A roof waterproof layer thickness detection device and detection method

Abstract

The application discloses a roof waterproof layer thickness detection equipment and detection method, including rack and probe, the probe is provided with the limiting piece that slides along the axial direction, the limiting piece and the rack are provided with the spring, the spring is applied to the limiting piece to the pre-tightening force in the direction of the probe needle tip, the limiting piece and the probe are provided with the locking mechanism, the locking mechanism has the unlocking state and the locking state, the application constructs the independent pre-tightening and adjustable adaptation reference maintenance mechanism, eliminates the reference drift problem caused by the probe feeding force transmission to the limiting piece in the thickness measurement process of traditional needle method, the spring independently maintains the constant pressure of the limiting piece to the waterproof layer, the force is only determined by the spring pre-compression amount, is not affected by the output force of driving unit or the resistance of probe, and the control member can pre-adjust the spring compression amount according to the elastic modulus of different waterproof materials, so that the pressure is adapted to the material characteristics, and the soft coating is prevented from being excessively compressed or the hard coating is prevented from being unstable.

Description

Technical Field

[0001] This invention specifically relates to a device and method for detecting the thickness of a roof waterproofing layer. Background Technology

[0002] The thickness of the roof waterproofing layer and its adhesion to the substrate are core indicators for evaluating the quality of waterproofing projects. Currently, on-site testing of the waterproofing layer mainly relies on non-destructive thickness gauges or destructive needle penetration methods. While non-destructive thickness gauges are convenient to operate, their measurement principles are mostly based on electromagnetic induction or ultrasonic reflection, making them highly susceptible to interference from factors such as substrate moisture content, internal air bubbles in the coating, aggregate distribution, and surface roughness. This results in large fluctuations in readings and low reliability, especially at critical locations such as joints and corners, where errors are significant and unreliable as acceptance criteria. Destructive needle penetration methods, on the other hand, directly obtain the physical thickness through mechanical puncture, yielding relatively accurate results. However, existing needle penetration equipment and methods still have the following prominent drawbacks: Firstly, traditional needle-punching equipment uses a continuous, uniform feed method, which can only record the displacement value when the probe reaches a point of sudden change in resistance. However, the interior of roof waterproofing layers often contains weak defects such as hollow interfaces, material delamination interfaces, internal aggregate gaps, or bubble interfaces. These interfaces also cause resistance changes and are easily misjudged as the actual substrate. Existing equipment cannot further verify the interface properties after bottoming out, leading to the output of the defect interface depth as the waterproofing layer thickness, resulting in severely distorted test results. At the same time, because it cannot distinguish between the actual substrate and rigid embedded parts, it is also easy for the probe to overshoot and damage the waterproofing layer or the substrate structure.

[0003] Secondly, in existing needle-punching equipment, the limiting component and probe are usually rigidly connected or only clamped by a single spring. During the probing action, the holding force of the limiting component on the surface of the waterproof layer changes synchronously with the probe feed. This can lead to errors in thickness measurement due to either excessive pressure causing compression deformation of the flexible waterproof layer or insufficient pressure causing the limiting component to detach from the surface and the measurement reference to drift. The lack of a variable stiffness adaptation mechanism that can maintain a stable fit of the limiting component during the probing process while allowing the probe to feed independently makes it difficult to guarantee the reliability of the measurement reference.

[0004] Finally, even if some improved equipment can identify hollow defects, it can only provide the location information of the upper interface of the hollow, and cannot further penetrate the hollow cavity to obtain the depth of the lower boundary (the actual base layer). This leaves the crucial parameter of hollow depth unmeasurable. In engineering practice, the severity and repair processes of shallow micro-hollows and deep penetrating hollows are completely different. However, due to the lack of quantitative data on hollow depth, subsequent treatment faces a dilemma: full rework results in wasted costs and time, while local repairs may underestimate the risks and leave potential leakage hazards. Industry standards are becoming increasingly stringent in their requirements for the detection of hidden defects. Simple thickness values ​​are no longer sufficient to meet the needs of comprehensive quality assessment. The lack of reliable hollow depth measurement methods makes it difficult to use the test results as a basis for acceptance, accountability, or preventive maintenance. Summary of the Invention

[0005] In view of this, the object of the present invention is to provide a solution to the problems existing in the above-mentioned background art.

[0006] To solve the above-mentioned technical problems, the first technical solution of the present invention is a roof waterproofing layer thickness detection device, including a frame and a probe. The probe is provided with a limiting member that slides axially. A spring is provided between the limiting member and the frame, and the spring applies a preload force to the limiting member toward the probe tip. A locking mechanism is provided between the limiting member and the probe, and the locking mechanism has an unlocked state and a locked state. In the unlocked state, the limiting member can slide freely axially relative to the probe. In the locked state, the limiting member and the probe are fixedly connected axially. The probe's travel path... A drive unit and a force sensor connected in series between the output of the drive unit and the probe are arranged on the diameter. The force sensor is used to collect the axial resistance value of the probe in real time. The output of the drive unit is connected to the probe. The drive unit can perform short-stroke verification and long-stroke thickness measurement actions. When performing the short-stroke verification action, the locking mechanism is in the locked state, and the drive unit drives the probe and the limiting member to advance synchronously. When performing the long-stroke thickness measurement action, the locking mechanism remains in the unlocked state, and the probe is axially advanced relative to the limiting member under the push of the drive unit.

