Looseness early warning device for high-altitude building scaffold

Abstract

The application provides a loosening early warning device of high-altitude building scaffold, and belongs to the technical field of building construction safety monitoring. The method collects multi-dimensional monitoring data of the scaffold connecting bolt, obtains a standardization factor by combining normalization processing, and then calculates the occurrence probability of bolt loosening based on a preset algorithm to finally complete the loosening early warning judgment. Through the method, early loosening hidden troubles of high-altitude scaffold bolts can be automatically and accurately identified without manual climbing and checking, which reduces the safety inspection labor cost, effectively improves the identification efficiency and early warning timeliness of scaffold loosening hidden troubles, and ensures the safety of high-altitude building construction operation.

Description

Technical Field

[0001] This application relates to the field of safety monitoring technology for high-altitude construction, and in particular to a loosening early warning device for high-altitude construction scaffolding. Background Technology

[0002] High-altitude construction scaffolding is a commonly used edge protection and work support facility in the construction process. Scaffolding is assembled and connected by uprights and horizontal bars with bolts and clamps. During long-term outdoor operations, the bolts at the connection points are prone to slow loosening due to wind swaying, temperature deformation, and construction load disturbance. If the loose bolts are not detected in time, it is very easy to cause safety accidents such as scaffolding instability and collapse.

[0003] Existing technologies for detecting loose scaffold bolts mostly rely on manual periodic inspections. High-altitude inspections are not only risky but also have long inspection cycles, making real-time monitoring impossible. Some scaffold inspection solutions that can achieve automatic monitoring require an independent power supply module for each monitoring point. Frequent battery replacements increase maintenance costs, while wiring power supplies significantly increases modification costs. Overall, these solutions are not economically viable and are difficult to promote and apply on a large scale. Summary of the Invention

[0004] The purpose of this invention is to solve the above-mentioned problems by providing a loosening early warning device for high-altitude construction scaffolding.

[0005] The technical solution of this application is implemented as follows: In a first aspect, this application provides a loosening early warning device for high-altitude construction scaffolding, the scaffolding including clamps and uprights, the uprights being connected by clamps, and the early warning device comprising: A thin-film pressure sensor is mounted on the connecting lug of the clamp; A passive RFID tag is installed on the clamp, and the passive RFID tag is electrically connected to the membrane pressure sensor. Each passive RFID tag has a unique identifier. An RFID reader is installed on the pole, and the radio frequency signal coverage of the RFID reader covers at least one of the passive RFID tags. The RFID reader transmits radio frequency signals to the passive RFID tag, activating the passive RFID tag and powering it. The passive RFID tag is activated in response to the radio frequency signal, generates a response signal based on the current resistance value of the thin-film pressure sensor, and sends the response signal along with its unique identifier back to the RFID reader. A data processing unit is installed on the scaffold and coupled to the RFID reader / writer. The data processing unit is configured to determine whether the clamp has become loose based on the change information of the resistance value in the response signal and to locate the loose clamp based on the unique identifier. The data processing unit is also configured to provide comprehensive early warning based on multi-dimensional parameters, specifically: The RFID reader receives the response signal and transmits it to the data processing unit. The data processing unit collects the historical resistance value sequence of the thin-film pressure sensor, calculates the pressure value change trend and continuous descent rate; collects the current time information of the scaffold location, distinguishing between daytime and nighttime periods; obtains climate information of the scaffold location through networked meteorology, including temperature, humidity, and rainfall; obtains wind speed data of the scaffold location through a wind speed sensor; and establishes a comprehensive judgment model based on the above data. The comprehensive judgment model is constructed as follows: using the pressure descent rate, the time period weight determined based on the current time information, the climate influence factor, and the wind speed influence factor as input parameters, it calculates the loosening risk score; when the loosening risk score exceeds a preset risk threshold, an early warning signal is issued.

[0006] In one embodiment, the early warning device further includes: A clock module, connected to the data processing unit, is used to obtain current time information to distinguish between daytime and nighttime periods; A meteorological data interface, connected to the data processing unit, is configured to obtain real-time climate information of the location of the scaffolding from a meteorological server via wireless communication. The climate information includes temperature, humidity, and rainfall. A wind speed sensor, connected to the data processing unit, is installed on the pole to acquire real-time wind speed data.

[0007] In one embodiment, the clamp includes two clamp bodies that can rotate relative to each other. Connecting ears are rotatably mounted on the clamp bodies. The connecting ears form an encircling space for fastening the upright with the clamp bodies by bolts. One end of the bolt is rotatably connected to the clamp body, and the other end is fastened to the connecting ear by a nut. A washer is also installed between the nut and the connecting ear. The thin-film pressure sensor is installed between the gasket and the connecting lug, and the thin-film pressure sensor converts the loose or tight state of the nut into a change in resistance value; The passive RFID tag is installed on the connecting ear or the clamp body; The RFID reader is configured to periodically transmit radio frequency signals to the passive RFID tag to activate the passive RFID tag and receive a response signal returned by the tag containing information about the change in resistance value and the unique identifier.

