Structural health monitoring system and method

Abstract

The invention discloses a structure health monitoring system and method, and belongs to the field of structure health intelligent monitoring. In order to realize health detection of a building wall, the invention provides an innovative monitoring system and method integrating a light source, a dynamic visual sensor and a pulse neural network. The system is used for carrying out long-term, real-time and distributed monitoring on inclination, settlement and vibration of building structures such as houses and bridges, and has the advantages of low power consumption, high real-time performance, high measurement precision and strong expandability.

Description

Technical Field

[0001] This invention relates to the field of intelligent structural health monitoring, and specifically to a structural health monitoring system and method. Background Technology

[0002] Health monitoring of building structures, especially real-time and long-term monitoring of tilt and minor vibrations, is an important technical field for disaster prevention and public safety.

[0003] Common detection methods include theodolite observation, leveling, total station observation, and tiltmeter monitoring. Theodolite and leveling methods rely heavily on manual on-site operation and data recording, making them more suitable for temporary spot checks or periodic verifications. While total station observation can achieve long-term monitoring through automated platforms, it requires additional equipment and software, and is primarily used for temporary measurements in regular applications. Tiltmeter monitoring is available in portable and fixed versions; portable versions are suitable for temporary measurements, while fixed versions can be embedded in critical building components for long-term automatic monitoring. Furthermore, monitoring methods based on vibrating wire tilt sensors and vibrating wire strain sensors, with their high precision and stability, can be integrated with data acquisition systems to achieve real-time transmission, automatic storage, and anomaly warnings. This is a commonly used new solution for long-term automatic monitoring, complementing traditional methods, but it is often more expensive and complex to install. Some known existing technologies can be found in the following list.

[0004] Prior art 1: CN115876159A, an adjustable building tilt early warning device.

[0005] Prior art 2: CN120470501A, a system for measuring the tilt of agricultural building wall structures.

[0006] Existing technology 3: CN104165619A, Laser tilt meter.

[0007] Prior art 4: CN108663021A, a device for detecting the verticality of building walls and a method for detecting using the device.

[0008] Prior art 5: CN113358094A, a method for detecting the verticality of load-bearing walls in building construction.

[0009] Prior art 6: CN219640938U, a level testing device for engineering supervision.

[0010] Existing technology 7: CN117760342A, a method for detecting the flatness of exterior walls based on laser point clouds.

[0011] There is a need in this field for a new solution that is ultra-low power consumption, low cost, can obtain detection results automatically and in a timely manner, has automatic early warning capabilities, and is scalable, in order to meet the actual deployment needs of modern large-scale building safety monitoring networks. Summary of the Invention

[0012] To alleviate or partially alleviate the above-mentioned technical problems, the solution of the present invention is as follows:

[0013] On one hand, this invention discloses a structural health monitoring system, comprising:

[0014] A light source emits light rays to an event-based sensor to obtain at least one pixel coordinate;

[0015] One of the event-based sensor and the light source is fixed to the structure, while the other is stationary relative to the ground.

[0016] On the other hand, this invention discloses a structural health monitoring system, comprising:

[0017] A light source emits light rays to an event-based sensor to obtain at least one pixel coordinate;

[0018] The event-based sensor is fixed to the structure;

[0019] The angle at which the light emitted by the light source is fixed relative to the earth.

[0020] On the other hand, this invention discloses a structural health monitoring system, comprising:

[0021] A light source emits light rays to an event-based sensor to obtain at least one pixel coordinate;

[0022] The light source is stationary relative to the earth, and the angle at which the light rays are emitted is fixed relative to the earth.

[0023] The event-based sensor is fixed to the structure.

[0024] On the other hand, this invention discloses a structural health monitoring system, comprising:

[0025] The light source will experience positional shift in the generated light spot due to the deformation of the structure.

[0026] An event-based sensor is used to sense the light spot.

[0027] On the other hand, this invention discloses a structural health monitoring system, comprising:

[0028] The light source is configured to form a light spot on the surface of the structure to be monitored;

[0029] An event-based sensor is configured to sense the movement of the light spot in an event-driven manner;

[0030] The processing unit is configured to calculate the deformation parameters of the surface of the structure to be monitored;

[0031] The light path of the light source and the surface of the structure to be monitored form a predetermined geometric relationship.