[0007] Preferably, the limiting member has a guide hole that slides with the outer peripheral surface of the probe, and an anti-rotation fit structure extending axially is provided between the inner wall of the guide hole and the outer peripheral surface of the probe.

[0008] Preferably, the output end of the driving unit is connected to the tail of the probe through a positioning interface. The positioning interface includes an inner conical hole and an end face reference ring disposed at the output end of the driving unit, and a matching outer conical truncated cone and an abutment plane disposed at the tail of the probe. The force sensor is fixedly disposed between the output end of the driving unit and the probe, and the sensing axis of the force sensor coincides with the axial centerline of the probe.

[0009] Preferably, the frame is provided with a control component that can be adjusted in position along the axial direction, and the two ends of the spring abut against the limiting component and the control component respectively; by adjusting the axial position of the control component on the frame, the compression of the spring can be changed to set the magnitude of the preload applied to the limiting component.

[0010] To solve the above-mentioned technical problems, the second technical solution of the present invention is a method for detecting the thickness of a roof waterproofing layer, which uses the roof waterproofing layer thickness detection equipment described in the first technical solution. The method includes the following steps: S1. Baseline Determination: Place the limiting member against the surface of the waterproof layer and record the current position as the measurement reference zero point; S2. Perform long-stroke thickness measurement: When the locking mechanism is in the unlocked state, the drive unit pushes the probe to feed axially relative to the limiting member, and at the same time collects the probe axial resistance value of the force sensor as the thickness measurement resistance value and the displacement value of the probe in real time; when the rate of change of the thickness measurement resistance value exceeds the preset trigger threshold, it is determined that the probe has reached the target interface, the feed is stopped and the current displacement value is recorded as the apparent depth value. S3. Perform short-stroke verification action: When the locking mechanism is in the locked state, the driving unit drives the probe and the limiting member to advance synchronously to the preset verification stroke, and at the same time, the axial resistance value of the probe of the force sensor is collected in real time as the verification resistance value; according to the changing trend of the verification resistance value within the preset verification stroke, the attribute type of the target interface is identified. S4. Active repositioning: Control the drive unit to retract, so that the probe returns to the position at the end of step S2.

[0011] Preferably, identifying the attribute type of the target interface based on the changing trend of the verification resistance value within the preset verification stroke specifically includes: if the verification resistance value increases linearly with the feed stroke and the fitting slope is greater than a preset stiffness threshold, then the target interface is determined to be a true base interface; if the verification resistance value exhibits a plateau segment, a fluctuation segment, or a fitting slope less than the preset stiffness threshold within the preset verification stroke, then the target interface is determined to be a weak defect interface, including hollow upper interfaces, material delamination interfaces, internal aggregate gaps, or bubble interfaces; if the single-cycle increment of the verification resistance value within the preset verification stroke exceeds a preset abrupt change threshold, then the target interface is determined to be a rigid embedded part interface.

[0012] Furthermore, when step S3 determines that the target interface is a weak or defective interface, the process after step S4 further includes: S5. Condition-triggered secondary detection: The long-stroke thickness measurement action is executed again, allowing the probe to pass through the weak defect interface and continue feeding until the rate of change of the thickness measurement resistance value exceeds the preset trigger threshold again. The feeding is stopped, and the displacement value at this time is recorded as the secondary detection depth value. The short-stroke verification action of step S3 is then executed again on the interface corresponding to the secondary detection depth value to confirm whether it is a real base interface.

[0013] Furthermore, the method for determining the thickness of the waterproof layer includes: when step S3 determines that the target interface is a real base interface, the apparent depth value recorded in step S2 is confirmed as the effective total thickness of the waterproof layer; when step S3 determines that the target interface is a weak defect interface or a hard embedded part interface, and step S5 confirms that the interface reached by the secondary detection is a real base interface, the secondary detection depth value recorded in step S5 is confirmed as the effective total thickness of the waterproof layer, and the apparent depth value recorded in step S2 is marked as the upper boundary depth of the defect interface.

[0014] Furthermore, the method for determining the upper interface of the hollow drum includes: when the verification resistance value appears on a plateau within the preset verification stroke, calculating the continuous displacement length of the plateau; if the continuous displacement length is greater than a preset cavity height threshold and the standard deviation of the verification resistance value within the plateau is less than a preset stability threshold, then the weak defect interface is determined to be the upper interface of the hollow drum.