[0008] In one embodiment, the passive RFID tag includes an RFID chip with universal input / output pins, and the thin-film pressure sensor is electrically connected to the universal input / output pins.

[0009] In one embodiment, the thin-film pressure sensor is a resistive thin-film pressure sensor.

[0010] In one embodiment, the data processing unit is configured to determine whether the clamp has become loose based on the change in resistance value in the response signal, including: S101. The RFID reader transmits a radio frequency signal to the passive RFID tag to activate the passive RFID tag and power it. S102. In response to being activated, the passive RFID tag generates a response signal based on the current resistance value of the thin-film pressure sensor, and sends the response signal along with its unique identifier back to the RFID reader. S103, The RFID reader receives the response signal and transmits it to the data processing unit; S104. The data processing unit parses the response signal, determines whether the clamp has become loose based on the change information of the resistance value, and locates the loose clamp based on the unique identifier. S105. If it is determined that the clamp is loose, the data processing unit issues an alarm and marks the location of the loose clamp.

[0011] In one embodiment, the data processing unit is set with a safe pressure threshold. When the pressure value corresponding to the resistance value is lower than the safe pressure threshold, the data processing unit triggers a loosening alarm. In one implementation, the comprehensive judgment model is a weighted scoring model, and the loosening risk score is calculated according to the following formula: Risk=w1·F1+w2·F2+w3·F3+w4·F4; Where Risk is the risk score for loosening, w1, w2, w3, and w4 are preset weight coefficients, and w1+w2+w3+w4=1, and F1, F2, F3, and F4 are the normalization functions of each parameter. When the loosening risk score exceeds a preset risk threshold, an early warning signal is issued.

[0012] In one embodiment, the data processing unit has a preset loosening risk threshold. When the loosening risk score exceeds the loosening risk threshold, it determines that there is a loosening risk and issues a warning signal.

[0013] The advantages or beneficial effects of the above technical solutions include at least the following: This invention employs a passive RFID tag combined with a thin-film pressure sensor. It eliminates the need for an independent power supply module for each monitoring point; the passive tag is powered by the radio frequency signal emitted by the RFID reader to complete data acquisition. This significantly reduces equipment modification and maintenance costs, making it more suitable for large-scale application. Furthermore, by incorporating multi-dimensional environmental parameters such as time, climate, and wind speed for comprehensive judgment, it can provide early warnings of bolt loosening risks based on environmental load factors. Compared to solutions relying solely on single-point pressure values, this further improves the accuracy of early warnings, enabling earlier detection of potential loosening risks and enhancing the safety of scaffolding operations. Attached Figure Description

[0014] The accompanying drawings illustrate exemplary embodiments of the present application and, together with the description thereof, serve to explain the principles of the present application. These drawings are included to provide a further understanding of the present application and are incorporated in and constitute a part of this specification.

[0015] Figure 1 A schematic diagram of the scaffolding structure according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of the installation of the clamp and the upright pole according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of the clamp structure according to an embodiment of the present invention is shown; Figure 4 A schematic diagram of a passive RFID tag and a thin-film pressure sensor according to an embodiment of the present invention is shown; Figure 5 A schematic diagram of the reading and writing range of an RFID reader / writer according to an embodiment of the present invention is shown, wherein the dashed circle represents the reading and writing range of the RFID reader / writer, and the center of the dashed circle represents the installation position of the RFID reader / writer.

[0016] Reference numerals: 10, clamp; 11, clamp body; 12, connecting ear; 13, bolt; 131, nut; 14, washer; 15, passive RFID tag; 16, membrane pressure sensor; 20, scaffolding. Detailed Implementation

[0017] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the drawings and embodiments of this application are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0018] It should be noted that, where there is no conflict, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] It should be understood that the term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc., mentioned in this application are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0020] It should be noted that the terms "one" and "more" used in this application are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of this application are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0022] This invention proposes a loosening early warning device for high-altitude construction scaffolding, which has the ability to autonomously monitor and warn of bolt loosening. The core idea is to integrate a thin-film pressure sensor 16 and a passive RFID tag 15 at the fastening connection of the scaffolding clamp 10, enabling wireless and passive sensing of the pre-tightening force of the bolt 13, and achieving intelligent early warning through multi-dimensional environmental information.

