the pressure value that maximizes active power, or the frequency-pressure combination that maximizes the average temperature difference. Parameter points that satisfy the preset extreme value conditions are marked as the initial optimal operating points.

[0091] Step 104: Within the neighborhood of the initial optimal operating point, calculate the sensitivity characteristics of the excitation performance evaluation index relative to the excitation control parameters, determine the parameter tolerance boundary based on the sensitivity characteristics, and obtain the optimal operating point configuration that includes the parameter center value and the parameter tolerance range.

[0092] This step not only identifies the optimal point but also provides engineering tolerances by analyzing the terrain features near that point. Specifically, the system analyzes the rate at which the excitation performance evaluation index changes with parameters within a small neighborhood of the initial optimal operating point. If the evaluation index is highly sensitive to frequency changes (i.e., a slight deviation in frequency causes a sharp drop in the index), the frequency tolerance range should be set narrower; if it is relatively insensitive to pressure changes, the pressure tolerance range can be appropriately widened. This differentiated tolerance design based on sensitivity analysis allows the system to maintain both high efficiency and necessary operational robustness in practical engineering applications. Based on this, the output optimal operating point configuration includes the center parameter and the allowable deviation range (±ΔP, ±Δf).

[0093] Optionally, based on the optimal operating point configuration, the system's operating status characteristics are monitored in real time during the detection process, and the parameter drift type is identified according to the operating status characteristics.

[0094] Alternatively, the method of this invention can also be:

[0095] Construct a parameter space that includes initial coupling pressure and operating frequency, and collect multiphysics response data of the system under different parameter combinations;

[0096] Based on the active power response of the transducer or the thermal image temperature difference, an excitation efficiency evaluation index is constructed, and the extreme value characteristics of the index response function are identified to lock the initial optimal operating point.

[0097] A second-order sensitivity analysis of the neighborhood of the optimal point is performed using the Hessian matrix, and a tolerance ellipse model is constructed to quantify the parameters to control the boundary.

[0098] A power-phase decoupling mechanism based on voltage and current is used to monitor frequency and pressure drift in real time.

[0099] On the other hand, it describes an alternative implementation method for using electrical response data to pinpoint the optimal operating point of a system, particularly by introducing an active power response extremum analysis and an extremum uniqueness determination mechanism, which solves the problems of low efficiency and uncertainty in traditional trial-and-error methods.

[0100] Step 201: The multiphysics response data is electrical response data, which includes voltage and current data under different excitation and control parameters.

[0101] In this embodiment, the system uses an external oscilloscope to acquire the voltage across the transducer. Data and Current Data. For each initial coupled pressure point P in the parameter space. _iThe system performs a complete frequency scan. During the scan, the voltage V and current I across the transducer, as well as the voltage-current phase difference θ, are measured in real time. Therefore, the acquired dataset contains sequences of V, I, and θ values ​​under different pressures and frequencies.

[0102] Step 202: Construct excitation effectiveness evaluation indicators based on multiphysics response data, including:

[0103] Under fixed initial coupling pressure parameters, the active power function is calculated based on the electrical response data obtained by frequency sweep, and it (the peak value of the active power response or its statistical characteristic) is used as the excitation efficiency evaluation index characterizing the electroacoustic energy conversion efficiency of the system.

[0104] In other words, under a fixed initial coupling pressure parameter selected in the excitation control parameter space, the active power data corresponding to different frequencies is calculated based on the voltage and current data obtained by frequency sweep, and the response function of active power changing with frequency is constructed to obtain the active power response.

[0105] Specifically, for any fixed pressure P _i The active power W is calculated from the electrical response data points obtained by frequency scanning, i.e.:

[0106] W(P _i f)=V(P _i f)×I(P _i ,f)×cosθ(P _i , f);

[0107] Among them, V(P) _i f) represents the pressure P _i and the effective value of voltage at frequency f, I(P) _i f) represents the corresponding effective value of the current, θ(P) _i f) represents the phase difference between the two. Based on this, the active power W(P) of the transducer can be obtained. _i f).

[0108] The active power W directly reflects the transducer's instantaneous energy conversion capability and final acoustic output intensity under the current load. A higher active power indicates that the transducer can generate stronger mechanical vibrations and acoustic energy output under the same electrical excitation. Therefore, selecting active power as an evaluation index for excitation effectiveness has a certain physical basis.

[0109] Step 203: Search for parameter points that make the incentive effectiveness evaluation index meet the extreme value conditions, including:

[0110] Call the active power response function that varies with the initial coupling pressure parameter; identify the stagnation point where the first-order difference sign of the active power response function changes from positive to negative with respect to the initial coupling pressure, and determine the pressure corresponding to the stagnation point as the optimal initial coupling pressure.