[0032] In one embodiment, the light source is a modulated light source driven by a pulse signal of a specific frequency or duty cycle.

[0033] In one embodiment, the modulation light source is an infrared laser.

[0034] In one type of embodiment, the event-based sensor is a dynamic vision sensor, or a vision sensor that includes DVS pixels and APS pixels.

[0035] In one embodiment, a spiking neural network is also included, which is configured to process the event stream of the event-based sensor output.

[0036] In one embodiment, the event-based sensor and the spiking neural network are integrated on the same hardware platform or chip to form an on-chip system for event perception and processing.

[0037] In one embodiment, the spiking neural network is configured to perform frequency-selective filtering on the event stream to extract valid events that match the modulation frequency of the light source.

[0038] In one embodiment, the structural health monitoring system is configured to calculate the displacement or vibration frequency of the structure based on the movement trajectory of the light spot.

[0039] In one embodiment, the event-based sensor is provided with a filter that allows light of at least the same wavelength as the light emitted by the light source to pass through.

[0040] In one embodiment, the structural health monitoring system is configured to perform edge computing, i.e., to process the event stream at the device end and output information indicating the deformation parameter results.

[0041] In one embodiment, a wireless communication module is also included for sending the information indicating the deformation parameter results to a remote server.

[0042] In one embodiment, the information indicating the deformation parameter result includes at least one of a device identifier, a timestamp, and a displacement or vibration frequency.

[0043] In one embodiment, the wireless communication module is configured to send information from multiple monitoring nodes regarding the results of the indicative deformation parameters to a shared gateway concentrator.

[0044] In one embodiment, the angle between the light path of the light source and the surface of the structure to be monitored forms an acute angle.

[0045] In one embodiment, the structural health monitoring system is configured to operate in an intermittent mode, wherein the light source and the event-based sensor are periodically woken up and activated by a timer.

[0046] In one embodiment, in the intermittent operating mode, the processing unit of the structural health monitoring system is configured to compare the currently acquired spot position with the stored initial reference position to calculate the long-term cumulative displacement.

[0047] On the other hand, the present invention discloses a structural health monitoring method, comprising the following steps: emitting light to an event-based sensor to obtain at least one pixel coordinate; wherein, one of the event-based sensor and the light source emitting the light is fixed to the structure, while the other is stationary relative to the ground.

[0048] The technical solution of this invention has one or more of the following beneficial technical effects:

[0049] (1) A novel technical approach is proposed, which relies on an event-based sensor and a specially designed light source, and can be further combined with a spiking neural network architecture. Edge computing is performed only when the position of the light spot is perceived, which greatly reduces the power consumption of the system.

[0050] (2) By designing the emission angle of the scintillation light source and processing the pulse neural network on-chip, it is possible to detect millimeter-level displacement and vibration frequency with high precision. Combined with the ultra-low latency characteristics of the event-based sensor, it is possible to achieve real-time monitoring and immediate early warning.

[0051] (3) The system architecture supports edge computing. Each monitoring device independently completes data processing and only sends back lightweight results. It has low communication pressure, excellent scalability, and is suitable for building a large-scale monitoring network.

[0052] Furthermore, other beneficial effects of the present invention will be mentioned in the specific embodiments. Attached Figure Description

[0053] Figure 1 This is a schematic diagram illustrating the deployment of a monitoring system according to one embodiment of the present invention;

[0054] Figure 2 This is a flowchart of the monitoring system of one embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram of the structure of a monitoring device according to one embodiment of the present invention;

[0056] Figure 4 This is a schematic diagram of the field of view of a monitoring device according to one embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram illustrating the principle of calculating wall displacement based on the movement of a light spot in one embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0059] To facilitate a clear description of the technical solutions in the embodiments of the present invention, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order.

[0060] Terminology Explanation:

[0061] The term "structural health monitoring" refers to the long-term, real-time, or periodic monitoring of the deformation of engineering structures (such as bridges, buildings, and machinery) through the integration of sensors and analysis systems.

