An ultrasonic-based method for detecting the thickness of a waterproof layer of an external wall
By using baseline correction and time-frequency domain feature fusion, the problem of waterproof layer thickness detection error caused by hardware error of ultrasonic testing equipment was solved, and accurate detection of the thickness of the external wall waterproof layer was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG SECOND CONSTR GRP CO LTD
- Filing Date
- 2026-04-28
- Publication Date
- 2026-07-03
AI Technical Summary
The hardware characteristics of existing ultrasonic testing equipment introduce systematic errors, resulting in significant errors in the thickness detection results of the external wall waterproofing layer. It is impossible to accurately locate the initial wave peak reflected from the surface of the waterproofing layer, leading to calculation errors.
By eliminating systematic errors through baseline correction, identifying the initial peak of reflection from the surface of the waterproof layer, extracting effective signal segments, and combining time-domain and frequency-domain characteristics to select the target echo interval, the two-way travel time is calculated to accurately detect the thickness of the waterproof layer.
Accurately extracting effective signal segments reduces the impact of interference signals, ensuring the relevance and effectiveness of echo interval identification and improving the accuracy of waterproof layer thickness detection.
Smart Images
Figure CN122108016B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of non-destructive testing technology in building engineering, specifically a method for detecting the thickness of exterior wall waterproofing layers based on ultrasonic waves. Background Technology
[0002] Currently, most methods for detecting the thickness of exterior wall waterproofing layers use ultrasonic testing. This involves directly acquiring the original ultrasonic echo signal of the waterproofing layer under test using ultrasonic testing equipment. Without performing baseline correction on the original signal, the method relies solely on a single signal feature to identify the interface echo between the waterproofing layer and the substrate. The thickness of the waterproofing layer is then calculated based on the time interval between the reflected wave peak on the waterproofing layer surface and the interface echo, combined with the ultrasonic wave propagation speed.
[0003] The inherent systematic errors in the hardware of ultrasonic testing equipment directly superimpose on the original ultrasonic echo signal, altering its true waveform and amplitude. Current testing methods cannot accurately pinpoint the initial peak of reflection from the waterproofing layer surface, nor can they extract effective signal segments containing only the echo information from the waterproofing layer interface. A large amount of redundant interference signals participate in subsequent echo analysis. Relying solely on single signal characteristics to divide echo intervals fails to accurately distinguish between the main interface echo between the waterproofing layer and the substrate and other interfering echoes. Significant biases exist in the selection of candidate echo intervals, directly causing errors in the two-way travel time calculation of ultrasonic waves propagating within the waterproofing layer, resulting in substantial errors in the waterproofing layer thickness detection results.
[0004] This invention addresses the shortcomings of existing detection technologies by eliminating system errors introduced by hardware characteristics, accurately extracting effective echo signal segments, and filtering out the target echo interval corresponding to the main interface through the fusion of time-domain and frequency-domain features, thereby achieving accurate detection of the thickness of the external wall waterproofing layer. Summary of the Invention
[0005] This invention aims to solve at least one of the technical problems existing in the prior art;
[0006] Therefore, this invention proposes an ultrasonic-based method for detecting the thickness of an exterior wall waterproofing layer, comprising:
[0007] Acquire the raw ultrasonic echo signal collected by the ultrasonic testing equipment on the surface of the waterproof layer of the exterior wall to be tested;
[0008] The original ultrasonic echo signal is baseline corrected to eliminate systematic errors introduced by the hardware characteristics of the ultrasonic testing equipment, resulting in a preprocessed ultrasonic echo signal.
[0009] Identify the initial peak position in the preprocessed ultrasonic echo signal corresponding to the reflection from the surface of the waterproof layer, and extract an effective signal segment containing the interface echo information of the waterproof layer from the preprocessed ultrasonic echo signal based on the initial peak position.
[0010] Determine the signal envelope of the effective signal segment, and identify multiple local peak and trough positions from the signal envelope;
[0011] Based on the local peak and trough locations, multiple candidate echo intervals are divided in the effective signal segment;
[0012] Calculate the fusion feature value of the time domain features and frequency domain features of each candidate echo interval, and select the target echo interval representing the main interface between the waterproof layer and the substrate from multiple candidate echo intervals based on the fusion feature value.
[0013] The time interval between the center position of the target echo interval in the effective signal segment and the initial peak position is calculated as the two-way travel time of the ultrasonic wave propagating in the waterproof layer.
[0014] The thickness of the waterproof layer on the exterior wall to be tested is calculated based on the known propagation speed of the ultrasonic wave and the two-way travel time.
[0015] Further, baseline correction of the original ultrasound echo signal includes:
[0016] In free space conditions without a target to be measured, a reference background signal is acquired using the ultrasonic detection device.
[0017] The baseline offset is obtained by calculating the average amplitude of the reference background signal over a preset time period.
[0018] The baseline offset is subtracted from the amplitude of each sampling point of the original ultrasonic echo signal to complete the baseline correction, thereby obtaining the preprocessed ultrasonic echo signal.
[0019] Furthermore, identifying multiple local peak and trough locations from the signal envelope includes:
[0020] The first-order difference of the signal envelope is calculated to obtain the slope change sequence of the signal envelope;
[0021] In the slope change sequence, find the zero-crossing point where the slope sign changes from positive to negative, and mark the position of the zero-crossing point in the signal envelope as a local peak position;
[0022] In the slope change sequence, find the zero-crossing point where the slope sign changes from negative to positive, and mark the position of the zero-crossing point in the signal envelope as a local trough position;
[0023] By traversing the entire slope change sequence, all local peak positions and local trough positions in the signal envelope are identified.