[0111] In this step, the system will apply different pressures P _i The active power W(P) calculated below _i Connecting f, we construct the discrete active power response function W(P, f). Optionally, the system calculates the first-order difference sequence ΔW of this function. _i =W _i+1 -W _i When the system detects ΔW _k >0 and ΔW _k+1 When <0, it indicates that in P _k To P _k+1 There exists a local maximum (stationary point) between P and P. To improve accuracy, polynomial interpolation can be used at P. _k By performing a local fit on W(P,f) in the vicinity and calculating the optimal initial coupling pressure P that maximizes active power by finding the derivative to zero, the optimal coupling pressure P is obtained. _opt .

[0112] Step 204, after identifying the stagnation point where the first-order difference sign of the active power response function changes from positive to negative with respect to the initial coupling pressure, further includes: counting the number of stagnation points identified in the excitation control parameter space to be searched; if the number of stagnation points is greater than 1, then calculating the active power value corresponding to each stagnation point, and selecting the stagnation point with the largest value as the global optimal initial coupling pressure point; if the number of stagnation points is equal to 1, then directly determining the current stagnation point as the global optimal initial coupling pressure point, and confirming the uniqueness of the optimal operating point.

[0113] In other embodiments, under optimal initial coupling pressure, the active power response function as a function of operating frequency is analyzed, and the frequency at which the maximum value is reached is determined as the optimal operating frequency.

[0114] The optimal initial coupling pressure and the optimal operating frequency are combined to form the initial optimal operating point.

[0115] In complex nonlinear coupled systems, the W(P,f) curve may exhibit a multi-peak shape along the pressure dimension axis. The system counts all stagnation points identified across the entire pressure search range. If the count result N>1, it indicates that multiple pressure points can achieve a locally favorable state for the system. In this case, the system compares the absolute values ​​of the active power corresponding to each stagnation point and selects the stagnation point with the highest active power as the globally optimal initial coupled pressure point P. _opt .

[0116] If the count result N=1, it indicates that a unique optimal operating point exists within the search range, providing a high degree of confidence for subsequent engineering applications. The optimal initial coupling pressure P has been confirmed. _opt Then, the system's active power response curve W(P) under this pressure as a function of frequency. _opt The frequency corresponding to the maximum point f is determined as the optimal operating frequency f. _opt Based on this, the initial optimal operating point (P) is obtained by combining these factors. _opt f _opt ).

[0117] On the other hand, this paper describes how to utilize second-order analysis tools from advanced mathematics to move beyond simply finding the optimal point and construct the optimal domain. This embodiment quantifies the sensitivity of system performance to control parameters, calculates the engineering tolerance boundary, and establishes a decoupled drift monitoring mechanism, thus solving the problems of parameter mismatch and unstable state in ultrasonic thermal excitation systems in practical applications.

[0118] Step 301: Calculate the sensitivity characteristics of the incentive effectiveness evaluation index relative to the incentive control parameters, including:

[0119] Centered on the initial optimal operating point, a second-order performance surface model of the incentive effectiveness evaluation index is established with respect to the dimensions of initial coupling pressure and operating frequency.

[0120] Calculate the Hessian matrix of the second-order performance surface model at the initial optimal operating point; perform eigenvalue decomposition on the Hessian matrix and extract the normalized curvature coefficients as sensitivity features characterizing the sensitivity of the parameter control.

[0121] In this embodiment, the system no longer treats the optimal operating point as an isolated point (P). _opt f _opt Instead of viewing it as a peak point on the performance surface W(P,f), the system considers it as such. To analyze the steepness near this peak, the system uses the Taylor expansion principle to construct a second-order polynomial model in the neighborhood of the optimal point. Based on this model, the system calculates the Hessian matrix H, which describes the second-order partial derivatives of the performance index W in both parameter dimensions. Specifically, the Hessian matrix can be expressed as: H = [[H _PP H _Pf ], [H _fP H _ff ]];

[0122] Among them, H _PP H is the second-order partial derivative of the active power W with respect to the initial coupling pressure P. _ff H is the second-order partial derivative of W with respect to the operating frequency f. _Pf and H _fPFor mixed second-order partial derivatives, and H _Pf =H _fP Partial derivatives can be calculated using the central difference method with sampled data from the neighborhood.

[0123] To eliminate the influence of different physical unit dimensions (Newton and Hertz) and achieve a unified comparison of sensitivity, this embodiment introduces the normalized curvature coefficient as a key sensitivity feature. The specific calculation formula is as follows:

[0124] κ _P =-(P 2 _opt / W _max )×H _PP ;

[0125] κ _f =-(f 2 _opt / W _max )×H _ff ;

[0126] Among them, κ _P P is the normalized curvature coefficient of the initial coupled pressure dimension. _opt For the optimal initial coupling pressure value, W _max H represents the maximum active power at the optimal point. _PP κ is the second-order partial derivative of the pressure dimension; _f f is the normalized curvature coefficient in the operating frequency dimension. _opt For the optimal operating frequency value, H _ff This is the second-order partial derivative in the frequency dimension. The larger the value of the normalized curvature coefficient, the steeper the performance surface in that direction, meaning the more sensitive the system performance is to changes in this parameter.