[0023] The scaffolding 20 mainly consists of uprights and clamps 10 for connecting adjacent uprights. Each clamp 10 includes two clamp bodies 11 that can rotate relative to each other, forming a space to accommodate the uprights when closed. Each clamp body 11 has a rotatable connecting lug 12 at its open end. Bolts 13 pass through holes in the connecting lugs 12 and engage with nuts 131 to fasten the two clamp bodies 11, thus securely holding the uprights. A washer 14 is also provided between the contact surfaces of the nut 131 and the connecting lug 12 to improve stress distribution and prevent loosening.

[0024] The key improvement of this invention is that the warning device includes a thin-film pressure sensor 16 installed between the gasket 14 and the connecting lug 12. The thin-film pressure sensor 16 directly bears the preload pressure from the nut 131, and therefore its resistance value changes with the tightness of the nut 131, i.e., the pressure applied to the gasket 14.

[0025] When the nut 131 gradually loosens due to vibration or other reasons, the clamping force acting on the washer 14 decreases, and the resistance value of the thin-film pressure sensor 16 changes accordingly, thereby converting the mechanical loosening amount into a measurable electrical signal. Preferably, the thin-film pressure sensor 16 is a resistive thin-film pressure sensor 16.

[0026] To enable wireless and battery-free reading of loosening signals, the warning device also includes passive RFID tags 15 mounted on the connecting lugs 12 or the body 11 of the clamp 10. Each passive RFID tag 15 is pre-written with a unique identification code to distinguish different monitoring points on the scaffolding 20 structure. A thin-film pressure sensor 16 is electrically connected to the universal input / output pins of the passive RFID tag 15 via wires. In one embodiment, the core of the passive RFID tag 15 is an RFID chip with universal input / output pins, capable of converting the input external resistance value into a digital signal.

[0027] by Figure 5 Taking the scaffolding 20 structure as an example, the upper layer of the entire monitoring system consists of at least one RFID reader installed on the uprights of the scaffolding 20, and a data processing unit coupled to the reader. The installation position of the RFID reader must ensure that the radio frequency signal it emits can effectively cover and activate at least one passive RFID tag 15 in its installation area. During operation, the reader is configured to periodically emit radio frequency signals of a specific frequency. When a passive RFID tag 15 enters the radio frequency field range of the reader, it will collect radio frequency energy through its antenna to briefly power itself, thereby activating the tag. The activated RFID tag will immediately measure the current resistance value of the thin-film pressure sensor 16 through its pins, convert the analog value into a digital value, and combine it with its unique identification code to package it into a response signal. Finally, the response signal is sent back to the RFID reader via radio frequency backscattering.

[0028] After receiving the response signal, the RFID reader forwards it to the data processing unit. The data processing unit parses the unique identifier and resistance value reflecting the pressure contained in the signal. By comparing the resistance value with the reference resistance value in the initial installation state or the set safety threshold, the data processing unit can determine whether the clamp 10 connection has become loose. For example, when the pressure value corresponding to the parsed resistance value drops below the preset safety pressure threshold, it can be determined as loose. At the same time, using the unique identifier attached to the signal, the data processing unit can quickly associate this warning information with the specific physical location of the scaffolding 20, thereby indicating the exact location where the loosening occurred to maintenance personnel.

[0029] In some more refined embodiments, the scaffolding 20 system also integrates environmental factor sensing capabilities to achieve earlier, predictive, and comprehensive early warnings. In this case, the data processing unit also connects to a clock module, a wireless weather data interface, and at least one wind speed sensor mounted on the uprights. The clock module provides the current date and time information, distinguishing between daytime and nighttime periods. The weather data interface continuously acquires real-time climate information, including temperature, humidity, and rainfall, from an external weather service via a wireless communication network. The wind speed sensor is responsible for collecting actual instantaneous wind speeds at the site.

[0030] In this embodiment, the data processing unit no longer relies solely on a single pressure threshold for judgment, but instead runs a comprehensive judgment model. This model correlates the historical resistance value change trend of the thin-film pressure sensor 16, especially the continuous rate of pressure decrease, with the aforementioned multi-source information such as time, climate, and wind speed, thereby outputting a loosening risk score or loosening probability value.

[0031] In one specific implementation, the data processing unit calculates the loosening risk score using a weighted scoring model. It can collect and calculate the normalized value of the pressure drop rate, a risk weighting factor based on the current day / night cycle, a normalized climate influence factor based on temperature and humidity, and a normalized wind speed influence factor based on real-time wind speed. These four factors are each assigned a preset weighting coefficient, and a comprehensive loosening risk score is obtained through weighted summation. Each weighting coefficient can be set based on experience or experimental data, and the sum is one. When this score exceeds a preset risk threshold, the data processing unit immediately issues an early warning signal, reminding maintenance personnel to conduct a preliminary inspection even before bolt 13 is completely loosened, provided the risk conditions are met.