[0062] The term "Event Based Sensor" (EBS) refers to a device that includes dynamic visual pixels capable of outputting event signals (or simply events) that reflect changes in light intensity. For example, it could be a Dynamic Vision Sensor (DVS, also known as an event camera) as commonly referred to in the art, or a sensor that simultaneously includes DVS pixels and Active Pixel Sensor (APS) pixels (commonly known in the art as a DAVIS). In an event-based sensor, the dynamic visual pixels sense changes in light intensity; when the change in light intensity (brightening or darkening) within the sensing area exceeds a certain range, an event (also known as a spike event) is generated.

[0063] Event-based sensors generate a sequence of pulse events, also known as a spike train, based on changes in light within the field of view (FOV). An event train consists of a series of events. An event can be represented by a quadruple (x, y, p, t), where x and y are the pixel coordinates that generated the event, p indicates the polarity, and t is the timestamp of the event.

[0064] The term "Spiking Neural Network" (SNN) refers to a third-generation neural network that mimics the workings of neurons in the biological brain, transmitting and processing information in the time domain through sparse, asynchronous pulses. SNNs are well-suited for processing event streams generated by event-based sensors and can perform on-chip computation on dedicated neuromorphic chips with extremely low power consumption.

[0065] Neuromorphic hardware refers to asynchronous, event-driven dedicated hardware that simulates the structure and information processing principles of biological neural networks, such as neuromorphic chips and spiking neural network processors, which natively support neural network operations such as pulse timing and membrane voltage dynamics.

[0066] The term "light source" refers to an object or device capable of emitting visible light. It can include modulated light sources modulated by various modulation methods, which can be various reasonable modulation modes. For example, a light source may emit light at a frequency of 300 Hz for 50% of each cycle (duty cycle), or it may emit light at different frequencies and duty cycles in different time intervals, which can be designed according to the application requirements. Unmodulated light sources are also possible in some embodiments, such as light sources used for periodically initiated detection tasks. However, the present invention is not limited to or construed as such.

[0067] The term "spot" refers to a bright area formed on the surface of a target or the imaging surface of a sensor after a beam of light emitted from a light source has passed through optical propagation (reflection, scattering, etc.). A spot has a certain size and shape, and may be circular, triangular, or other irregular shapes. The brightness of the spot area is significantly higher than the surrounding environment, and the light energy is relatively concentrated in this area.

[0068] In one embodiment of the present invention, a structural health monitoring system is disclosed, comprising: a light source that emits light to an event-based sensor to obtain at least one pixel coordinate; one of the event-based sensor and the light source is fixed to the structure, while the other is stationary relative to the ground.

[0069] In one embodiment of the present invention, a structural health monitoring system is disclosed, comprising: a light source that emits light to an event-based sensor to obtain at least one pixel coordinate; the event-based sensor being fixed to the structure; and the angle at which the light source emits light being fixed relative to the ground.

[0070] In one embodiment of the present invention, a structural health monitoring system is disclosed, comprising: a light source that emits light to an event-based sensor to obtain at least one pixel coordinate; the light source being stationary relative to the ground and the angle at which the emitted light is emitted being fixed relative to the ground; and the event-based sensor being fixed to the structure.

[0071] In one embodiment of the present invention, a structural health monitoring system is disclosed, comprising: a light source, the position of which is shifted due to deformation of the structure; and an event-based sensor for sensing the light spot.

[0072] In one embodiment of the present invention, a structural health monitoring system is disclosed, comprising: a light source configured to form a light spot on the surface of a structure to be monitored; an event-based sensor configured to sense the movement of the light spot in an event-driven manner; and a processing unit configured to calculate deformation parameters of the surface of the structure to be monitored; wherein the optical path of the light source and the surface of the structure to be monitored form a predetermined geometric relationship.

[0073] In one embodiment, the present invention provides a monitoring method that integrates a light source, an event-based sensor, and a spiking neural network. By employing an event-driven sensor and an SNN computation model, the present invention achieves ultra-low power consumption operation and real-time data processing at the monitoring device level, transmitting only lightweight monitoring results.

[0074] Various embodiments of the present invention can be used for long-term, real-time, and distributed monitoring of tilt, settlement, and minor vibrations of building structures such as houses and bridges. They have the advantages of extremely low power consumption, low response delay, high measurement accuracy, and ease of large-scale networking.