[0127] Step 302, the sensitivity feature also includes a parameter coupling index, which is calculated based on the ratio of the off-diagonal elements to the diagonal elements of the Hessian matrix; the method also includes:

[0128] When the parameter coupling index exceeds the preset coupling threshold, a coordinated adjustment strategy is generated between the frequency parameter and the initial coupling pressure parameter.

[0129] When the parameter coupling index is lower than the preset coupling threshold, an independent adjustment strategy is generated for the frequency parameter and the initial coupling pressure parameter.

[0130] In real-world physical systems, pressure and frequency are not independent. For example, a change in pressure may slightly affect the frequency characteristics of a transducer. This embodiment defines a parameter coupling index ρ to quantify the mutual influence. The formula for calculating this index is: ρ = abs(H _Pf ) / sqrt(H _PP×H _ff );

[0131] Where ρ is the parameter coupling index, abs(H _Pf H is the absolute value of the mixed second-order partial derivative. _PP and H _ff These are the principal second-order partial derivatives for pressure and frequency, respectively.

[0132] Optionally, the system presets a coupling threshold ρ _th For example, 0.1. When the calculated ρ is less than ρ _th When ρ is greater than ρ, it indicates that the interaction between the two parameters is negligible, and the system generates an independent adjustment strategy, meaning that adjusting pressure and frequency separately will achieve the optimal result. _th When this occurs, it indicates a strong coupling between the two. In this case, a coordinated adjustment strategy must be generated. For example, while adjusting the pressure, the frequency must be compensated synchronously according to a predetermined functional relationship to prevent the system performance from slipping out of the optimal range.

[0133] Step 303, determining the parameter tolerance boundary based on sensitivity characteristics, including:

[0134] Based on the eigenvalues ​​and eigenvectors of the Hessian matrix, a tolerance ellipse model centered on the initial optimal operating point is constructed.

[0135] Calculate the projection intercepts of the tolerance ellipse model on the initial coupling pressure dimension axis and the operating frequency dimension axis; determine the projection intercepts as the allowable tolerance ranges of the initial coupling pressure parameter and the operating frequency parameter, respectively, where the dimension axis with larger eigenvalues ​​has a smaller allowable tolerance range.

[0136] This step translates the mathematical curvature into an engineering-operable tolerance. The system performs eigenvalue decomposition on the Hessian matrix H, obtaining two eigenvalues ​​λ. _1 and λ _2 And the corresponding eigenvectors. The two eigenvalues ​​represent the magnitude of the principal curvatures of the performance surface.

[0137] Based on a preset performance degradation tolerance (e.g., allowing a 5% decrease in active power), a tolerance ellipse equation can be defined, specifically: λ _1 ×x' 2 +λ _2 ×y' 2 =C _tol ;

[0138] Where x' and y' are local coordinates along the direction of the eigenvector, C _tol This is a constant related to the tolerance for performance degradation. By projecting this ellipse back onto the original Pf coordinate system, the intercepts on the pressure axis and frequency axis can be obtained, denoted as ΔP and Δf, respectively.

[0139] The two intercepts constitute the final tolerance boundary, and the initial coupling pressure is allowed at P. _opt Fluctuations within ±ΔP range, operating frequency allowed within f _opt Fluctuations occur within the range of ±Δf. Based on the properties of eigenvalue decomposition, directions with large eigenvalues ​​(steep surfaces) necessarily have smaller tolerance ranges; conversely, directions with small eigenvalues ​​(flat surfaces) have larger tolerance ranges. This design allows for strict control in sensitive directions and relaxed control in insensitive directions, thus optimizing control costs.

[0140] Step 304: During the detection process, monitor the system's operational status characteristics in real time, including:

[0141] The system acquires current voltage and current data in real time and calculates the change in active power and phase relative to the initial optimal operating point. When the change in active power exceeds a preset power threshold but the phase change does not exceed a preset phase threshold, the system is determined to have experienced coupling pressure drift; when the phase change exceeds a preset phase threshold, the system is determined to have experienced operating frequency drift.

[0142] In other embodiments, the method of the present invention further includes: based on the optimal operating point configuration, monitoring the operating status characteristics of the system in real time during the detection process, and identifying the parameter drift type according to the operating status characteristics.