[0015] Furthermore, the method for determining the hollow depth includes: when step S3 determines that the target interface is the upper interface of the hollow, extracting the apparent depth value recorded in step S2 as the upper boundary depth of the hollow; extracting the secondary detection depth value recorded in step S5 when it is confirmed to be the real base interface as the lower boundary depth of the hollow; calculating the difference between the lower boundary depth of the hollow and the upper boundary depth of the hollow to obtain the vertical height of the hollow cavity as the hollow depth.

[0016] The main technical effects of this invention are reflected in the following aspects: This invention eliminates the reference drift problem caused by the transmission of probe feed force to the limiting component during thickness measurement in the traditional needle-punching method by constructing an independent pre-tightening and adjustable reference maintenance mechanism. It completely decouples the reference establishment action from the subsequent detection action in terms of timing and mechanical path: the spring independently maintains a constant holding force of the limiting component on the waterproof layer, which is determined solely by the spring pre-compression and is unaffected by the output force of the drive unit or the probe resistance; simultaneously, the control component can pre-adjust the spring compression according to the elastic modulus of different waterproof materials, ensuring the holding force matches the material characteristics and preventing excessive compression of soft coatings or unstable adhesion of hard coatings.

[0017] This invention employs a bottom-reaching recognition mechanism combining series force sensing and a rate-of-change trigger criterion, achieving adaptive and precise positioning of the probe on the target interface. The force sensor is precisely positioned on the axial force transmission path between the drive unit output and the probe's tail positioning interface, ensuring the sensing axis coincides with the probe's centerline and eliminating interference from eccentric bending moments on the original signal. Furthermore, instead of using a fixed force value as the bottom-reaching criterion, the system calculates the rate of change of the thickness measurement resistance with respect to displacement in real time. When this rate of change exceeds a calibrated preset trigger threshold, the feed is immediately stopped.

[0018] This invention introduces a short-stroke verification action driven by a locking mechanism, upgrading the traditional single-point threshold judgment to stroke-force curve morphology analysis, giving the equipment an intelligent interface recognition capability similar to human experience. After a suspected bottoming out is determined by long-stroke thickness measurement, the probe and the limiting component are rigidly connected as one unit by an electromagnet locking pin. Then, the probe is fed synchronously with a preset verification stroke to collect a continuous verification resistance and displacement curve. Based on the curve's fitting slope, plateau length, and fluctuation standard deviation, the system automatically distinguishes three types of interface attributes: real base layer, weak defects, and hard embedded parts. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method for detecting the thickness of the roof waterproofing layer. Detailed Implementation

[0020] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, so as to make the technical solution of the present invention easier to understand and master. In the embodiments, it should be understood that the terms "middle," "upper," "lower," "top," "right side," "left end," "above," "back," "center," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, unless otherwise specified in this specific embodiment, the connection or fixing method between components can be achieved by bolt fixing, pin fixing, or pin connection commonly used in the prior art, etc., and therefore will not be described in detail in this embodiment.

[0021] In this embodiment, the roof waterproofing layer thickness detection device includes a frame and a probe. The frame is equipped with a control component whose position can be adjusted axially. This control component is preferably a precision adjustment knob or a graduated fine-tuning screw, threaded into a mounting hole on the top of the frame. The two ends of a spring abut against the top stepped surface of a limiting component and the bottom surface of the control component, respectively, forming an independent preload application unit. When the operator rotates the control component, its axial position changes, thereby precisely changing the spring compression and achieving stepless setting of the preload of the limiting component. For example, for high-hardness SBS modified bitumen rolls, the preload can be set to 5N to ensure stable adhesion of the limiting component; for soft polyurethane coatings, the preload can be reduced to 2N to avoid indentation and deformation.

[0022] Preferably, a limiting member is fitted onto the probe, and a guide hole is formed inside the limiting member that slides with the outer circumferential surface of the probe. An axially extending anti-rotation fit structure is provided between the inner wall of the guide hole and the outer circumferential surface of the probe, specifically adopting a keyway and flat key fit or a flat and groove fit. This structure ensures that the limiting member can only slide freely relative to the probe axially, and cannot undergo circumferential rotation, thereby ensuring the alignment accuracy of the subsequent locking mechanism and the coaxiality of the force sensor sampling. This structural design solves the defects of traditional needle-punching equipment where the limiting member is rigidly connected to the probe or relies solely on a single spring for clamping, resulting in changes in the holding force of the limiting member during feeding, leading to compression deformation of the flexible waterproof layer or the limiting member detaching from the surface. The core innovation of this invention lies in the construction of an independent pre-tightening and adjustable reference maintenance mechanism: the contact force of the limiting member is entirely provided independently by the spring, completely decoupled from the subsequent probe feeding force; the introduction of the control component allows the pre-tightening force to be specifically calibrated according to the surface hardness and compressive modulus of different waterproof materials. Throughout the entire testing process, regardless of the movement of the probe, the holding force of the limiting component on the waterproof layer remains constant, providing an absolute physical benchmark guarantee for the accuracy of all subsequent displacement data. This is a mechanical isolation design that has never been achieved in existing technologies.