[0024] Furthermore, based on the locations of the local peaks and troughs, multiple candidate echo intervals are divided within the effective signal segment, including:
[0025] According to the time sequence, all the identified local peak and trough positions are sorted to obtain an alternating sequence of peak and trough positions.
[0026] In the position sequence, starting from any local peak position, it extends to the next adjacent local trough position, forming a half-cycle interval from the peak to the trough.
[0027] The half-cycle interval is extended forward and backward by a preset number of sample points to form an initial interval with the local peak position as the core.
[0028] Merge all initial intervals that overlap on the time axis, and define each merged or unmerged initial interval as a candidate echo interval.
[0029] Further, calculating the fused feature value of the time-domain and frequency-domain features for each candidate echo interval includes:
[0030] Perform a short-time Fourier transform on the signal within the candidate echo interval to obtain its spectral distribution;
[0031] Calculate the time-domain energy of the candidate echo interval signal, and calculate the frequency-domain energy of its spectrum within the preset waterproof layer characteristic frequency band;
[0032] The time-domain energy and the frequency-domain energy are respectively normalized.
[0033] The normalized time-domain energy and frequency-domain energy are weighted and summed, and the resulting sum is the fused characteristic value of the candidate echo interval.
[0034] Furthermore, selecting the target echo interval representing the main interface between the waterproof layer and the substrate from multiple candidate echo intervals based on the fusion feature value includes:
[0035] A fusion feature threshold is set, which is obtained based on the echo interval characteristics of a known standard waterproof layer sample.
[0036] Compare the fusion feature value of each candidate echo interval with the fusion feature threshold;
[0037] The candidate echo interval with the fusion feature value greater than the fusion feature threshold and the largest fusion feature value among all candidate echo intervals that meet this condition is selected and determined as the target echo interval.
[0038] Further, calculating the time interval between the center position of the target echo interval in the effective signal segment and the initial peak position includes:
[0039] Determine the start and end time points of the target echo interval within the valid signal segment;
[0040] Calculate the average of the start time point and the end time point to obtain the center time point of the target echo interval;
[0041] Calculate the time point corresponding to the initial peak position;
[0042] Calculate the difference between the center time point and the time point corresponding to the initial wave peak position. The difference is the two-way travel time of the ultrasonic wave propagating in the waterproof layer.
[0043] Furthermore, based on the known propagation speed of ultrasound and the two-way travel time, the thickness of the waterproof layer on the exterior wall to be measured is calculated, including:
[0044] Obtain the average propagation speed of ultrasonic waves in the material constituting the waterproof layer of the exterior wall to be tested;
[0045] Multiplying the two-way travel time by the average propagation speed yields the total path length of the ultrasonic wave propagating back and forth in the waterproof layer material.
[0046] Divide the total path length by two to get the thickness of the waterproof layer on the exterior wall to be measured.
[0047] Furthermore, acquiring the raw ultrasonic echo signals collected by the ultrasonic testing equipment on the surface of the waterproof layer of the exterior wall to be tested includes:
[0048] The transducer probe of the ultrasonic testing equipment is attached to the clean surface of the waterproof layer on the exterior wall to be tested.
[0049] The transducer probe is controlled by an ultrasonic testing device to emit ultrasonic pulse signals with a specific center frequency and bandwidth.
[0050] The transducer probe receives ultrasonic echo signals reflected from inside the waterproof layer of the exterior wall under test.
[0051] The received ultrasonic echo signal is converted from analog to digital to generate a digital original ultrasonic echo signal.
[0052] Furthermore, the step of attaching the transducer probe of the ultrasonic testing equipment to the clean surface of the exterior wall waterproofing layer to be tested includes:
[0053] Apply a sufficient amount of acoustic coupling agent to the acoustic wave emitting end of the transducer probe to ensure that the acoustic coupling agent evenly covers the entire acoustic wave emitting end face;
[0054] Place the transducer probe stably on the clean surface of the waterproof layer of the exterior wall to be tested, and apply a moderate pressure perpendicular to the surface so that the acoustic coupling agent forms a uniform and bubble-free film between the probe end face and the surface of the waterproof layer.
[0055] The transducer probe is kept in a stable position and orientation at the detection point, and no relative movement occurs between it and the surface of the waterproof layer of the exterior wall to be tested during the acquisition of the original ultrasonic echo signal.
[0056] Compared with the prior art, the beneficial effects of the present invention are:
[0057] Baseline correction of the raw ultrasonic echo signal directly eliminates systematic errors introduced by the hardware characteristics of the ultrasonic testing equipment, removing interference from hardware deviations on the waveform and amplitude of the raw ultrasonic echo signal. This ensures that the pre-processed ultrasonic echo signal can completely and accurately reflect the propagation state of ultrasound within the waterproofing layer of the exterior wall under test. Based on the identified initial peak position of the reflection from the waterproofing layer surface, effective signal segments can be accurately extracted from the pre-processed ultrasonic echo signal. This operation directly filters out redundant signals that do not contain echo information from the waterproofing layer interface, retaining only the signal content related to the reflection from the interface between the waterproofing layer and the substrate. This reduces the interference of irrelevant signals on subsequent echo interval identification, feature calculation, and other processes, ensuring that the data used for subsequent signal analysis is targeted and effective.