[0143] Among them, the operational status characteristics of the real-time monitoring system during the detection process include:

[0144] The system acquires the current voltage and current data in real time, calculates the corresponding active power and voltage-current phase difference, and compares them with the active power and phase difference at the initial optimal operating point to obtain the change in active power and phase change.

[0145] When the change in active power exceeds a preset power threshold, but the change in phase does not exceed a preset phase threshold, the system is determined to have experienced coupling pressure drift.

[0146] When the phase change exceeds the preset phase threshold, the system is determined to have experienced operating frequency drift, regardless of whether the active power change exceeds the preset power threshold.

[0147] The system monitors the current active power (W) in real time. _curr and voltage-current phase difference θ _curr The active power W at the optimal point _opt and θ _opt Compare them.

[0148] Calculate the phase change Δθ: Δθ = θ _curr -θ _opt ;

[0149] Calculate the change in active power Δ|W|: Δ|W|=W _curr -W _opt ;

[0150] Based on the physical characteristics of piezoelectric transducers, frequency drift primarily causes changes in the reactance matching effect of the ultrasonic system near its optimal operating frequency, while having a smaller impact on resistance matching, further altering the voltage and current phase angles. Conversely, changes in coupling pressure primarily alter the resistance matching effect of the ultrasonic system near its optimal operating frequency, leading to changes in active power, while having a relatively smaller impact on reactance matching. Based on this mechanism, the system sets a phase threshold θ. _th and active power threshold W _th .

[0151] The specific judgment logic is as follows:

[0152] If abs(Δ|W|)>W _th And abs(Δθ) < θ _th If the force is too high, it is primarily determined to be coupling pressure drift, and the system should prompt the user to check the tightness of the loading device or automatically adjust the applied force. If abs(Δθ|)>θ _th If the active power changes, it is determined to be frequency drift, and the system should prioritize adjusting the generator's output frequency to track it. This decoupling diagnostic strategy improves the targeted nature and efficiency of system maintenance.

[0153] This step provides an online diagnostic mechanism to identify the physical causes of system performance degradation.

[0154] One example describes how infrared thermal imaging technology can be used to evaluate the energy conversion efficiency of an ultrasonic thermal excitation system by constructing a visualized temperature field response, thereby determining the optimal operating parameters for different loading devices. Although this method is more cumbersome than electrical methods, its results are intuitive and reliable, and it is often used as a verification benchmark for electrical optimization results.

[0155] Step 401: The multiphysics response data is infrared thermal imaging temperature data, which includes a sequence of thermal images obtained by infrared imaging of the surface of the object under test under different excitation control parameters.

[0156] Furthermore, the experimental system incorporates an uncooled infrared thermal imager as a response acquisition device, building upon its hardware foundation. The thermal imager is positioned directly over the surface of the test object (e.g., a concrete specimen containing real hidden cracks) and continuously acquires a sequence of thermal images at a fixed frame rate (e.g., 50 Hz).

[0157] Optionally, the acquisition process is conducted in a temperature-controlled, sealed space to eliminate environmental interference. The thermal image sequence records the spatiotemporal evolution of the temperature field in the crack region caused by stimulated heating under ultrasonic excitation. Each thermal image frame is a two-dimensional temperature matrix T(x, y, t), where (x, y) are pixel coordinates and t is a timestamp. This data reflects the cumulative effect of ultrasonic energy being converted into heat energy at the hidden crack.

[0158] Step 402: Construct an excitation performance evaluation index based on multiphysics response data, including: identifying the heat-generating regions before and after the hidden crack excitation in the thermal image sequence; calculating the average temperature difference between the heat-generating region after the hidden crack excitation and the background heat-generating region before the hidden crack excitation, and using it as an excitation performance evaluation index characterizing the thermal excitation conversion efficiency of the system.

[0159] In this step, the process is as follows:

[0160] Image processing is performed on the original thermal image sequence, i.e., using image segmentation algorithms, such as the Otsu method or gradient-based edge detection, to automatically identify areas of increased temperature in the excitation thermal images and mark them as hidden crack heat-generating regions R. _defect Simultaneously, the same region before excitation was selected as the background heat generation region R. _background .

[0161] Furthermore, construct an incentive effectiveness evaluation index ΔT _avg Specifically:

[0162] Calculate R at each time step _defect The average temperature T in the region _d (t) and the time before excitation R _background The average temperature T in the region _b The difference between (t) and the two is the instantaneous temperature difference of the hidden crack. The temperature difference value at the end of the excitation (e.g., at the end of the 15th second) or the maximum temperature difference value during the entire excitation process is selected as the final evaluation index.