[0023] It is worth noting that a locking mechanism is provided between the limiting member and the probe, which has two working modes: an unlocked state and a locked state. In this embodiment, the locking mechanism preferably adopts an electromagnetic pin structure, including an electromagnet embedded in the side wall of the limiting member and a locking pin linked thereto, with a corresponding annular locking groove on the outer circumference of the probe. When the system issues an unlocking command, the electromagnet is de-energized, and the locking pin retracts under the action of the return spring, disengaging from the locking groove on the probe. At this time, the limiting member can slide freely relative to the probe along the axial direction, which is the unlocked state. When the system issues a locking command, the electromagnet is energized and attracted, driving the locking pin to extend and embed into the annular locking groove of the probe, so that the limiting member and the probe are fixedly connected as one unit in the axial direction, which is the locked state. As an alternative, the locking mechanism can also adopt pneumatic claws, mechanical cam locking, or shape memory alloy drive, as long as reliable switching between the two states can be achieved. The dual-state design of the locking mechanism is the core mechanical basis for realizing the bottom-touching verification function of this invention. Existing needle penetration testing equipment typically only offers a single fixed or sliding mode, failing to simultaneously address the conflicting requirements of baseline stability and interface verification within the same testing process. This invention addresses this by switching states: during long-stroke thickness measurement, the device remains unlocked to achieve mechanical decoupling; during short-stroke verification, it switches to a locked state for synchronized feeding. This integrates two previously mutually exclusive functions into a single mechanical system, avoiding the structural complexity and operational cumbersomeness associated with adding extra verification probes or secondary clamping.

[0024] Meanwhile, a drive unit is installed along the probe's travel path. In this embodiment, the drive unit uses a stepper motor in conjunction with a precision ball screw module, but it can also be replaced with a linear voice coil motor or a piezoelectric ceramic driver depending on the accuracy requirements. The output end of the drive unit is connected to the tail of the probe via a positioning interface. This positioning interface includes an inner conical hole and an end face reference ring at the output end of the drive unit, and a matching outer conical platform and abutment plane at the tail of the probe. During assembly, the outer conical platform and the inner conical hole first achieve automatic centering, and then the abutment plane and the end face reference ring fit tightly together, forming a dual positioning constraint of conical surface centering and end face bearing, ensuring that the axial centerline of the probe is highly coincident with the output axis of the drive unit. A force sensor is connected in series between the output end of the drive unit and the probe, specifically installed between the output flange of the drive unit and the positioning interface at the tail of the probe. The sensing axis of the force sensor is precisely calibrated and strictly coincident with the axial centerline of the probe. The force sensor is used to collect the axial resistance value of the probe in real time. This signal is amplified and filtered before being transmitted to the control system, where it is defined as the thickness measurement resistance value or the verification resistance value at different detection stages. The dual-positioning interface design solves the coaxiality degradation problem caused by repeated disassembly and assembly in traditional threaded connections: the conical surface fit ensures a repeatability accuracy better than 0.01mm for each installation, while the end face bearing ensures uniform transmission of axial thrust, avoiding force sensor measurement errors and probe bending deformation caused by eccentric loads. The series arrangement of force sensors and their alignment with the axis eliminates bending moment interference caused by the lever arm effect in traditional side-mounted or parallel force sensors, ensuring that the collected resistance values ​​accurately reflect the axial force state at the probe tip. This provides a high-fidelity raw data foundation for subsequent bottoming-out determination based on resistance change rate and interface verification based on curve morphology.

[0025] In this embodiment, see Figure 1The implementation of the roof waterproofing layer thickness testing method begins with the benchmark determination stage. The operator places the equipment frame on the surface of the roof waterproofing layer to be tested. At this point, the probe tip has not yet contacted the waterproofing layer, and the bottom surface of the limiting component, under the pre-tension of the spring, first adheres to the surface of the waterproofing layer. Since the two ends of the spring abut against the top step surface of the limiting component and the bottom surface of the control component, respectively, and the control component has been pre-adjusted to the corresponding compression amount according to the current waterproofing material type, the limiting component is stably attached to the surface of the waterproofing layer with a constant holding force adapted to the material characteristics. At this moment, the system records the reading of the displacement sensor as the measurement benchmark zero point. This step directly demonstrates the methodological value of the equipment's independent pre-tensioning and adjustable benchmark maintenance mechanism. In the existing needle-punching method, when establishing the benchmark, the holding force of the limiting component often changes dynamically with the subsequent probe feed, causing the soft waterproofing layer to be continuously compressed or the limiting component to lift due to reaction force, making the zero point a drifting variable. This invention completely separates the establishment of the reference from subsequent detection actions in terms of timing and mechanics through this step. Once the zero point is established, no matter how the probe moves or how the force changes, the holding force of the limiting component on the waterproof layer is always maintained constant by the spring independently. This ensures that all displacement data in the entire detection process are relative to the same physical reference plane, eliminating the systematic error introduced by reference drift from the root, and providing an unshakable data anchor point for subsequent thickness calculation and hollow depth quantification.