[0058] Extracting the signal envelope of the effective signal segment clearly reveals the amplitude variation pattern of the signal. By identifying the local peaks and troughs within the signal envelope, the echo region within the effective signal segment can be finely divided, forming multiple candidate echo intervals. Calculating the time-domain and frequency-domain characteristics of each candidate echo interval and obtaining a fused feature value allows for the comprehensive utilization of signal representation information in both time and frequency dimensions, avoiding the limitations of single-feature judgment. Filtering the target echo interval based on the fused feature value accurately pinpoints the echo interval at the main interface between the waterproof layer and the substrate, effectively distinguishing the main interface echo from other interfering echoes and preventing non-target echoes from affecting the detection results. Accurately obtaining the time interval between the center position and the initial peak position of the target echo interval ensures the accuracy of the two-way travel time calculation of ultrasonic waves propagating in the waterproof layer, thereby improving the accuracy of the calculated thickness of the external wall waterproof layer. Attached Figure Description
[0059] Figure 1 This is a flowchart illustrating the steps of an ultrasonic-based method for detecting the thickness of an external wall waterproofing layer according to the present invention.
[0060] Figure 2 A flowchart for identifying the locations of local peaks and troughs in the signal envelope;
[0061] Figure 3 A flowchart for calculating the fusion feature values of candidate echo intervals;
[0062] Figure 4 A graph showing the relationship between two-way travel time and thickness for different waterproofing materials;
[0063] Figure 5 This is a diagram showing the baseline correction effect of the ultrasonic echo signal. Detailed Implementation
[0064] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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 are within the scope of protection of the present invention.
[0065] See Figure 1 The process involves acquiring raw ultrasonic echo signals from the surface of the waterproofing layer on the exterior wall under test using an ultrasonic testing device. Baseline correction is performed on this raw ultrasonic echo signal to eliminate systematic errors introduced by the hardware characteristics of the ultrasonic testing device, resulting in a preprocessed ultrasonic echo signal. The initial peak position corresponding to the reflection from the waterproofing layer surface is identified in the preprocessed ultrasonic echo signal, and an effective signal segment containing the interface echo information of the waterproofing layer is extracted from the preprocessed ultrasonic echo signal based on this initial peak position. The signal envelope of this effective signal segment is determined, and multiple local peak and trough positions are identified from this signal envelope. Based on these local peak and trough positions, multiple candidate echo intervals are divided within the effective signal segment. The fusion feature value of the time-domain and frequency-domain characteristics of each candidate echo interval is calculated, and the target echo interval representing the main interface between the waterproofing layer and the substrate is selected from the multiple candidate echo intervals based on this fusion feature value. The time interval between the center position of the target echo interval in the effective signal segment and the initial peak position is calculated as the two-way travel time of the ultrasonic wave propagating in the waterproofing layer. Based on the known propagation speed of ultrasound and the two-way travel time, the thickness of the waterproof layer on the exterior wall to be measured is calculated.
[0066] In one embodiment of the present invention, a reference background signal is acquired using an ultrasonic testing device under free-space conditions without a target. The average amplitude of the reference background signal over a preset time period is calculated to obtain the baseline offset. The baseline offset is subtracted from the amplitude of each sampling point of the original ultrasonic echo signal to complete baseline correction, resulting in a preprocessed ultrasonic echo signal. In a specific implementation, during the baseline correction operation, the ultrasonic testing device acquires the background signal in free space without any target. Free space conditions refer to the absence of solid medium contact at the probe tip, with coupling only to air. In a specific implementation, the ultrasonic testing device controls the transducer probe to emit and receive ultrasonic signals under free-space conditions, and the acquired original voltage sequence is the reference background signal. The reference background signal reflects the circuit noise of the ultrasonic testing device itself, the spontaneous vibration of the transducer, and the environmental background signal, and in terms of signal morphology, it is a DC bias fluctuating near zero. In practice, the reference background signal is processed, and a signal segment of a preset time length is extracted. This preset time length should cover the main stable section of the signal. The arithmetic mean of the amplitudes of all sampling points within this signal segment is calculated, and the result is defined as the baseline offset. The baseline offset is a numerical value that characterizes the overall DC offset level of the signal.
[0067] In practice, the baseline offset is subtracted from the amplitude of each sampling point of the original ultrasonic echo signal. It can be understood that the original ultrasonic echo signal is a mixed signal containing useful echo information and the system's inherent DC offset. Subtracting the baseline offset from each sampling point eliminates the fixed-level offset introduced by hardware circuit characteristics. The baseline correction process is performed point-by-point, and the processed signal sequence is the preprocessed ultrasonic echo signal. The zero baseline of the preprocessed ultrasonic echo signal is corrected, making the determination of peak and trough positions in subsequent signal analysis more accurate and unaffected by the instrument's own DC bias. The baseline offset is calculated using the following formula:
[0068]
[0069] Where: symbol Represents the calculated baseline offset, symbol This indicates the total number of sampling points of the reference background signal within a preset time period, with the symbol... Indicates the first reference background signal The amplitude of each sampling point. In practical implementation, the selection of the preset time length needs to ensure that the reference background signal is in a stable state. Typically, a stable interval is selected after the ultrasonic pulse emission ends and there is no external reflected signal interference for calculation. After completing the baseline offset subtraction operation, the amplitude of each sampling point in the original ultrasonic echo signal becomes... ,in It is the first of the original ultrasonic echo signals Amplitude at each sampling point, It is the amplitude after correction for the corresponding sampling point, composed of all The resulting sequence is the preprocessed ultrasonic echo signal.