[0163] ΔT _avg =mean(T _d )-mean(T _b );

[0164] Where, ΔT _avg The mean (T) is the average temperature difference index. _d ) represents the average temperature value of the pixels in the hidden crack region, mean(T) _b The value represents the average temperature of the pixels in the background heat-generating region. Under the same excitation energy input, a larger temperature difference indicates a stronger ability of the system to generate heat through hidden cracks, meaning a higher thermal excitation conversion efficiency.

[0165] Step 403: Search for parameter points that enable the excitation performance evaluation index to meet preset extreme conditions, including: performing frequency sweep excitation within a preset frequency range under discrete initial coupling pressure levels; comparing the average temperature difference under different pressure levels and frequency combinations; and determining the pressure level and frequency combination that produces the maximum average temperature difference as the initial optimal operating point.

[0166] In this step, a full factorial experiment was designed to find the optimal operating point.

[0167] Accordingly, discrete initial coupling pressure levels were set. Based on the mechanical characteristics of the device, 2500N, 2700N, 2900N, 3100N, 3300N, and 3500N were selected as test points.

[0168] Next, set the frequency scan range. Perform a fine scan in 0.1 kHz increments within the range of 37.5 kHz to 38.5 kHz.

[0169] Based on this, the experimental procedure is as follows:

[0170] Fixed initial coupling pressure level P _i Adjust the operating frequency to f. _j The ultrasonic excitation was triggered, with a duration set to 15 seconds and a constant output power of 70W. The thermal imager simultaneously recorded the data, calculated, and saved the average temperature difference ΔT under this condition. _avg (P _i f _j Cool the test block to room temperature, change the frequency or pressure, and repeat the above steps.

[0171] After completing tests under all operating conditions, the system generates a two-dimensional temperature rise distribution table. By comparing the data in the table, ΔT is found. _avg The cell with the largest value corresponds to the initial optimal operating point for that loading device based on the combination of pressure and frequency.

[0172] Furthermore, for the aluminum alloy loading device, the temperature rise in the hidden crack area is the largest in the range of initial coupling pressure of 2900N and operating frequency of 38.1kHz to 38.2kHz, with an average temperature difference of more than 0.6℃, which is higher than other pressure points, such as 0.3℃ at 2500N.

[0173] For the glass fiber reinforced nylon loading device, the optimal operating point has shifted, the initial optimal coupling pressure has increased, and the optimal operating frequency range has shifted slightly upward.

[0174] According to one aspect of this application, after determining the initial optimal operating point, the same method as in the foregoing embodiments can be used to average the temperature difference ΔT. _avgInstead of active power W, a Hessian matrix is ​​constructed to calculate the tolerance elliptical boundary.

[0175] Another example describes the specific optimal operating parameters determined by the system after adaptive optimization when using a high-stiffness aluminum alloy loading device. The data in this embodiment comes from the experimental verification of the aforementioned embodiments and represents the optimal physical state under this specific hardware combination.

[0176] Step 501, the ultrasonic thermal excitation system includes an externally excited topological ultrasonic generator and a sandwich piezoelectric transducer, and the loading device is selected from an aluminum alloy loading device or a glass fiber nylon loading device.

[0177] Alternatively, the ultrasonic thermal excitation system includes a self-excited topological ultrasonic generator that supports wide frequency adjustment from 17kHz to 135kHz and a PZT-4 sandwich piezoelectric transducer with a nominal frequency of 40kHz and a rated power of 50W. The loading device is selected from an aluminum alloy loading device or a glass fiber nylon loading device.

[0178] In this embodiment, the system hardware configuration is as follows: the ultrasonic generator is a KMD-M3 type with automatic frequency tracking function; the transducer is a KMD 40kHz 50W PZT-4 type sandwich piezoelectric ceramic transducer with a resonant frequency of approximately 40kHz.

[0179] Specifically, the loading device selected in this embodiment is an aluminum alloy loading device. This device is made of aluminum alloy, which has extremely high elastic modulus and structural stiffness. The high stiffness characteristic introduces a large additional force disturbance into the transducer during vibration, but it has good structural strength and high durability.

[0180] Step 502: When using an aluminum alloy loading device, the initial optimal operating point determined by the method is: an initial coupling pressure of 2900N and an operating frequency range of 38.1kHz to 38.2kHz.

[0181] Furthermore, the system employs an adaptive configuration method, whether based on electrical power response or infrared thermal imaging, and ultimately converges and locks in the optimal operating point as follows:

[0182] Optionally, the optimal initial coupling pressure P _opt =2900N. At this pressure, the transducer and generator can achieve a better resistance matching effect, improving active power. Too low a pressure (such as 2500N) will cause the resistance matching effect between the transducer and generator to decrease; while too high a pressure (such as 3100N) will introduce excessive losses at the coupling interface and affect the resistance matching stability, reducing energy transfer efficiency. 2900N is the balance point that maximizes active power and maximizes thermal excitation temperature rise.