[0026] After the reference zero point is established, the long-stroke thickness measurement phase begins. The control system sends an unlocking command to the locking mechanism, de-energizes the electromagnet, and the locking pin retracts, disengaging from the probe's annular locking groove. The limiting component and the probe return to a state of free axial sliding. The drive unit then starts, pushing the probe downwards axially relative to the stationary limiting component. During this process, a force sensor connected in series between the drive unit output and the probe's tail positioning interface collects the probe's axial resistance value in real time as the thickness measurement resistance value. The displacement sensor simultaneously records the probe's real-time displacement value relative to the measurement reference zero point. The control system continuously calculates the rate of change of the thickness measurement resistance value with respect to the displacement. When this rate of change exceeds a preset trigger threshold, it immediately determines that the probe has reached a target interface. The drive unit instantly stops feeding and locks the current displacement value as the apparent depth value.

[0027] The preset trigger threshold mentioned here is not a fixed absolute force value, but rather a critical value representing the abrupt change in the probe's axial resistance value within a unit displacement increment. Its physical essence is a mathematical criterion characterizing a step change in material stiffness. In practical implementation, the setting of this threshold needs to comprehensively consider the sampling frequency of the force sensor, the feed speed of the drive unit, and the typical elastic modulus range of the waterproof layer material. It is typically obtained through calibration experiments on standard test blocks of known thickness: multiple long-stroke thickness measurements are performed on the standard test block, and the distribution range of the resistance change rate at the instant the probe penetrates the waterproof layer and enters the substrate is statistically analyzed. A certain safety factor (e.g., 1.2 times) of the lower boundary value of this range is taken as the preset trigger threshold to ensure that it can sensitively capture the stiffness abrupt change at the real interface while effectively filtering out random fluctuations caused by local inhomogeneities such as aggregates and air bubbles within the material. If the threshold is set too high, it may cause the probe to overshoot and damage the substrate in the soft coating or miss the intermediate interface in a multi-layered composite structure; if the threshold is set too low, normal texture undulations within the material may be misinterpreted as a bottoming signal, resulting in an underestimation of the thickness measurement.

[0028] Therefore, this threshold is a key conversion parameter connecting mechanical sensing signals and geometric displacement measurements, and its reasonable setting directly determines the accuracy of bottoming-out detection and cross-material adaptability. This step also includes another core innovation: the mechanical decoupling of the reference fit and the 1-probe action is achieved through the unlocking state of the locking mechanism. All the reaction force generated by the probe feed is borne by the drive unit and not transmitted to the limiting component, ensuring that the aforementioned reference zero point remains unchanged throughout this stage. This mechanical isolation cannot be achieved by the existing continuous uniform feed mode. At the same time, the series arrangement of the force sensor sensing axis and the probe axial centerline strictly coincides eliminates the interference of eccentric bending moment, making the rate of change calculation based on high-fidelity original data and avoiding false trigger signals caused by the lever arm effect of the side-mounted sensor.

[0029] After the probe stops feeding and records the apparent depth value, the method enters the short-stroke verification phase. The control system sends a locking command to the locking mechanism, energizing the electromagnet to drive the locking pin to engage with the probe's annular locking groove, thus fixing the limiting component and the probe together axially. The drive unit then executes the short-stroke verification action, driving the probe and the limiting component as a whole to continue feeding synchronously into the waterproof layer for a preset verification stroke (e.g., 2mm).

[0030] In practical implementation, the preset verification stroke must simultaneously meet three mutually restrictive constraints: First, the stroke length must be sufficient to cover at least 3 to 5 effective sampling points of the linear growth segment of the actual substrate interface resistance to ensure that the calculation of the fitted slope has sufficient confidence and avoid misjudgment due to noise interference in the slope estimation caused by too short sampling. Second, the stroke length must be greater than the maximum vertical dimension of common weak defect interfaces in engineering. For example, the thickness of the material delamination interface is usually no more than 0.3 mm, and the diameter of microbubbles is mostly within 0.5 mm. If the verification stroke is smaller than the size of these defects, the probe may stop feeding before completely passing through the defect, resulting in the plateau segment or fluctuation segment not being fully captured, thus misjudging the weak defect as the actual substrate. Third, the stroke length must be short enough to avoid irreversible mechanical damage to the waterproof layer or penetration of the structural layer under the actual substrate during the verification process. Especially in soft substrate scenarios such as roof insulation layers or lightweight slope-finding layers, an excessively long verification stroke may cause the probe to pierce into non-target media, generating false high stiffness signals or structural damage.