[0070] In one embodiment of the present invention, see [reference] Figure 2 The first-order difference of the signal envelope is calculated to obtain the slope change sequence of the signal envelope. Within this slope change sequence, zero-crossing points where the slope sign changes from positive to negative are identified, and the corresponding positions in the signal envelope are marked as local peak positions. Similarly, zero-crossing points where the slope sign changes from negative to positive are identified, and the corresponding positions are marked as local trough positions. The entire slope change sequence is traversed to identify all local peak and trough positions in the signal envelope. These identified local peak and trough positions are sorted chronologically to obtain an alternating sequence of peak and trough positions. Starting from any local peak position, the sequence extends to the next adjacent local trough position, forming a half-period interval descending from a peak to a trough. This half-period interval is then expanded forward and backward by a predetermined number of sample points to form an initial interval centered on the local peak position. Merge all initial intervals that overlap on the time axis, and define each merged or unmerged initial interval as a candidate echo interval.
[0071] In specific implementations, some embodiments involve performing first-order difference calculations on the signal envelope. The signal envelope is a curve reflecting the amplitude changes of the ultrasonic echo signal, and first-order difference calculation involves calculating the amplitude difference between adjacent sampling points in the signal envelope sequence. It can be understood that first-order difference calculation can reveal the slope change of the signal envelope. By calculating the difference value at each position, a slope change sequence of the signal envelope is obtained, where each element of the slope change sequence represents the instantaneous rate of change of the signal envelope at that point. In specific implementations, a traversal search operation is performed in the slope change sequence to find the zero-crossing point where the slope sign changes from positive to negative. It can be understood that a zero-crossing point refers to the specific location in the slope change sequence where the value changes from positive to zero and then to negative. When it is detected that the value at a certain position in the slope change sequence changes from a positive value at the previous position to a zero or negative value at the current position, and subsequent positions remain negative, this position is marked as a zero-crossing point where the slope sign changes from positive to negative. In practice, the position corresponding to the zero-crossing point in the signal envelope is marked as a local peak position, representing the turning point where the signal envelope transitions from rising to falling. In practice, within the slope change sequence, zero-crossing points where the slope sign changes from negative to positive are searched. When a value at a certain position in the slope change sequence changes from a negative value at the previous position to a zero or positive value at the current position, and subsequent positions remain positive, this position is identified as a zero-crossing point where the slope sign changes from negative to positive. In practice, the position corresponding to this zero-crossing point in the signal envelope is marked as a local trough position, representing the turning point where the signal envelope transitions from falling to rising. By traversing the entire slope change sequence, all local peak and trough positions in the signal envelope are identified; these positions constitute the basic set of positions for subsequent division of candidate echo intervals.
[0072] In specific implementation, all identified local peak and trough positions are sorted chronologically. Local peak and trough positions alternate on the time axis, resulting in a sequential sequence of alternating peak and trough positions. Within this sequence, starting from any local peak position and ending at the first adjacent local trough position, a half-cycle interval is formed, traversing from peak to trough. This half-cycle interval covers the signal envelope's descent from a peak to an adjacent trough. This half-cycle interval is then expanded forward and backward by a preset number of sample points, which can be set based on the echo signal width. This expansion creates an initial interval centered on the local peak position. This initial interval contains the complete waveform information of the peak and its immediate vicinity, preventing the loss of effective echo information due to boundary truncation. In some embodiments, all overlapping initial intervals on the time axis are merged. When two or more initial intervals intersect or contain each other within a time range, they are merged into a larger continuous interval. In practice, each initial interval, whether merged or not, is defined as a candidate echo interval. A candidate echo interval represents the temporal distribution range of a possible interface reflection echo on the signal envelope. The division of candidate echo intervals does not rely on threshold judgments for individual sampling points, but is based on structural analysis of the signal envelope shape, which better adapts to changes in signal amplitude.
[0073] In practice, the number and range of candidate echo intervals are determined by the number and distribution of local peaks and troughs. One candidate echo interval corresponds to one local peak, and multiple adjacent local peaks may be merged into one candidate echo interval due to interval overlap. The central difference method is used to calculate the slope change sequence of the signal envelope. The formula for the central difference method is as follows:
[0074]
[0075] Where: symbol Indicates the signal envelope sequence at the th The first-order difference value at the nth sampling point, i.e., the first-order difference value of the slope change sequence. One element, symbol Indicates the first element in the signal envelope sequence. The amplitude of each sampling point, sign Indicates the first element in the signal envelope sequence. The amplitude at each sampling point. Slope change sequence. Used for subsequent detection of zero crossing.
[0076] In one embodiment of the present invention, see [reference] Figure 3The signal within the candidate echo interval is subjected to a short-time Fourier transform to obtain its spectral distribution. The time-domain energy of the candidate echo interval signal and its frequency-domain energy within the preset characteristic frequency band of the waterproof layer are calculated. Both the time-domain and frequency-domain energies are normalized. The normalized time-domain and frequency-domain energies are then weighted and summed; the sum is the fusion characteristic value of the candidate echo interval. A fusion characteristic threshold is set, which is obtained based on the echo interval characteristics of a known standard waterproof layer sample. The fusion characteristic value of each candidate echo interval is compared with this fusion characteristic threshold. The candidate echo interval with the largest fusion characteristic value among all candidate echo intervals satisfying this condition is selected as the target echo interval.