[0075] The following text will further elaborate on the technical solution of the present invention by taking the real-time monitoring of house tilt or vibration as an example, such as house tilt caused by foundation collapse or house vibration caused by earthquakes.

[0076] The present invention will exemplify the use of DVS as an event-based sensor in the following description, and will use this example to illustrate the technical concept of the present invention. Those skilled in the art will understand that the scheme described below is also applicable to hybrid sensors such as DAVIS that simultaneously contain DVS pixels and APS pixels.

[0077] Figure 1 This is a schematic diagram illustrating the deployment of a monitoring system according to one embodiment of the present invention. Figure 3This is a schematic diagram of the structure of a monitoring device according to one embodiment of the present invention. Figure 1 As shown, the light source is fixed on a stable base near the wall to be monitored, rather than on the wall itself. During installation, its emission angle needs to be precisely adjusted so that the light path of the light source forms an acute angle α (e.g., 20 degrees) with the wall surface (the reflector of the monitoring device). The light emitted by the light source, after being reflected by the reflector (or propagated by other optical means), will form a light spot on the DVS imaging surface of the monitoring device. This light spot will form an imaging area on the DVS imaging surface that is significantly brighter than the background.

[0078] At the same time, the monitoring device of the present invention (such as Figure 3 As shown, the plate-shaped mounting base (used to fix the equipment) is fixedly installed in a stable position on the building wall, and the observation normal of the DVS of the monitoring equipment forms another fixed angle with the wall surface (the reflector of the monitoring equipment), as shown in the figure. Figure 3 In the monitoring equipment shown, the fixed angle between the observation normal of its DVS and the wall surface (the reflector of the monitoring equipment) is 70 degrees.

[0079] Figure 4 This is a schematic diagram of the field of view of a monitoring device according to one embodiment of the present invention. Figure 4 As shown, the field of view of its DVS is a conical region. Therefore, the fixed angle formed by the observation normal of its DVS and the wall, as well as the angle formed by the light path generated by the light source and the wall, must ensure that the conical region can completely cover and continuously observe the light spot on the reflector. Specifically, the light spot formed by the light source on the reflector must always be located within the intersection section of the conical field of view of the DVS and the reflector (i.e., the dashed elliptical part in the figure). This arrangement constitutes the geometric reference of the monitoring system. It is particularly important to note that the angle α between the light path generated by the light source and the wall is a key design parameter. The smaller its value, the higher the system's sensitivity to detecting small displacements of the wall. In a specific embodiment, when the angle α between the light path generated by the light source and the wall is designed to be 20 degrees, the system can achieve extremely high detection accuracy, that is, the movement of one pixel on the DVS imaging surface of the monitoring device corresponds to an actual wall displacement of about one millimeter.

[0080] In one embodiment, the light source is a laser that emits infrared light at a specific frequency.

[0081] Figure 2 This is a flowchart of a monitoring system according to one embodiment of the present invention. After physical deployment is completed, the system enters an automated workflow, as shown below. Figure 2As shown. First, the light source begins operation, emitting flickering infrared light in a specific modulation mode (e.g., at a frequency of 300 Hz and a duty cycle of 50%). This modulation characteristic helps the monitoring device's DVS to more stably and resiliently identify and track light spots against complex ambient light backgrounds. To further enhance anti-interference performance, in one embodiment, a narrowband filter is also positioned before the optical path of the event-based sensor. The central transmission band of this filter matches the emission band of the light source (such as an infrared laser), effectively suppressing other bands of light in the ambient light while allowing at least the specific band of light emitted by the light source to enter the sensor, thereby pre-filtering out most background interference at the optical level. Furthermore, compared to continuous emission, this flickering mode of the light source itself also helps to reduce the overall power consumption of the light source.

[0082] When the monitored wall structure is in a stable state, the position of the light spot on the reflector remains unchanged. At this time, a traditional CMOS sensor would continuously acquire image frames containing a large amount of repetitive background information, consuming unnecessary power and bandwidth. However, for the event-based sensor at the core of this system (DVS as an example in this embodiment), since its working principle is "sensitive only to changes in light intensity," the DVS generates almost no output data when the position of the light spot on the reflector remains unchanged. The system maintains an extremely low static power consumption level, thus achieving the low power consumption goal of this invention.