[0183] The optimal operating frequency can be f. _opt =38.1kHz to 38.2kHz: Due to the pressure load effect of the loading device and the impedance characteristics of the generator, the optimal frequency of the ultrasonic generation system shifts downward relative to the nominal frequency (approximately 40kHz). Under the constraint of the aluminum alloy loading device, the system achieves the highest efficiency in converting electrical energy to acoustic energy near 38.15kHz.

[0184] In summary, when the system recognizes that the loading device is made of aluminum alloy, it can directly call or recommend the above parameters as the initial settings, thus shortening the on-site debugging time.

[0185] Another example describes the drift of the system's optimal operating parameters and the final determination of the optimal configuration when using a low-stiffness glass fiber reinforced nylon loading device.

[0186] The ultrasonic thermal excitation system includes an externally excited topological ultrasonic generator and a sandwich piezoelectric transducer, and the loading device is selected from an aluminum alloy loading device or a glass fiber nylon loading device.

[0187] In this embodiment, the hardware is the same as in the previous embodiment, but the loading device is replaced with a glass fiber reinforced nylon loading device. This device is made of reinforced nylon material (PA66+GF30, polyamide 66+30% glass fiber), which has lower stiffness compared to aluminum alloy.

[0188] Building upon this, the introduction of a glass fiber reinforced nylon loading device can utilize the material's low stiffness to suppress additional force disturbances, resulting in higher amplitude stability of the transducer output. However, the low stiffness of nylon material causes changes in the mechanical response, leading to a shift in the optimal operating point.

[0189] When using a glass fiber reinforced nylon loading device, the initial optimal operating point determined by the method is: an initial coupling pressure of 3100N and an operating frequency range of 38.3kHz to 38.4kHz.

[0190] The experimental data and the algorithm optimization results consistently show that the optimal operating point of the glass fiber reinforced nylon loading device has the following characteristics:

[0191] Optimal initial coupling pressure P _opt =3100N, compared to the aluminum alloy loading device (2900N), the nylon device requires a greater preload force (an increase of approximately 200N) to achieve optimal coupling. Due to the softer nature of nylon, greater pressure is needed to establish the optimal resistance match between the transducer and the generator.

[0192] Optimal operating frequency f _opt=38.3kHz to 38.4kHz, the glass fiber nylon loading device has relatively small additional force disturbance and has a smaller effect on the frequency of the transducer. Therefore, its optimal operating frequency (about 38.35kHz) is slightly higher than that of the aluminum alloy loading device (about 38.15kHz).

[0193] This embodiment reveals the decisive influence of the overall stiffness of the loading device on the operating point of the ultrasonic thermal excitation system. A specific ultrasonic system can identify the location of the optimal point (e.g., if the optimal point is found to be near 3100N rather than 2900N), and can even infer the type of loading device currently installed, thus achieving intelligent parameter adaptation.

[0194] In some embodiments, the test process may also employ the following methods:

[0195] Within the amplitude system control framework, the following ultrasonic thermal excitation tests were conducted to verify and determine the optimal electrical operating state of the ultrasonic system under the condition of a glass fiber nylon loading device. The specific implementation method is as follows:

[0196] For the glass fiber reinforced nylon loading device, within the initial coupling pressure range of 2500N to 3500N, four pressure levels of 2700N, 2900N, 3100N, and 3300N are selected with a step size of 200N. These four pressures are applied respectively to form four basic working conditions.

[0197] For each basic operating condition, a scan is performed near the operating frequency tracked by the generator. Accordingly, based on the generator's frequency tracking function, the optimal operating frequency range of the generator-transducer system is initially determined. Using a frequency step size of 0.1 kHz, four operating frequency points are selected within this initial optimal operating frequency range to form the frequency test set for that operating condition. This step size setting can capture the key changing trend of the thermal excitation efficiency curve near the optimal operating point, balancing experimental efficiency with engineering accuracy.

[0198] A KMD-M3 externally excited topological ultrasonic generator and a KMD 40kHz 50W PZT-4 sandwich piezoelectric transducer were used for short-duration excitation at a constant output power of 70W, with each excitation lasting 15 seconds and an interval of 6 minutes. Air cooling was employed to rapidly dissipate heat from the system. An infrared thermal imager was used to monitor the temperature field in the microcrack area on the surface of the concrete specimen, and temperature data before and after excitation were recorded.

[0199] The average temperature difference under all experimental conditions was compared and analyzed. For ultrasonic thermal excitation applications, the optimal electrical operating state of the ultrasonic system is ultimately manifested in the most significant conversion of acoustic energy input into the test object into heat at the hidden cracks. Both theoretical analysis and experiments show that, given the concrete material and ultrasonic thermal excitation conditions, the average temperature difference in the concrete hidden cracks is directly positively correlated with the effective acoustic energy output by the ultrasonic system.