[0032] The warning steps of the warning device 20 of the present invention include: S101. The passive RFID tag 15 is activated and powered by the RFID reader / writer by transmitting radio frequency signals to the passive RFID tag 15. S102, In response to being activated, the passive RFID tag 15 generates a response signal based on the current resistance value of the thin-film pressure sensor 16, and sends the response signal along with its unique identifier back to the RFID reader. S103, The RFID reader receives the response signal and transmits it to the data processing unit; S104. The data processing unit analyzes the response signal, determines whether the clamp 10 has become loose based on the change in resistance value, and locates the loose clamp 10 based on the unique identifier. S105. If it is determined that the clamp 10 is loose, the data processing unit issues an alarm and marks the location of the loose clamp 10.

[0033] The data processing unit is set with a safe pressure threshold. When the pressure value corresponding to the resistance value is lower than the safe pressure threshold, a loosening alarm is triggered.

[0034] Scaffolding 20 also includes: The clock module is used to obtain current time information to distinguish between daytime and nighttime periods; The meteorological data interface is configured to obtain real-time climate information of the location of the scaffold 20 from the meteorological server via wireless communication. The climate information includes temperature, humidity, and rainfall. A wind speed sensor, installed on the pole, is used to acquire real-time wind speed data. The data processing unit is also configured to: perform correlation analysis on the resistance value change information of the thin-film pressure sensor 16 with at least one of the time information, climate information, and wind speed information. The data processing unit outputs a loosening risk score or loosening probability value based on a comprehensive judgment model. When the score or probability value exceeds a preset threshold, an early warning signal is issued.

[0035] The above comprehensive judgment model is a weighted scoring model. The loosening risk score is calculated according to the following formula: Risk=w1·F1+w2·F2+w3·F3+w4·F4; Where Risk is the risk score for loosening, w1, w2, w3, and w4 are preset weight coefficients, and w1+w2+w3+w4=1, and F1, F2, F3, and F4 are the normalization functions of each parameter. When the risk score for loosening exceeds the preset risk threshold, an early warning signal is issued.

[0036] The data processing unit has a preset loosening risk threshold. When the loosening risk score exceeds the loosening risk threshold, it determines that there is a loosening risk and issues a warning signal. For example, when the loosening risk score exceeds the preset loosening risk threshold but the pressure value corresponding to the current resistance value has not yet fallen below the safe pressure threshold, a warning prompt is issued to require the bolt 13 at the corresponding position to be pre-inspected, reminding maintenance personnel to check for potential hazards in advance. If the pressure value corresponding to the resistance value has already fallen below the safe pressure threshold, a loosening alarm is directly triggered, requiring immediate tightening and maintenance.

[0037] The comprehensive judgment model described above can be implemented using a weighted scoring model. The loosening risk score is defined as the weighted sum of four normalized factors: Risk = w1·F1 + w2·F2 + w3·F3 + w4·F4; where F1 is a pressure-related factor, which can be normalized based on the ratio of the current pressure value to the initial pressure value or the continuous rate of pressure decrease; F2 is a time factor, for example, a higher risk coefficient is assigned during nighttime when there is no inspection and the wind is stronger; F3 is a climate factor, which comprehensively considers the effects of temperature and humidity on metal creep and corrosion, as well as the aggravating effect of rainfall on vibration; F4 is a wind speed factor, which is normalized based on the real-time wind speed. The sum of each weight coefficient w1, w2, w3, and w4 is 1. The specific values ​​can be set based on historical data or engineering experience, for example, w1=0.5, w2=0.1, w3=0.15, w4=0.25. A loosening risk threshold is preset in the data processing unit, for example, 0.7. When the calculated Risk value exceeds 0.7, it is determined that there is a high risk of loosening, and an early warning signal is issued.

[0038] The following example illustrates the process of comprehensive early warning judgment. For example, on a high-rise building construction scaffold (20), the system monitored a clamp (10) at 2 AM. The membrane pressure sensor (16) showed a resistance value sequence indicating that the pressure had continuously decreased from 5 kN to 4 kN over the past 6 hours, at a rate of 0.167 kN / hour. The current time factor is marked as a high-weight period at night. Data from the meteorological interface shows a real-time temperature of 28℃, humidity of 85%, and light rain, resulting in a relatively high value for the climate factor after normalization. The wind speed sensor measured an average wind speed of 12 m / s, and the wind speed factor also increased accordingly. The system substitutes these parameters into the weighted scoring model and calculates the current loosening risk score: Risk = 0.5 × (4 / 5 normalized value approximately 0.8) + 0.1 × 1 + 0.15 × 0.9 + 0.25 × 0.8 ≈ 0.835. Since it exceeds the preset threshold of 0.7, the data processing unit immediately issues an early warning, prompting the inspection personnel to prioritize tightening the node during the daytime of the following day, thereby avoiding the risk of bolt 13 completely loosening.