[0083] Even the slightest displacement of the wall due to tilting, settlement, or vibration will change the position of the light spot on the reflector relative to the monitoring device. This brightness change is instantly captured by the DVS (Distributed Visual System) of the monitoring device. Because the pixel array of the DVS operates asynchronously, only those pixels whose brightness changes along the light spot's movement path will independently generate output. This output is not a complete image frame, but a series of lightweight pulse events. Each event can be described by a quadruple (x, y, p, t), where (x, y) are the pixel coordinates that generated the event, p indicates the polarity of brightness change (brightening or darkening), and t is a timestamp accurate to the microsecond. These events are arranged chronologically, forming an event stream that precisely describes the trajectory of the light spot.

[0084] These pulse event streams are immediately and directly fed into the spiking neural network (SNN) within the integrated monitoring device for processing. In a preferred embodiment, the event-based sensor (DVS) and the spiking neural network (SNN) are co-integrated on a single hardware platform or neuromorphic chip, forming a compact on-chip system for event perception and processing. This highly integrated architecture significantly reduces the latency and power consumption of data transmission between components, which is key to achieving ultra-low power consumption and real-time processing. As a third-generation neural network, the SNN's neuron model simulates the biological brain, performing sparse and asynchronous computation only when it receives sufficient event pulse stimuli. This perfectly matches the spatiotemporal sparsity of the event stream generated by the DVS, forming an ultra-low power closed loop from perception to computation. The SNN runs a pre-trained dedicated algorithm that can process the input event stream in real time, perform spot detection, recognition, and tracking tasks, accurately lock the image of the spot, and continuously track the coordinate change sequence of its centroid on the DVS imaging plane, thereby obtaining the movement trajectory of the spot.

[0085] In one embodiment, the spiking neural network can be configured with frequency-selective filtering capabilities. Its neuron dynamic parameters or synaptic weights are pre-trained to filter the temporal patterns of the input event stream, prioritizing responses to events that match the modulation frequency of the light source (e.g., 300Hz). This effectively extracts valid events generated by the modulated light source from the original event stream containing ambient light noise, significantly improving the system's signal-to-noise ratio and anti-interference capability.

[0086] In one embodiment, the algorithm within the SNN first accumulates the input event stream over time to form a short event frame, and then extracts the features of the light spot through the convolutional layers within the SNN. The activity of neurons in the output layer of the SNN corresponds to the possible target location on the DVS imaging plane. By identifying the most active group of neurons, the precise coordinates (x, y) of the current light spot centroid can be calculated in real time. This process is continuous, thus obtaining a sequence of changes in the light spot centroid over time.

[0087] After acquiring the movement trajectory of the light spot at the edge, the system then calculates the displacement or frequency. Figure 5This is a schematic diagram illustrating the principle of calculating wall displacement based on the movement of a light spot, according to one embodiment of the present invention. As shown in the figure, the diagram clearly illustrates how the light spot moves accordingly on the DVS imaging surface when the wall shifts. Based on the displacement vector (Δx, Δy) of the tracked light spot centroid on the DVS imaging surface, and combined with pre-calibrated system geometric parameters (including the installation angle of the light source, the installation angle of the DVS, and the baseline distance between the light source and the DVS), the system can calculate the actual displacement vector of the wall in the real world (including the direction and magnitude of the displacement) in real time using a built-in geometric optics and triangulation model. If the wall is vibrating, the light spot will reciprocate at high speed on the DVS imaging surface. By analyzing the high-precision timestamp information contained in the event stream and statistically analyzing the period of the light spot's reciprocating motion, the vibration frequency of the wall can be further calculated. This process has extremely high real-time performance, thanks to the low latency characteristics of the DVS and the parallel processing capabilities of the SNN.

[0088] Finally, the microcontroller within the integrated monitoring device transmits the refined, lightweight, structured data—such as displacement, vibration frequency, or alarm signals triggered when displacement exceeds a preset threshold—to a remote server (a remote signal receiving terminal or cloud platform) via its integrated low-power wireless communication module. This "edge computing, result feedback" model contrasts sharply with the traditional "full upload of raw data" model. It significantly reduces the bandwidth pressure on communication links and the computational load on data centers, enabling the system to economically and efficiently support hundreds or thousands of monitoring nodes operating simultaneously. It possesses excellent scalability and is highly suitable for building large-scale, distributed building safety monitoring networks.