[0200] Therefore, the average temperature difference can be used as a performance indicator to characterize the overall operating point of the ultrasonic thermal excitation system. When a certain operating frequency at a given initial coupling pressure level produces the highest average temperature difference, the ultrasonic thermal excitation system is considered to be in its optimal operating state. This frequency is determined as the optimal operating frequency of the system under the current pressure, and this pressure is identified as the optimal initial coupling pressure corresponding to the system. Experimental results show that the optimal initial coupling pressure of the system under the glass fiber nylon loading device is 3100 N, and the corresponding optimal operating frequency range is 38.3 kHz to 38.4 kHz.

[0201] In some other embodiments, after detecting parameter drift, the system may adopt the following compensation strategy:

[0202] When coupling pressure drift is detected, the system issues a warning signal, instructing the operator to check the tightness of the loading device, or automatically adjust the applied force to restore the active power to within the tolerance range. Specifically, if ΔW < 0 (power decrease), the initial coupling pressure should be appropriately increased or decreased to bring the system back close to the optimal matching state.

[0203] When frequency drift is detected, the system can activate the automatic frequency tracking function to perform a small-range frequency sweep around the current frequency to find the frequency corresponding to the new power maximum point, and adjust the generator's output frequency to that new frequency. If the generator supports real-time frequency phase-locked tracking, this function can be directly enabled to achieve dynamic frequency compensation.

[0204] In some other embodiments, the following anomalies may be encountered during the optimal operating point search process described above:

[0205] If no stationary point (i.e., the active power response function changes monotonically) is identified within the entire space of excitation control parameters to be searched, it indicates that the current search range does not cover the optimal operating point, and the system should prompt the user to expand the parameter search range or check the device connection status.

[0206] If the active power values ​​corresponding to the identified multiple stagnation points are very similar, for example, the difference is less than 2% of the maximum value, it indicates that there are multiple approximately equivalent operating points in the system. In this case, any one of them can be selected as the initial optimal operating point, or the stagnation point located in the central area of ​​the search range can be selected first to retain greater adjustment leeway.

[0207] The above-described solution employs a multi-physics parameter spatial scanning and adaptive optimization strategy to address the problem of low energy transfer efficiency caused by the drift of the optimal operating point with load variations. Specifically, it traverses and scans within a preset pressure and frequency space, using active power or thermal imaging temperature rise as performance indicators to capture the specific optimal operating point (e.g., 2900N / 38.1kHz and 3100N / 38.3kHz) under different loading devices (e.g., aluminum alloy loading devices or glass fiber reinforced nylon loading devices). In particular, it introduces differential optimization and uniqueness determination logic based on active power, ensuring the system is always locked in the state of maximum power transfer, thus eliminating blind reliance on nominal parameters.

[0208] This scheme also proposes a second-order sensitivity analysis and decoupling monitoring mechanism based on the Hessian matrix, solving the problems of lack of tolerance basis for parameter adjustment and inability to distinguish the root cause of drift. A second-order performance surface is constructed, and the normalized curvature coefficients are calculated to quantify the sensitivity of system performance to frequency and pressure. Based on this, a scientific tolerance elliptical boundary is generated, achieving strict control of sensitive dimensions and relaxed management of non-sensitive dimensions. Simultaneously, utilizing the physical characteristics of voltage and current phase difference being sensitive to frequency and active power being sensitive to pressure, a power-phase drift decoupling diagnostic method is established. This method can distinguish and locate frequency drift or pressure relaxation in real time, ensuring the long-term stability and high robustness of the detection process.

[0209] It should be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

Claims

1. An adaptive configuration method for the optimal operating point of an ultrasonic thermal excitation system for hidden cracks in concrete, characterized in that, include: Obtain the excitation control parameter space of the ultrasonic thermal excitation system to be searched. The excitation control parameter space shall include at least the initial coupling pressure dimension and the operating frequency dimension. The ultrasonic thermal excitation system is driven to operate within the excitation control parameter space to be searched, and multi-physics response data of the system under different parameter combinations are collected; among which, the multi-physics response data are electric power data or infrared thermal image temperature data; Based on multiphysics response data, an excitation effectiveness evaluation index is constructed. The parameter points that make the excitation effectiveness evaluation index meet the extreme value conditions are searched and determined as the initial optimal operating points. Within the neighborhood of the initial optimal operating point, the sensitivity characteristics of the excitation performance evaluation index relative to the excitation control parameters are calculated. Based on the sensitivity characteristics, the parameter tolerance boundary is determined, and the optimal operating point configuration including the parameter center value and the parameter tolerance range is obtained. The calculation of the sensitivity characteristics of the incentive performance evaluation index relative to the incentive control parameters includes: establishing a second-order performance surface model of the incentive performance evaluation index with respect to the initial coupling pressure dimension and the operating frequency dimension, centered on the initial optimal operating point; calculating the Hessian matrix of the second-order performance surface model at the initial optimal operating point; performing eigenvalue decomposition on the Hessian matrix and extracting the normalized curvature coefficients as sensitivity characteristics characterizing the sensitivity of parameter control. The parameter tolerance boundary is determined based on sensitivity characteristics, including: constructing a tolerance ellipse model centered on the initial optimal operating point based on the eigenvalues ​​and eigenvectors of the Hessian matrix; calculating the projection intercepts of the tolerance ellipse model on the initial coupling pressure dimension axis and the operating frequency dimension axis; and determining the projection intercepts as the allowable tolerance ranges of the initial coupling pressure parameter and the operating frequency parameter, respectively, wherein the dimension axis with larger eigenvalues ​​has a smaller allowable tolerance range.