[0039] In addition to the weighted scoring model, this implementation also provides a comprehensive early warning scheme based on a machine learning model. First, the resistance value sequence of the thin-film pressure sensor 16 is continuously collected over a historical period to form a pressure change feature vector, extracting statistical features such as mean, variance, rate of decline, and number of consecutive days of decline. Simultaneously, time features (time of day, whether it is a working day), climate features (temperature, humidity, rainfall), and wind speed features (average wind speed, gust wind speed) are collected within the corresponding time period to form an environmental feature vector. These pressure change feature vectors are then concatenated and fused with the environmental feature vector as input samples.

[0040] A classification model is pre-trained using a large number of historical normal and loose fault samples. The classification model can be a logistic regression model, support vector machine, random forest, gradient boosting tree, or lightweight neural network. The trained model can output the probability value of the clamp 10 being in a loose state based on the input feature vector. In actual monitoring, the system repeats the above feature collection process at fixed intervals, inputting the data into the classification model to obtain a looseness probability value between 0 and 1. The data processing unit presets a probability threshold, for example, 0.8. When the output looseness probability value exceeds 0.8, even if the pressure has not yet fallen below the safe pressure threshold, the system will still issue an early warning signal to remind maintenance.

[0041] In another embodiment, the early warning method further includes a step of making a comprehensive early warning judgment based on multi-dimensional parameters: S201. Collect the historical resistance value sequence of the thin-film pressure sensor 16, and calculate the changing trend and continuous rate of decrease of the pressure value; S202. Collect the current time information of the location of scaffolding 20, and distinguish between daytime and nighttime periods; S203. Obtain climate information of the location of scaffolding 20 through networked meteorology, including temperature, humidity and rainfall. S204. Obtain wind speed data at the location of scaffold 20 using a wind speed sensor; S205. Based on the above data, the comprehensive judgment model is constructed as follows: the loosening risk score is calculated using the pressure drop rate, the time period weight determined based on the current time information, the climate influence factor, and the wind speed influence factor as input parameters. S206. When the loosening risk score exceeds the preset risk threshold, an early warning signal is issued.

[0042] To further illustrate the specific execution process of the weighted scoring model based on multi-dimensional parameters, the following section uses a construction example to explain in detail the complete process from raw data collection to final score calculation.

[0043] Scene setup and raw data collection: During the construction of the facade of a high-rise building, scaffolding 20 was located in a windy area with harsh environmental conditions. The system performed a periodic comprehensive risk assessment at 1:00 AM on a certain day. For a clamp node 10 (unique identifier RFID_NW_1503) at the northwest corner of the 15th floor of scaffolding 20, the system simultaneously collected the following four types of raw data: First, pressure data. Information was extracted from the historical resistance value sequence of the 16 membrane pressure sensor at node 10 of the clamp. The system record shows that the initial preload at this node was 5.0 kN. Over the past 12 consecutive hours, the pressure value gradually decreased from 4.8 kN to 4.2 kN. Based on this, the data processing unit calculated the continuous rate of decrease in pressure as (4.8 - 4.2) kN ÷ 12 hours = 0.05 kN / hour. Simultaneously, the ratio of the current pressure value to the initial preload is 4.2 ÷ 5.0 = 0.84.

[0044] Second, time data. The clock module obtains the current time as 1:00 AM. According to the system's preset time period division rules (for example, 6:00 AM to 10:00 PM is the daytime period, and 10:00 PM to 6:00 AM the next day is the nighttime period), the current time is determined to be a high-weight nighttime period.

[0045] Third, climate data. Real-time data obtained from the meteorological service via the meteorological data interface shows: the current temperature is 32℃, the relative humidity is 90%, the weather condition is continuous moderate rain, and the rainfall is 5 mm / hour.

[0046] Fourth, wind speed data. The wind speed sensor installed at this pole measured the current average wind speed to be 10 m / s, with gusts reaching 15 m / s.

[0047] Normalization process for each factor: The system then inputs the raw data into the corresponding normalization functions to convert them into dimensionless factor values ​​between 0 and 1.

[0048] For the pressure-related factor F1, the system employs a normalization rule that integrates the pressure ratio and the rate of decrease. This rule uses a pressure ratio of 0.84 and a rate of decrease of 0.05 kN / h as inputs. The rate of decrease of 0.05 kN / h is mapped to a rate score of 0.6 (for example, according to a pre-defined linear mapping table: a rate ≥ 0.1 kN / h scores 1.0 point, a rate 0.05 kN / h scores 0.6 points, and a rate of 0 scores 0 points). Finally, F1 is calculated as the weighted average of the pressure ratio (0.84) and the rate score (0.6), assuming each has a weight of 0.5, resulting in F1 = 0.5 × 0.84 + 0.5 × 0.6 = 0.72. This value reflects both the degree to which the current pressure deviates from its initial state and incorporates the recent trend of weakening pressure.