[0089] In one embodiment, the present invention can be used in large-scale monitoring networks. Dozens or even hundreds of monitoring devices described in this invention can be deployed within the same large building (such as a factory or bridge) or community. Each monitoring device is a node, and each node independently completes its own sensing, calculation, and result generation. All nodes asynchronously send lightweight monitoring results (including device ID, timestamp, displacement, status flags, vibration frequency, and other information indicating deformation parameters) to a shared gateway concentrator via wireless transmission modules. The gateway then aggregates the data to a cloud-based monitoring platform. Because each node performs edge computing, the amount of data transmitted back is extremely small, avoiding data flooding. This makes it possible to build a dense monitoring network covering a wide area using low-bandwidth, low-power networks, and to analyze the overall structural deformation pattern based on the correlation of displacement data from different nodes.

[0090] In one embodiment, the system is configured to monitor long-term house tilting caused by slow foundation settlement. The operating strategy of the system is optimized: the light source and monitoring equipment do not operate continuously, but are controlled by a low-power timer in the microcontroller, waking up and operating for a period of time only at a fixed time each day (e.g., 2 AM) to collect data and calculate the cumulative displacement relative to an initial reference at that moment. This intermittent operating strategy significantly reduces power consumption, enabling battery-powered systems with long endurance, for example, up to several years. Furthermore, the SNN algorithm in this mode is configured to perform high-precision "static position" comparisons, i.e., comparing the current spot centroid position with the initial reference position stored in memory, thereby eliminating short-term vibration interference and accurately reflecting long-term, slow tilting trends.

[0091] In one embodiment, the system is used to monitor high-frequency structural vibrations such as those caused by earthquakes. In this case, the system is set to continuous full-power operation. The time resolution of the monitoring device's DVS is set to the highest level to ensure the capture of rapid, minute movements of the light spot. The SNN algorithm is optimized for extremely low processing latency, focusing on real-time output of the light spot's position sequence without complex historical trajectory smoothing. By analyzing the timestamps of the high-density event stream and the light spot's position sequence, the microcontroller can quickly perform Fourier transform or zero-crossing detection analysis, calculating the dominant frequency and maximum amplitude of the wall vibration in real time at the edge side. When the vibration exceeds a safety threshold, an emergency warning signal is issued, which can be used for disaster early warning.

[0092] One implementation process of this invention starts with system deployment, proceeds through light emission from the light source, DVS event-driven perception, SNN edge intelligent computing, geometric parameter calculation, and finally result feedback. The entire chain deeply integrates the core concepts of event-driven and edge computing, thereby achieving low power consumption, high real-time performance, high precision, and powerful large-scale deployment capabilities that are difficult for traditional CMOS image solutions to match.

[0093] In one embodiment, the angle at which the light source emits light is fixed relative to the ground. The position of the light source emitting device can be changed during operation according to specific needs. Specifically, the light source is configured such that the optical axis direction of its emitted light remains fixed relative to the Earth's coordinate system. This configuration ensures the high precision of the monitoring system, thereby providing a reliable absolute reference for displacement calculation. It is worth noting that the physical position of the light source emitting device itself can be flexibly adjusted according to the specific monitoring scenario and installation conditions. For example, it can be precisely positioned during system initialization, or its position can be finely adjusted during long-term monitoring to meet maintenance needs. As long as the angle of its emitted light remains unchanged, and through specific calculations by the processing unit, the accuracy of the monitoring system will not be affected.

[0094] In one embodiment, one of the event-based sensor and the light source is fixed to the structure, while the other is stationary relative to the ground. This means that, in actual deployment, the roles of the two can be interchanged depending on specific needs. For example, in one implementation, the event-based sensor can be fixed to the structure, while the light source is placed in a stationary environment; in another equivalent implementation, the light source can also be fixed to the structure, while the event-based sensor is placed in a stationary environment. These two deployment methods are physically symmetrical, essentially constructing a "motion-stationary" relative measurement system that accurately calculates the deformation and displacement of the structure by monitoring changes in the optical path information between the component fixed to the structure and the stationary component. This design flexibility allows the invention to easily adapt to various complex field installation conditions.