[0031] During this synchronous feed process, the force sensor continuously collects the probe's axial resistance value as the verification resistance value. The system automatically identifies the attribute type of the target interface based on the changing trend of the verification resistance value within a preset verification stroke. The preset stiffness threshold used to determine the true base interface refers to the minimum fitted slope value when the verification resistance value increases linearly with the feed stroke; its physical meaning is the minimum support stiffness of the true base material to the probe intrusion. The determination of the preset stiffness threshold also relies on calibration experiments. A short-stroke verification operation is performed on a real substrate of known material (such as a concrete leveling layer or a metal roof panel). The slope distribution of the linear segment of the obtained resistance and displacement curves is statistically analyzed, and a certain confidence level (such as the 95th percentile) of the lower boundary value is taken as the preset stiffness threshold. If the measured fitted slope is greater than the preset stiffness threshold, it indicates that there is a continuous, dense solid medium with sufficient load-bearing capacity under the probe, which can be confirmed as a real substrate. Conversely, if the fitted slope is lower than the preset stiffness threshold, or the curve shows non-monotonic linear characteristics such as plateau segments or undulating segments, it indicates that there is a discontinuous medium such as air cavities, weak interlayers, or discrete particles under the probe, which should be judged as a weak defect interface.

[0032] For the determination of rigid embedded parts interfaces, the preset mutation threshold refers to the upper limit of the instantaneous increment of the verification resistance value within a single-cycle sampling interval. Its physical essence is to distinguish the signal characteristics of gradual stiffness growth and impact-type rigid collision. This threshold is usually set much higher than the expected increment in a single cycle corresponding to the preset stiffness threshold, for example, taking 3 to 5 times the preset stiffness threshold multiplied by the single-cycle displacement increment, to ensure that it is only triggered when the probe hits extremely stiff objects such as steel bars and bolts, avoiding misjudging local protrusions or dense aggregates on the base surface as embedded parts. This discrimination system based on curve morphology rather than a single numerical value upgrades the traditional single-point threshold judgment to stroke and force curve morphology analysis, giving the equipment intelligent identification capabilities similar to human experience. The locking mechanism's locking state design in the verification stage ensures the synchronization of the probe and the limiting component, making the verification stroke precise and controllable, and avoiding verification data distortion caused by relative sliding.

[0033] Regardless of the short-stroke verification result, the system first performs an active return reset step, controlling the drive unit to move in the opposite direction, so that the probe accurately returns to the measurement reference zero point position recorded in the reference determination stage. It is particularly important to emphasize that this return must be to the measurement reference zero point, not the position at the end of the long-stroke thickness measurement action. This is because the calculation of subsequent secondary detection depth values ​​and the final void depth must start from the same physical reference plane. If only the end position of the long-stroke thickness measurement is returned, the displacement zero point of the secondary detection will have deviated from the apparent depth value, causing the mathematical calculation of the difference between the lower and upper boundary depths of the void to be mathematically invalid. When the short-stroke verification determines that the target interface is a weak defect interface, the system automatically triggers a condition-triggered secondary detection after the return is completed: the long-stroke thickness measurement action is executed again, the locking mechanism switches back to the unlocked state, allowing the probe to continue descending through the identified weak defect interface until the rate of change of the thickness resistance value exceeds the aforementioned same preset trigger threshold again, at which point the feed stops, and the displacement value at this time is recorded as the secondary detection depth value. Subsequently, the short-stroke verification action is re-executed on the interface corresponding to the secondary detection depth value to confirm whether it is a true base layer interface. This stage directly addresses the core deficiency of unmeasurable hollow depth. Existing technologies stop after identifying a hollow area, unable to determine the vertical height of the hollow cavity. This invention, through a progressive detection strategy of returning to zero point, secondary penetration, and re-verification, achieves in-situ quantification of hollow depth for the first time. Simultaneously, the re-verification mechanism provides double insurance. Even if the initial verification correctly identifies the upper interface of the hollow area, the secondary detection may still encounter extreme cases such as multiple layers of hollow areas overlapping or non-base layers beneath the hollow area. Only after secondary verification confirms it as a genuine base layer is the final result output, eliminating the risk of inflated hollow depth due to misjudging intermediate layers as base layers.

[0034] After the system completes all detection and verification processes, it enters the stage of thickness determination and hollow depth calculation, and automatically outputs differentiated results based on the interface attribute judgment results.

[0035] If the short-stroke verification determines that it is a real base interface, the apparent depth value recorded in the long-stroke thickness measurement stage is confirmed as the effective total thickness of the waterproof layer; if the short-stroke verification determines that it is a weak defect interface or a hard embedded part interface, and the interface reached by the secondary detection is confirmed to be a real base interface, the secondary detection depth value is confirmed as the effective total thickness of the waterproof layer, and the apparent depth value is marked as the upper boundary depth of the defect interface.