[0077] In practical implementation, a short-time Fourier transform (SFT) is performed on the signal within the candidate echo interval. The candidate echo interval signal is a finite-length ultrasonic echo sequence truncated in the time domain. The SFT obtains its spectral distribution by windowing the candidate echo interval signal and calculating its spectrum. The spectral distribution describes the energy intensity of the signal at different frequency components within the interval. In practical implementation, the time-domain energy of the candidate echo interval signal is calculated, representing the total intensity of the signal on the time axis. In practical implementation, the frequency-domain energy of the spectrum within a preset waterproof layer characteristic frequency band is calculated. The preset waterproof layer characteristic frequency band is a pre-defined frequency range based on the acoustic characteristics of ultrasonic wave propagation in typical waterproof layer materials. The frequency-domain energy represents the concentration of the signal within this preset waterproof layer characteristic frequency band. In practical implementation, both the time-domain and frequency-domain energies are normalized. Normalization eliminates the influence of absolute amplitude differences in the signal, mapping the time-domain and frequency-domain energies to a unified numerical range. In practice, the normalized time-domain energy and frequency-domain energy are weighted and summed. The weighting coefficients can be set, and the resulting sum is the fusion characteristic value of the candidate echo interval. The fusion characteristic value comprehensively reflects the intensity and frequency distribution characteristics of the candidate echo interval signal.
[0078] In specific implementations, a fusion feature threshold is set, which is a preset numerical judgment standard. In some embodiments, the fusion feature threshold is obtained based on the echo interval characteristics of a known standard waterproof layer sample. The known standard waterproof layer sample refers to a standard test block with a known thickness and material. The fusion feature threshold is determined by collecting its ultrasonic echo and calculating the fusion feature value of the corresponding echo interval, followed by statistical analysis. In specific implementations, the fusion feature value of each candidate echo interval is compared with the fusion feature threshold, and the comparison operation traverses all the divided candidate echo intervals. In specific implementations, the candidate echo interval with a fusion feature value greater than the fusion feature threshold and having the largest fusion feature value among all candidate echo intervals that meet this condition is selected. The selected candidate echo interval is determined as the target echo interval, which is considered to be most likely to represent the main interface reflection signal between the waterproof layer and the substrate.
[0079] It is understandable that comparing the fusion characteristic value of each candidate echo interval with the fusion characteristic threshold filters out weak reflection or interference signal intervals with fusion characteristic values below the threshold. These weak reflection or interference signals may originate from inhomogeneous structures or clutter within the waterproofing layer. It is also understandable that selecting the candidate echo interval with the largest fusion characteristic value among all candidate echo intervals with fusion characteristic values greater than the fusion characteristic threshold implies that the target echo interval possesses the strongest time-frequency domain comprehensive characteristics within the frequency band of interest. This aligns with the characteristic that reflections from the main interface typically have strong energy and concentrated frequency components. In some embodiments, the calibration of the fusion characteristic threshold needs to be performed on representative samples of various known standard waterproofing layers to obtain a universally applicable threshold level. In specific implementations, the weighted summation formula can be expressed as:
[0080]
[0081] Where: symbol The calculated fusion feature value is represented by the symbol. Represents the weighting coefficients assigned to the normalized time-domain energy, with the symbol... Represents the time-domain energy after normalization, with the symbol... This represents the weighting coefficient assigned to the normalized frequency domain energy, with the symbol... This represents the frequency domain energy after normalization. Weighting coefficients. and The sum can be set to a fixed value, such as 1, to ensure the comparability of fused feature values.
[0082] In one embodiment of the present invention, the start and end time points of the target echo interval within the effective signal segment are determined. The average of the start and end time points is calculated to obtain the center time point of the target echo interval. The time point corresponding to the initial peak position is calculated. The difference between the center time point and the time point corresponding to the initial peak position is calculated; this difference is the two-way travel time of the ultrasonic wave propagating in the waterproof layer. The average propagation speed of the ultrasonic wave in the material constituting the waterproof layer of the exterior wall to be tested is obtained. The two-way travel time is multiplied by the average propagation speed to obtain the total path length of the ultrasonic wave propagating round trip in the waterproof layer material. The total path length is divided by two to obtain the thickness of the waterproof layer of the exterior wall to be tested.
[0083] In practical implementation, the start and end time points of the target echo interval within the effective signal segment are determined. The target echo interval is the interval representing the reflection from the main interface, selected from candidate echo intervals. The start and end time points directly correspond to the start and end sampling point numbers of the target echo interval in the digital signal, which can be converted into specific time values by combining the signal sampling frequency. In practical implementation, the average value between the start and end time points is calculated. The average value is the time value at the midpoint between the start and end time points, and the calculation result is the center time point of the target echo interval. The center time point serves as an estimate of the position of the main interface reflected echo in the time domain. In practical implementation, the time point corresponding to the initial peak position is calculated. The initial peak position is the first obvious peak corresponding to the reflection from the waterproof layer surface in the preprocessed ultrasonic echo signal. In practical implementation, the difference between the center time point and the time point corresponding to the initial peak position is calculated. The difference represents the time it takes for the ultrasonic wave to propagate from the waterproof layer surface to the interface between the waterproof layer and the substrate and back to the surface. The difference is the two-way travel time of the ultrasonic wave propagating in the waterproof layer.