[0049] For the time factor F2, the system directly maps the nighttime period to F2=1.0 according to the preset time period and risk weight correspondence table.

[0050] For climate factor F3, the system inputs three parameters—temperature 32℃, humidity 90%, and rainfall 5 mm / hour—into a predefined climate normalization function. The function's rules are as follows: temperatures exceeding 30℃ contribute 0.2, humidity exceeding 80% contribute 0.3, and rainfall exceeding 2 mm / hour contributes 0.5. The sum of these scores is truncated to 1.0. Under the current conditions, 0.2 + 0.3 + 0.5 = 1.0, therefore F3 is set to 1.0.

[0051] For the wind speed factor F4, the system comprehensively evaluates and normalizes the average wind speed and gust wind speed. A gust of 15 m / s corresponds directly to a higher wind speed factor score according to the system's preset mapping relationship (for example, a gust of ≥12 m / s is considered high risk). The final normalized value of the wind speed factor F4 is 0.85.

[0052] Calculation and early warning triggering of loosening risk score: The summaries of each factor value and the preset weight coefficients are shown in Table 1 below:

[0053] Table 1 The data processing unit calculates the current loosening risk score as Risk = 0.360 + 0.100 + 0.150 + 0.2125 = 0.8225.

[0054] The loosening risk score of 0.8225 exceeded the system's preset risk threshold of 0.7, prompting the data processing unit to trigger an early warning mechanism. The system automatically generated an early warning work order, which included: the unique identifier of the warning node (RFID_NW_1503), the loosening risk score of 0.8225, the risk level of "high," and the recommended action of "prioritizing pre-tightening checks and tightening." The work order was then pushed to the handheld terminal of the on-duty management personnel.

[0055] The following morning at 7:00 AM, following the instructions of the early warning work order, maintenance personnel arrived at the designated clamp location 10 in the northwest corner of the 15th floor. After using a calibrated torque wrench for inspection, they found that nut 131 at that location had visibly loosened, and the actual preload had indeed decreased, verifying the accuracy and timeliness of this multi-dimensional comprehensive early warning. The maintenance personnel then retightened the point, eliminating the safety hazard.

[0056] As a parallel embodiment, the early warning method also includes a step of making a comprehensive early warning judgment based on a machine learning model: S301. Collect the resistance value sequence of the thin-film pressure sensor 16 within a historical time period to form a pressure change feature vector; S302. Synchronously collect the time characteristics, climate characteristics and wind speed characteristics of the location of scaffolding 20 to form an environmental feature vector; S303. Fuse the pressure change feature vector with the environmental feature vector and input it into a pre-trained classification model; S304. The classification model outputs the probability value of the loosening of clamp 10 in its current state; S305. When the probability value of loosening exceeds the preset probability threshold, an early warning signal is issued; The classification model can be any of the following: logistic regression, support vector machine, random forest, gradient boosting tree, or lightweight neural network; the random forest model is preferred. For example, after three months of operation, a clamp 10, although there was no obvious single pressure drop, showed a slow downward trend in pressure characteristics. It had also been in a high-temperature and high-humidity environment for several consecutive days recently, with short-term strong winds at night. These features, when input into a trained random forest model, resulted in a loosening probability of 0.85, exceeding the 0.8 threshold, prompting an early warning from the system. On-site inspection by maintenance personnel revealed that nut 131 was indeed slightly loose, and timely retightening prevented further deterioration.

[0057] The calculation process for the loosening probability output by the above model is as follows: Step 1: Data Preparation and Feature Construction The system selects the past 7 days as a monitoring window and generates a complete feature data for the 10 nodes of the currently monitored clamp to determine its loosening status.

[0058] First, pressure change characteristics are extracted. These are four values ​​calculated from the resistance value sequence of the 16-phase membrane pressure sensor: First, the pressure drop rate, calculated as the average rate of pressure drop over the past 24 hours, yielding a result of 0.12 kN / day; second, the pressure fluctuation degree, calculated as the standard deviation of pressure values ​​over the past 7 days, yielding a result of 0.05 kN, indicating that the pressure fluctuates but not drastically; third, the current pressure ratio, calculated as the ratio of the current pressure of 4.0 kN to the initial preload of 5.0 kN, yielding a result of 0.8. These three values ​​constitute the pressure change characteristic vector [0.12, 0.05, 0.8].

[0059] Secondly, environmental features were extracted. The system synchronously acquired monitored environmental data: the clock module identified the time as 3:00 AM, which is nighttime, so the "nighttime" feature was marked as 1; real-time climate information obtained through the meteorological data interface was 30℃, humidity 85%, and rainfall 2 mm / hour, which, after normalization, yielded a climate feature vector of high temperature, high humidity, and rain; the real-time average wind speed obtained through the wind speed sensor was 8 m / s, with gusts of 12 m / s, forming a wind speed feature vector. Finally, the pressure change feature vector and the environmental feature vector were concatenated end-to-end to form a complete feature vector used for model judgment.