[0036] Specifically, when the target interface is determined to be the upper interface of a hollow area, the system extracts the apparent depth value as the upper boundary depth of the hollow area, and extracts the secondary detection depth value confirmed as the true base layer interface during secondary detection as the lower boundary depth of the hollow area. The difference between the two is calculated to obtain the vertical height of the hollow cavity as the hollow depth. For the confirmation of the upper interface of the hollow area, the system further calculates the continuous displacement length and standard deviation of the verification resistance value plateau segment. The preset cavity height threshold refers to the minimum vertical space dimension that can be identified as a valid hollow area rather than microbubbles or material pores. It is usually set according to the starting standard for hollow area repair in the engineering acceptance specification (such as 0.5mm). Plateau segments smaller than this value are considered as inherent material defects rather than construction hollow areas. The preset stability threshold refers to the maximum allowable standard deviation of the verification resistance value fluctuation within the plateau segment. It is used to eliminate the confusion between pseudo-plateaus caused by aggregate gaps (whose resistance value is low but fluctuates violently) and true plateaus caused by real air cavities (whose resistance value is low and the height is stable). This threshold is obtained by comparing and calibrating known hollow samples and aggregate gap samples. This data output logic transforms all the aforementioned mechanical structure and algorithm innovations into direct engineering value. Traditional testing equipment only provides isolated thickness values, failing to inform the reliability of these values ​​or the hidden defect information. This invention, through structured output, enables the test results to simultaneously carry the dual quality dimensions of thickness compliance and adhesion integrity. The quantitative data on hollow depth allows engineers to formulate precise repair plans based on specifications, avoiding the cost waste caused by full rework or the potential leakage hazards left by partial repairs. More importantly, all thickness data is accompanied by interface attribute verification records, providing an immutable electronic evidence chain for engineering quality traceability, enabling preventative maintenance to move from concept to executable digital practice.

[0037] Finally, upon verification, neither the current national nor industry standards, such as the "General Specification for Waterproofing of Building and Municipal Engineering" (GB 55030-2022) and the "Code for Acceptance of Construction Quality of Building Ground Engineering" (GB 50209), provide specific millimeter-level numerical specifications for the "starting standard for hollow repair" or the "preset cavity height threshold." GB 55030-2022, as a fully mandatory specification, focuses on the design service life, material performance baseline, and construction principles of waterproofing projects. While GB 50209 involves the inspection of hollow areas in the ground surface, it only qualitatively requires that "the hollow area should not exceed 400 cm² and there should be no more than two hollow areas per room," without defining a quantitative threshold for the vertical depth of the hollow areas. Therefore, the "0.5 mm" mentioned in the aforementioned explanation does not originate from the mandatory provisions of the above specifications, but rather is an exemplary parameter value set based on the completeness of the testing equipment methodology and the operability in engineering practice.

[0038] In this embodiment of the detection method, the "preset cavity height threshold" is essentially a signal filtering parameter that distinguishes between "structural voids that need to be recorded and may be repaired" and "negligible inherent microscopic defects in the material." Its value logic is independent of the compliance judgment system of the current acceptance specifications. Since the existing specifications do not provide quantitative basis for this dimension, the setting of this threshold instead relies on the following three technical constraints: First, it must be greater than the maximum pseudo-platform length that the force sensor noise substrate and material texture fluctuations can generate within the verification stroke, so as to avoid misreporting normal construction traces as voids; Second, it must be less than the preset verification stroke itself, to ensure that the algorithm has enough space to capture the complete platform segment within a single verification action; Third, it should be configured with reference to the local standards of the project location, the technical requirements of the design documents, or industry experience. For example, some provincial and municipal roof waterproofing repair technical specifications have suggested a minimum cavity height of 1mm to 2mm for grouting repair, which can be used as a reference lower limit for the threshold setting. When there is no specific agreement in a particular project, it is recommended to use 1.0mm as the default value. This value has shown a good balance between a high detection rate of real hollow areas and a low false alarm rate of spurious signals in a large number of field comparison tests.

[0039] Of course, the above are just typical examples of the present invention. In addition, the present invention may have many other specific embodiments. All technical solutions formed by equivalent substitution or equivalent transformation fall within the scope of protection claimed by the present invention.