[0084] In specific implementations, the average propagation speed of ultrasonic waves in the material constituting the waterproof layer of the exterior wall to be tested is obtained. The average propagation speed is the velocity of sound when ultrasonic waves propagate in a specific waterproof layer material. In some embodiments, the average propagation speed can be calculated by measuring a sample of the same material with a known thickness and combining it with the propagation time of ultrasonic waves within it. In specific implementations, the two-way travel time is multiplied by the average propagation speed; the product is the total path length traveled by the ultrasonic wave in one round trip within the waterproof layer material, and the total path length is equal to twice the thickness of the waterproof layer. In some embodiments, the total path length is divided by two, and the quotient is the thickness of the waterproof layer to be tested. It can be understood that calculating the thickness of a medium by measuring the propagation time of ultrasonic waves in the medium and the known velocity of sound is based on the fundamental physical principle of ultrasonic thickness measurement. In specific implementations, the calculation of the center time point can avoid the positional deviation that may be caused by directly using a single extreme point such as a peak or trough. The center time point can better represent the energy center of the entire echo waveform, improving the robustness of the time interval measurement. It can be understood that the accuracy of the two-way travel time calculation method directly determines the accuracy of the final thickness calculation result.
[0085] In practice, the propagation speed of ultrasonic waves is related to factors such as the material type, temperature, and density of the waterproof layer. The average propagation speed needs to be matched to the actual waterproof layer material being tested. The thickness calculation process can be expressed using the following formula:
[0086]
[0087] Where: symbol This represents the calculated thickness of the exterior wall waterproofing layer, indicated by the symbol. The speed of sound propagation is indicated by the average speed of ultrasound waves in the material constituting the waterproof layer of the exterior wall being tested. The symbol is... This represents the calculated two-way travel time. The formula reflects the relationship that thickness equals the product of sound speed and propagation time. In practice, different materials have different average propagation velocities. Refer to Table 1, which shows the reference values of the average ultrasonic propagation velocity of several common exterior wall waterproofing materials under typical conditions.
[0088] Table 1: Reference Values for Average Ultrasonic Propagation Velocity of Several Common Exterior Wall Waterproofing Materials
[0089]
[0090] The data in Table 1 are reference values under typical conditions. In actual testing applications, the specific value of the average propagation speed should be determined by calibration or by querying relevant material acoustic parameter databases, based on the material type, formulation, state, and environmental conditions of the actual test object.
[0091] See Figure 4This is a graph showing the relationship between the ultrasonic two-way travel time and thickness of different waterproofing materials. It illustrates the variation of ultrasonic two-way travel time for four common exterior wall waterproofing materials at different thicknesses, representing the core visualization result of ultrasonic thickness measurement. The differences in the curve slopes of different materials intuitively verify the correspondence between sound velocity and material type, providing a visual reference for selecting the correct sound velocity parameter during on-site testing and avoiding thickness errors caused by incorrect sound velocity selection. The curve slope can predict the time measurement sensitivity of different materials, allowing for the establishment of differentiated testing accuracy standards for different materials. It can serve as a visual reference for the quality acceptance of waterproofing layer construction, comparing the measured travel time with the theoretical curve to quickly determine whether the thickness meets the standard, significantly improving acceptance efficiency. Given the two-way travel time and sound velocity, the thickness can be measured using the formula d=v. The thickness of the waterproof layer is accurately calculated using t / 2. When the thickness is known, the material type can be deduced by measuring the travel time, aiding in quality acceptance. Different materials exhibit significant differences in travel time, providing a basis for setting parameters for ultrasonic testing equipment and avoiding thickness calculation errors caused by material misjudgment.
[0092] In one embodiment of the invention, the transducer probe of an ultrasonic testing device is attached to a clean surface of the waterproof layer of the exterior wall to be tested. The ultrasonic testing device controls the transducer probe to emit ultrasonic pulse signals with a specific center frequency and bandwidth. The transducer probe receives ultrasonic echo signals reflected from inside the waterproof layer of the exterior wall to be tested. The received ultrasonic echo signals are converted from analog to digital to generate digitized raw ultrasonic echo signals. Sufficient acoustic coupling agent is applied to the acoustic wave emitting end of the transducer probe to ensure that the acoustic coupling agent uniformly covers the entire acoustic wave emitting end face. The transducer probe is placed stably on the clean surface of the waterproof layer of the exterior wall to be tested, and a moderate pressure perpendicular to the surface is applied to form a uniform and bubble-free film of acoustic coupling agent between the probe end face and the surface of the waterproof layer. The position and orientation of the transducer probe at the test point are kept stable, and no relative movement occurs between the transducer probe and the surface of the waterproof layer of the exterior wall to be tested during the acquisition of raw ultrasonic echo signals.
[0093] In practice, the transducer probe of the ultrasonic testing equipment is attached to the clean surface of the waterproof layer on the exterior wall to be tested. The transducer probe is the core component for ultrasonic wave transmission and reception. The attachment process must ensure good acoustic contact between the transducer probe and the surface of the waterproof layer, avoiding any air gaps. The ultrasonic testing equipment controls the transducer probe to emit ultrasonic pulse signals with a specific center frequency and bandwidth, which are preset based on the acoustic characteristics and expected thickness range of the waterproof layer material. The transducer probe receives ultrasonic echo signals reflected from within the waterproof layer. Ultrasonic waves are reflected when they encounter the internal interface of the waterproof layer, and the reflected echoes are received by the same transducer probe. The received ultrasonic echo signals undergo analog-to-digital conversion, converting continuous analog voltage signals into discrete digital signal sequences, generating digitized raw ultrasonic echo signals for subsequent processing.