[0095] In one embodiment, the event-based sensor and the light source are fixedly mounted on two different monitored structures. Although this embodiment cannot accurately detect which specific monitored structure has tilted, the monitoring result indicates that at least one monitored structure has tilted. This provides timely and necessary information for early warning in the future. Specifically, the light source is fixed to a first structure (e.g., the wall of building A), and the event-based sensor is fixed to a second structure (e.g., the wall of building B). The light emitted by the light source enters the sensor's field of view directly or after reflection, forming a light spot on its imaging surface. If a relative displacement (including relative tilting, settlement, or vibration) occurs between the two structures, the position of the light spot on the sensor's imaging surface will change, thereby being captured by the event-driven sensor and generating an event stream. If the system detects a relative displacement between the two monitored structures, the system can issue an early warning, prompting maintenance personnel to further investigate and measure the two structures to determine which monitored structure has a problem, or to determine whether both have displaced. In actual engineering monitoring, this can be applied to scenarios where it is necessary to monitor the relative displacement between two adjacent or related structures (e.g., adjacent buildings, adjacent bridge piers, etc.). For example, adjacent buildings may tilt relative to each other due to differential settlement of the foundation; relative displacement monitoring between adjacent piers in a bridge, etc. This embodiment also has the advantages of low power consumption, high real-time performance and edge computing, and is especially suitable for preliminary screening of the health status and long-term trend monitoring of paired structures.

[0096] In response to the possibility of fine water droplets forming on the reflector due to humid weather, in practice, to reduce the impact of this adverse condition on the present invention, the actual effect of the solution can be optimized from the following perspectives: design a beam with a sufficiently small beam diameter, sufficiently high beam energy, and sufficiently consistent beam diameter to minimize pixel output events; and allow for greater tolerance in actual tilt judgment after experimental testing of multiple sets of data to address the impact of humid weather on imaging results.

[0097] To better illustrate the present invention, numerous specific details have been provided in the detailed embodiments described above. Those skilled in the art should understand that the present invention can be practiced even without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of the present invention.

[0098] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A structural health monitoring system, characterized in that, include: A light source emits light rays to an event-based sensor to obtain at least one pixel coordinate; One of the event-based sensor and the light source is fixed to the structure, while the other is stationary relative to the ground.

2. A structural health monitoring system, characterized in that, include: The light source is configured to form a light spot on the surface of the structure to be monitored; An event-based sensor is configured to sense the movement of the light spot in an event-driven manner; The processing unit is configured to calculate the deformation parameters of the surface of the structure to be monitored; The light path of the light source and the surface of the structure to be monitored form a predetermined geometric relationship.

3. The structural health monitoring system according to any one of claims 1 to 2, characterized in that: The light source is a modulated light source, which is driven by a pulse signal of a specific frequency or duty cycle.

4. The structural health monitoring system according to claim 2, characterized in that: The structural health monitoring system is configured to calculate the displacement or vibration frequency of the structure based on the movement trajectory of the light spot.

5. The structural health monitoring system according to any one of claims 1 to 2, characterized in that: It also includes a spiking neural network configured to process the event stream of the event-based sensor output.

6. The structural health monitoring system according to any one of claims 1 to 2, characterized in that: The structural health monitoring system is configured to operate in an intermittent mode, wherein the light source and the event-based sensor are periodically woken up and activated by a timer.

7. The structural health monitoring system according to claim 6, characterized in that: In the intermittent working mode, the processing unit of the structural health monitoring system is configured to compare the currently acquired spot position with the stored initial reference position to calculate the long-term cumulative displacement.

8. The structural health monitoring system according to claim 2, characterized in that: The event-based sensor and the light source are respectively fixedly installed on two different monitored structures.

9. The structural health monitoring system according to any one of claims 1 to 2, characterized in that: The event-based sensor is provided with a filter that allows light of at least the same wavelength as the light emitted by the light source to pass through.

10. A method for monitoring structural health, characterized in that, Includes the following steps: Emit light rays to an event-based sensor to obtain at least one pixel coordinate; In this configuration, one of the event-based sensor and the light source emitting the light is fixed to the structure, while the other is stationary relative to the ground.

Images