[0060] Step 2: Independent judgment within the random forest model The constructed feature vector is simultaneously fed into 100 decision trees within a pre-trained random forest model. Each decision tree independently determines "yes" or "no" to whether the clamp 10 will loosen in its current state, based on its own rules. This process is akin to a vote by 100 independent experts.

[0061] The decision-making process of several decision trees can be illustrated by example: The rule for the first decision tree might be: First, determine if the current pressure ratio is less than 0.85? Since the current value is 0.8, the condition is met; then, determine if the nighttime period is 1? Since the current value is 1, the condition is met; finally, the decision tree gives the judgment "Yes, it will loosen".

[0062] The rules of the second decision tree may focus more on environmental factors. For example, it first determines whether the gust speed is greater than 10 m / s. Since the current gust speed is 12 m / s, the condition is met. Then it determines whether the humidity is greater than 80%. Since the current humidity is 85%, the condition is met. Finally, the decision tree also gives a "yes" judgment.

[0063] The rules of the third decision tree may be more complex, taking into account both the pressure drop rate and humidity. For example, it may ask whether the pressure drop rate is greater than 0.1 kN / day and the humidity is greater than 80%. Since the current value is 0.12 kN / day and the humidity is 85%, both conditions are met, so the decision is also "yes".

[0064] Of course, some of the 100 decision trees will also judge "no, will not loosen" based on their characteristics and rules.

[0065] Step 3: Calculating Probability Values ​​and Making the Final Decision The model analyzed the judgments of these 100 decision trees. The results showed that 85 trees gave the judgment "yes, it will loosen," and 15 trees gave the judgment "no, it will not loosen."

[0066] At this point, the model calculates the "probability of loosening" for node 10 of the clamp using a simple formula: Probability of loosening = (Number of decision trees that judge it to be "loose") / (Total number of decision trees). That is, 85 / 100 = 0.85. This value of 0.85 represents the model's assessment that the current state of clamp 10 has 85% similarity to historical cases that are about to loosen and require immediate attention.

[0067] Finally, the data processing unit compares the calculated probability value of 0.85 with the system's preset warning threshold of 0.8. Since 0.85 > 0.8, the system determines that the probability is true and immediately generates and sends an early warning signal containing the unique identifier of the clamp 10 and the probability value of loosening.

[0068] Through the above technical solution, the loosening early warning device 20 for high-altitude building scaffolding combines passive wireless sensing technology, environmental perception technology and data analysis methods to achieve all-round bolt 13 loosening monitoring from single-point threshold alarm to multi-dimensional intelligent early warning, which significantly improves construction safety and maintenance efficiency.

[0069] In this scheme, the data processing unit continuously collects the resistance value sequence of the thin-film pressure sensor 16 over a historical period and extracts pressure change features from it to form a pressure change feature vector. Simultaneously, it collects time features, climate features, and wind speed features within the same time period to form an environmental feature vector. Then, these two types of feature vectors are fused and input into a pre-trained classification model. This classification model can be a logistic regression model, support vector machine, random forest, gradient boosting tree, or lightweight neural network, etc. The model output is a loosening probability value between zero and one. When this probability value exceeds a preset probability threshold, the system determines that the clamp 10 is highly likely to loosen and issues an early warning.

[0070] As can be seen from the above embodiments, the present invention changes the traditional working method of relying on manual visual inspection or torque wrench to check the bolts 13 of the scaffolding 20 one by one, and provides an intelligent scaffolding 20 system that can be monitored in real time, accurately located, and has environmental perception and prediction capabilities, which significantly improves the safety of building construction.

[0071] In the description of this application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, 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. Therefore, they should not be construed as limitations on this application.

[0072] Those skilled in the art should understand that the above embodiments are merely for illustrative purposes and are not intended to limit the scope of this application. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of this application.