[0094] In practice, a sufficient amount of acoustic coupling agent is applied to the acoustic emitting end of the transducer probe. The acoustic coupling agent is a medium used to fill the gap between the probe and the surface under test. In some embodiments, the acoustic coupling agent is a viscous liquid or paste, such as water, glycerin, or a specialized ultrasonic coupling agent. Applying a sufficient amount of acoustic coupling agent ensures that it uniformly covers the entire acoustic emitting end face, which is the surface of the transducer that emits ultrasonic waves. In practice, the transducer probe is placed stably on the clean surface of the waterproof layer of the exterior wall to be tested, and moderate pressure is applied perpendicular to the surface. The purpose of applying pressure is to expel air between the probe end face and the surface of the waterproof layer and to spread the acoustic coupling agent into a uniform thin layer. Applying moderate pressure perpendicular to the surface causes the acoustic coupling agent to form a uniform, bubble-free film between the probe end face and the surface of the waterproof layer. This uniform, bubble-free film ensures that ultrasonic energy is effectively transmitted to the waterproof layer under test and that effective reflected signals are received. In practice, the transducer probe should be kept in a stable position and orientation at the detection point. During the acquisition of the original ultrasonic echo signal, the transducer probe should not move relative to the surface of the waterproof layer of the exterior wall to be tested. Relative movement will introduce additional noise or signal distortion.
[0095] It is understandable that cleaning the surface of the exterior wall waterproofing layer to be tested is to remove dust, oil, loose coatings, and other adhering substances. These adhering substances can hinder the formation of a uniform film by the acoustic coupling agent and attenuate the ultrasonic signal. It is also understandable that applying a moderate pressure perpendicular to the surface is necessary; too little pressure may result in air bubbles in the coupling layer, while too much pressure may compress the acoustic coupling agent, causing the coupling layer to become too thin or the probe to contact the surface, affecting the coupling effect. In specific implementations, the selection of a specific center frequency is related to the thickness of the waterproofing layer. Generally, thinner waterproofing layers require a higher center frequency to obtain better time resolution, while thicker waterproofing layers can choose a lower center frequency to obtain stronger penetration. In some embodiments, the bandwidth of the ultrasonic pulse signal determines the time-domain resolution of the signal. A specific center frequency and bandwidth can be achieved by controlling the parameters of the pulse generation module in the ultrasonic testing equipment. In specific implementations, the sampling frequency of the analog-to-digital conversion needs to satisfy the Nyquist sampling theorem, and the sampling frequency must be at least twice the highest frequency component of the ultrasonic pulse signal to ensure that no signal information is lost. In specific implementations, the accuracy of the analog-to-digital conversion, i.e., the number of bits of the analog-to-digital converter, affects the dynamic range and signal-to-noise ratio of the original ultrasonic echo signal. In practice, the acoustic impedance of the acoustic coupling agent should be matched as closely as possible to the acoustic impedance of the transducer probe and the waterproofing layer material under test to maximize the transmission efficiency of ultrasonic energy. During the acquisition of the original ultrasonic echo signal, the transducer probe should not move relative to the surface of the waterproofing layer under test. This requires the operator to maintain a stable handheld position or use a probe fixing device during signal acquisition.
[0096] See Figure 5 This is a baseline correction diagram of an ultrasonic echo signal, visually demonstrating the changes in the original echo signal after baseline correction during ultrasonic testing of exterior wall waterproofing layers. It is a core visualization result of the signal preprocessing stage. The original ultrasonic echo signal has a significant DC bias, including baseline offset errors introduced by the system hardware. After baseline correction, the signal mean is pulled close to 0 by subtracting the baseline offset, eliminating system errors. The baseline represents the 0 amplitude reference line, which is the ideal mean position of the corrected signal. The original signal is generally raised, while the corrected signal fluctuates around the 0 baseline, restoring the true echo amplitude characteristics. The shape and relative amplitude of key echo peaks such as surface reflection and interface reflection are completely preserved, and no detection features are lost due to correction. The noise distribution of the corrected signal is more uniform, facilitating subsequent peak identification and feature extraction. After eliminating the baseline offset, the position and amplitude measurement of echo peaks are more accurate, avoiding misjudgment of peaks or thickness calculation errors caused by baseline offset.
[0097] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.
Claims
1. An ultrasonic-based method for detecting the thickness of a waterproofing layer of an exterior wall, characterized by, The method is implemented based on ultrasonic testing equipment and includes: Acquire the raw ultrasonic echo signal collected by the ultrasonic testing equipment on the surface of the waterproof layer of the exterior wall to be tested; The original ultrasonic echo signal is baseline corrected to eliminate systematic errors introduced by the hardware characteristics of the ultrasonic testing equipment, resulting in a preprocessed ultrasonic echo signal. Identify the initial peak position in the preprocessed ultrasonic echo signal corresponding to the reflection from the surface of the waterproof layer, and extract an effective signal segment containing the interface echo information of the waterproof layer from the preprocessed ultrasonic echo signal based on the initial peak position. Determine the signal envelope of the effective signal segment, and identify multiple local peak and trough positions from the signal envelope; Based on the local peak and trough locations, multiple candidate echo intervals are divided in the effective signal segment; Calculate the fusion feature value of the time domain features and frequency domain features of each candidate echo interval, and select the target echo interval representing the main interface between the waterproof layer and the substrate from multiple candidate echo intervals based on the fusion feature value. The time interval between the center position of the target echo interval in the effective signal segment and the initial peak position is calculated as the two-way travel time of the ultrasonic wave propagating in the waterproof layer. The thickness of the waterproof layer on the exterior wall to be tested is calculated based on the known propagation speed of the ultrasonic wave and the two-way travel time.