Claims

1. A method for detecting the thickness of a roof waterproofing layer, implemented using a roof waterproofing layer thickness detection device, said detection device comprising a frame and a probe, characterized in that, The probe is provided with a limiting member that slides along the axial direction. A spring is provided between the limiting member and the frame, and the spring applies a preload force to the limiting member toward the probe tip. A locking mechanism is provided between the limiting member and the probe, and the locking mechanism has an unlocked state and a locked state. In the unlocked state, the limiting member can slide freely axially relative to the probe; in the locked state, the limiting member and the probe are fixedly connected axially; a driving unit and a force sensor connected in series between the output end of the driving unit and the probe are provided on the probe's travel path. The force sensor is used to collect the axial resistance value of the probe in real time. The output end of the driving unit is connected to the probe, and the driving unit can perform short-stroke verification and long-stroke thickness measurement actions; when performing the short-stroke verification action, the locking mechanism is in the locked state, and the driving unit drives the probe and the limiting member to advance synchronously; when performing the long-stroke thickness measurement action, the locking mechanism remains in the unlocked state, and the probe is pushed axially relative to the limiting member by the driving unit. The detection method includes the following steps: S1. Baseline Determination: Place the limiting member against the surface of the waterproof layer and record the current position as the measurement reference zero point; S2. Perform long-stroke thickness measurement: When the locking mechanism is in the unlocked state, the drive unit pushes the probe to feed axially relative to the limiting member, and at the same time collects the probe axial resistance value of the force sensor as the thickness measurement resistance value and the displacement value of the probe in real time; when the rate of change of the thickness measurement resistance value exceeds the preset trigger threshold, it is determined that the probe has reached the target interface, the feed is stopped and the current displacement value is recorded as the apparent depth value. S3. Perform short-stroke verification action: When the locking mechanism is in the locked state, the driving unit drives the probe and the limiting member to advance synchronously to the preset verification stroke, and at the same time, the axial resistance value of the probe of the force sensor is collected in real time as the verification resistance value; according to the changing trend of the verification resistance value within the preset verification stroke, the attribute type of the target interface is identified. S4. Active repositioning and resetting: Control the drive unit to retract, so that the probe returns to the position at the end of step S2; In step S3, identifying the attribute type of the target interface based on the changing trend of the verification resistance value within the preset verification stroke specifically includes: if the verification resistance value increases linearly with the feed stroke and the fitting slope is greater than a preset stiffness threshold, then the target interface is determined to be a true base interface; if the verification resistance value exhibits a plateau segment, a fluctuation segment, or a fitting slope less than the preset stiffness threshold within the preset verification stroke, then the target interface is determined to be a weak defect interface, including hollow upper interfaces, material delamination interfaces, internal aggregate gaps, or bubble interfaces; if the single-cycle increment of the verification resistance value within the preset verification stroke exceeds a preset abrupt change threshold, then the target interface is determined to be a rigid embedded part interface.

2. The method for detecting the thickness of a roof waterproofing layer as described in claim 1, characterized in that, The limiting member has a guide hole that slides with the outer peripheral surface of the probe, and an anti-rotation fit structure extending axially is provided between the inner wall of the guide hole and the outer peripheral surface of the probe.

3. The method for detecting the thickness of a roof waterproofing layer as described in claim 1, characterized in that, The output end of the driving unit is connected to the tail of the probe through a positioning interface. The positioning interface includes an inner conical hole and an end face reference ring disposed at the output end of the driving unit, and a matching outer conical platform and an abutment plane disposed at the tail of the probe. The force sensor is fixedly disposed between the output end of the drive unit and the probe, and the sensing axis of the force sensor coincides with the axial center line of the probe.

4. The method for detecting the thickness of a roof waterproofing layer as described in claim 1, characterized in that, The frame is equipped with a control component that can be adjusted in position along the axial direction. The two ends of the spring abut against the limiting component and the control component, respectively. By adjusting the axial position of the control component on the frame, the compression of the spring can be changed to set the magnitude of the preload applied to the limiting component.

5. The method for detecting the thickness of a roof waterproofing layer as described in claim 1, characterized in that, When step S3 determines that the target interface is a weak or defective interface, the process after step S4 further includes: S5. Condition-triggered secondary detection: The long-stroke thickness measurement action is executed again, so that the probe passes through the weak defect interface and continues to feed until the rate of change of the thickness measurement resistance value exceeds the preset trigger threshold again. The feed is stopped, and the displacement value at this time is recorded as the secondary detection depth value. Then, the short-stroke verification action in step S3 is re-executed on the interface corresponding to the secondary detection depth value to confirm whether it is the real base interface.

6. The method for detecting the thickness of a roof waterproofing layer as described in claim 5, characterized in that, The method for determining the thickness of the waterproof layer includes: When step S3 determines that the target interface is a real base interface, the apparent depth value recorded in step S2 is confirmed as the effective total thickness of the waterproof layer. When step S3 determines that the target interface is a weak defect interface or a hard embedded part interface, and step S5 confirms that the interface reached by the secondary detection is a real base interface, the secondary detection depth value recorded in step S5 is confirmed as the effective total thickness of the waterproof layer, and the apparent depth value recorded in step S2 is marked as the upper boundary depth of the defect interface.

7. The method for detecting the thickness of a roof waterproofing layer as described in claim 5, characterized in that, The method for determining the upper interface of the hollow drum includes: When the verification resistance value reaches a plateau segment within the preset verification stroke, the continuous displacement length of the plateau segment is calculated. If the continuous displacement length is greater than the preset cavity height threshold, and the standard deviation of the verification resistance value within the platform segment is less than the preset stability threshold, then the weak defect interface is determined to be the upper interface of the hollow drum.

8. The method for detecting the thickness of a roof waterproofing layer as described in claim 7, characterized in that, The method for determining the hollow depth includes: When step S3 determines that the target interface is the upper interface of the hollow drum, the apparent depth value recorded in step S2 is extracted as the upper boundary depth of the hollow drum. Extract the secondary detection depth value recorded in step S5 when it is confirmed to be the real base interface as the lower boundary depth of the hollow area; The difference between the lower boundary depth and the upper boundary depth of the hollow cavity is calculated to obtain the vertical height of the hollow cavity as the hollow depth.

Images