Claims

1. A loosening early warning device for high-altitude construction scaffolding, the scaffolding comprising clamps and uprights, the uprights being connected by clamps, characterized in that: The early warning device includes: A thin-film pressure sensor is mounted on the connecting lug of the clamp; A passive RFID tag is installed on the clamp, and the passive RFID tag is electrically connected to the membrane pressure sensor. Each passive RFID tag has a unique identifier. An RFID reader is installed on the pole, and the radio frequency signal coverage of the RFID reader covers at least one of the passive RFID tags. The RFID reader transmits radio frequency signals to the passive RFID tag, activating the passive RFID tag and powering it. The passive RFID tag is activated in response to the radio frequency signal, generates a response signal based on the current resistance value of the thin-film pressure sensor, and sends the response signal along with its unique identifier back to the RFID reader. A data processing unit is installed on the scaffold and coupled to the RFID reader / writer. The data processing unit is configured to determine whether the clamp has become loose based on the change information of the resistance value in the response signal and to locate the loose clamp based on the unique identifier. The data processing unit is also configured to provide comprehensive early warning based on multi-dimensional parameters, specifically: The RFID reader receives the response signal and transmits it to the data processing unit. The data processing unit collects the historical resistance value sequence of the thin-film pressure sensor, calculates the pressure value change trend and continuous descent rate; collects the current time information of the scaffold location, distinguishing between daytime and nighttime periods; obtains climate information of the scaffold location through networked meteorology, including temperature, humidity, and rainfall; obtains wind speed data of the scaffold location through a wind speed sensor; and establishes a comprehensive judgment model based on the above data. The comprehensive judgment model is constructed as follows: using the pressure descent rate, the time period weight determined based on the current time information, the climate influence factor, and the wind speed influence factor as input parameters, a loosening risk score is calculated; when the loosening risk score exceeds a preset risk threshold, an early warning signal is issued.

2. The early warning device for loosening of high-altitude construction scaffolding according to claim 1, characterized in that: The early warning device also includes: A clock module, connected to the data processing unit, is used to obtain current time information to distinguish between daytime and nighttime periods; A meteorological data interface, connected to the data processing unit, is configured to obtain real-time climate information of the location of the scaffolding from a meteorological server via wireless communication. The climate information includes temperature, humidity, and rainfall. A wind speed sensor, connected to the data processing unit, is installed on the pole to acquire real-time wind speed data.

3. The early warning device for loosening of high-altitude construction scaffolding according to claim 1, characterized in that: The clamp includes two clamp bodies that can rotate relative to each other. Connecting ears are rotatably mounted on the clamp bodies. The connecting ears form an encircling space for fastening the upright with the clamp bodies by bolts. One end of the bolt is rotatably connected to the clamp body, and the other end is fastened to the connecting ear by a nut. A washer is also installed between the nut and the connecting ear. The thin-film pressure sensor is installed between the gasket and the connecting lug, and the thin-film pressure sensor converts the loose or tight state of the nut into a change in resistance value; The passive RFID tag is installed on the connecting ear or the clamp body; The RFID reader is configured to periodically transmit radio frequency signals to the passive RFID tag to activate the passive RFID tag and receive a response signal returned by the tag containing information about the change in resistance value and the unique identifier.

4. The early warning device for loosening of high-altitude construction scaffolding according to claim 1, characterized in that: The passive RFID tag includes an RFID chip with universal input / output pins, and the thin-film pressure sensor is electrically connected to the universal input / output pins.

5. The early warning device for loosening of high-altitude construction scaffolding according to claim 2, characterized in that: The thin-film pressure sensor is a resistive thin-film pressure sensor.

6. The early warning device for loosening of high-altitude construction scaffolding according to claim 2, characterized in that: The data processing unit is configured to determine whether the clamp has become loose based on the change in resistance value in the response signal, including: S101. The RFID reader transmits a radio frequency signal to the passive RFID tag to activate the passive RFID tag and power it. S102. In response to being activated, the passive RFID tag generates a response signal based on the current resistance value of the thin-film pressure sensor, and sends the response signal along with its unique identifier back to the RFID reader. S103, The RFID reader receives the response signal and transmits it to the data processing unit; S104. The data processing unit parses the response signal, determines whether the clamp has become loose based on the change information of the resistance value, and locates the loose clamp based on the unique identifier. S105. If it is determined that the clamp is loose, the data processing unit issues an alarm and marks the location of the loose clamp.

7. The early warning device for loosening of high-altitude construction scaffolding according to claim 2, characterized in that: The data processing unit is set with a safety pressure threshold. When the pressure value corresponding to the resistance value is lower than the safety pressure threshold, the data processing unit triggers a loosening alarm.

8. The early warning device for loosening of high-altitude construction scaffolding according to claim 6, characterized in that: The comprehensive judgment model is a weighted scoring model, and the loosening risk score is calculated according to the following formula: Risk=w1·F1+w2·F2+w3·F3+w4·F4; Where Risk is the risk score for loosening, w1, w2, w3, and w4 are preset weight coefficients, and w1+w2+w3+w4=1, and F1, F2, F3, and F4 are the normalization functions of each parameter. When the loosening risk score exceeds a preset risk threshold, an early warning signal is issued.

9. The early warning device for loosening of high-altitude construction scaffolding according to claim 7, characterized in that: The data processing unit is preset with a loosening risk threshold. When the loosening risk score exceeds the loosening risk threshold, it determines that there is a loosening risk and issues a warning signal.

Images