2. The ultrasonic-based method of claim 1, wherein, Baseline correction of the original ultrasound echo signal includes: In free space conditions without a target to be measured, a reference background signal is acquired using the ultrasonic detection device. The baseline offset is obtained by calculating the average amplitude of the reference background signal over a preset time period. The baseline offset is subtracted from the amplitude of each sampling point of the original ultrasonic echo signal to complete the baseline correction, thereby obtaining the preprocessed ultrasonic echo signal.
3. The ultrasonic-based method of claim 2, wherein, The identification of multiple local peak and trough locations from the signal envelope includes: The first-order difference of the signal envelope is calculated to obtain the slope change sequence of the signal envelope; In the slope change sequence, find the zero-crossing point where the slope sign changes from positive to negative, and mark the position of the zero-crossing point in the signal envelope as a local peak position; In the slope change sequence, find the zero-crossing point where the slope sign changes from negative to positive, and mark the position of the zero-crossing point in the signal envelope as a local trough position; By traversing the entire slope change sequence, all local peak positions and local trough positions in the signal envelope are identified.
4. The ultrasonic-based method of claim 3, wherein, Based on the locations of the local peaks and troughs, multiple candidate echo intervals are divided within the effective signal segment, including: According to the time sequence, all the identified local peak and trough positions are sorted to obtain an alternating sequence of peak and trough positions. In the position sequence, starting from any local peak position, it extends to the next adjacent local trough position, forming a half-cycle interval from the peak to the trough. The half-cycle interval is extended forward and backward by a preset number of sample points to form an initial interval with the local peak position as the core. Merge all initial intervals that overlap on the time axis, and define each merged or unmerged initial interval as a candidate echo interval.
5. The ultrasonic-based method of claim 4, wherein, Calculating the fused feature value of the time-domain and frequency-domain features for each candidate echo interval includes: Perform a short-time Fourier transform on the signal within the candidate echo interval to obtain its spectral distribution; Calculate the time-domain energy of the candidate echo interval signal, and calculate the frequency-domain energy of its spectrum within the preset waterproof layer characteristic frequency band; The time-domain energy and the frequency-domain energy are respectively normalized. The normalized time-domain energy and frequency-domain energy are weighted and summed, and the resulting sum is the fused characteristic value of the candidate echo interval.
6. The ultrasonic-based method of claim 5, wherein, Based on the fusion feature values, the target echo intervals representing the main interface between the waterproof layer and the substrate are selected from multiple candidate echo intervals, including: A fusion feature threshold is set, which is obtained based on the echo interval characteristics of a known standard waterproof layer sample. Compare the fusion feature value of each candidate echo interval with the fusion feature threshold; The candidate echo interval with the fusion feature value greater than the fusion feature threshold and the largest fusion feature value among all candidate echo intervals that meet this condition is selected and determined as the target echo interval.
7. The ultrasonic-based method of claim 6, wherein, Calculating the time interval between the center position of the target echo interval in the effective signal segment and the initial peak position includes: Determine the start and end time points of the target echo interval within the valid signal segment; Calculate the average of the start time point and the end time point to obtain the center time point of the target echo interval; Calculate the time point corresponding to the initial peak position; Calculate the difference between the center time point and the time point corresponding to the initial wave peak position. The difference is the two-way travel time of the ultrasonic wave propagating in the waterproof layer. 8.The ultrasonic-based method for detecting the thickness of the waterproof layer of an external wall according to claim 7, wherein, Based on the known propagation speed of ultrasound and the two-way travel time, the thickness of the waterproof layer on the exterior wall to be measured is calculated as follows: Obtain the average propagation speed of ultrasonic waves in the material constituting the waterproof layer of the exterior wall to be tested; Multiplying the two-way travel time by the average propagation speed yields the total path length of the ultrasonic wave propagating back and forth in the waterproof layer material. Divide the total path length by two to get the thickness of the waterproof layer on the exterior wall to be measured.
9. The ultrasonic-based method of claim 8, wherein, The raw ultrasonic echo signals collected by the ultrasonic testing equipment on the surface of the waterproof layer of the exterior wall to be tested include: The transducer probe of the ultrasonic testing equipment is attached to the clean surface of the waterproof layer on the exterior wall to be tested. The transducer probe is controlled by an ultrasonic testing device to emit ultrasonic pulse signals with a specific center frequency and bandwidth. The transducer probe receives ultrasonic echo signals reflected from inside the waterproof layer of the exterior wall under test. The received ultrasonic echo signal is converted from analog to digital to generate a digital original ultrasonic echo signal.
10. The ultrasonic-based method of claim 9, wherein, The step of attaching the transducer probe of the ultrasonic testing equipment to the clean surface of the exterior wall waterproofing layer to be tested includes: Apply a sufficient amount of acoustic coupling agent to the acoustic wave emitting end of the transducer probe to ensure that the acoustic coupling agent evenly covers the entire acoustic wave emitting end face; Place the transducer probe stably on the clean surface of the waterproof layer of the exterior wall to be tested, and apply a moderate pressure perpendicular to the surface so that the acoustic coupling agent forms a uniform and bubble-free film between the probe end face and the surface of the waterproof layer. The transducer probe is kept in a stable position and orientation at the detection point, and no relative movement occurs between it and the surface of the waterproof layer of the exterior wall to be tested during the acquisition of the original ultrasonic echo signal.