2. The method according to claim 1, characterized in that, The multiphysics response data is electrical power data, which includes voltage data, current data, and active power data under different excitation and control parameters. The excitation efficiency evaluation index is constructed based on multi-physics response data, including: under fixed initial coupling pressure parameters, calculating active power data corresponding to different frequencies based on voltage and current data obtained by frequency sweep, constructing a response function of active power as a function of frequency, and obtaining active power response; extracting the peak value or its statistical characteristics of the active power response, and using it as an excitation efficiency evaluation index characterizing the system's effective power conversion capability.

3. The method according to claim 2, characterized in that, Search for parameter points that allow the incentive effectiveness evaluation index to satisfy extreme value conditions, including: Call the active power response function that varies with the initial coupling pressure parameters; Identify the stagnation point where the first-order difference sign of the active power response function changes from positive to negative with respect to the initial coupling pressure, and determine the pressure corresponding to the stagnation point as the optimal initial coupling pressure; Under optimal initial coupling pressure, the active power response function as a function of operating frequency is analyzed, and the frequency at which the maximum value is reached is determined as the optimal operating frequency. The optimal initial coupling pressure and the optimal operating frequency are combined to form the initial optimal operating point.

4. The method according to claim 1, characterized in that, Also includes: Based on the optimal operating point configuration, the system's operational status characteristics are monitored in real time during the probing process, and the parameter drift type is identified according to these characteristics. Among them, the operational status characteristics of the real-time monitoring system during the detection process include: The system acquires the current voltage and current data in real time, calculates the corresponding active power and voltage-current phase difference, and compares them with the active power and phase difference at the initial optimal operating point to obtain the change in active power and phase change. When the change in active power exceeds a preset power threshold, but the change in phase does not exceed a preset phase threshold, the system is determined to have experienced coupling pressure drift. When the phase change exceeds the preset phase threshold, the system is determined to have experienced operating frequency drift, regardless of whether the active power change exceeds the preset power threshold.

5. The method according to claim 1, characterized in that, The multiphysics response data is infrared thermal imaging temperature data, which includes a sequence of thermal images obtained by infrared imaging the surface of the object under different excitation control parameters. Excitation effectiveness evaluation indicators are constructed based on multiphysics response data, including: Identify the heat-generating regions before and after the excitation of hidden cracks in the thermal image sequence; The average temperature difference between the heat-generating region after excitation of the hidden crack and the background heat-generating region before excitation is calculated and used as an evaluation index of excitation efficiency characterizing the thermal excitation conversion efficiency of the system.

6. The method according to claim 5, characterized in that, Search for parameter points that allow the incentive effectiveness evaluation index to satisfy extreme value conditions, including: Under discrete initial coupling pressure levels, frequency sweep excitation is performed within a preset frequency range. Compare the average temperature difference under different combinations of pressure levels and frequencies; The combination of pressure level and frequency that produces the maximum average temperature difference is determined as the initial optimal operating point.

7. The method according to claim 1, characterized in that, The ultrasonic thermal excitation system includes a self-excited topological ultrasonic generator that supports wideband adjustment from 17kHz to 135kHz and a PZT-4 sandwich piezoelectric transducer with a nominal frequency of 40kHz and a rated power of 50W. The loading device is selected from an aluminum alloy loading device or a glass fiber nylon loading device.

8. The method according to claim 7, characterized in that, When using an aluminum alloy loading device, the initial optimal operating point determined by the method is: an initial coupling pressure of 2900N and an operating frequency range of 38.1kHz to 38.2kHz. When using a glass fiber reinforced nylon loading device, the initial optimal operating point determined by the method is: an initial coupling pressure of 3100N and an operating frequency range of 38.3kHz to 38.4kHz.

Images