Inclination survey safety monitoring method based on multi-parameter fusion

Abstract

The invention discloses a multi-parameter fusion inclinometry safety monitoring method, and relates to the technical field of data processing, and the method comprises the steps: integrating an inclinometer, a displacement sensor, a stress sensor and a temperature sensor to construct a multi-sensing monitoring array, monitoring a road construction process at a fixed point, collecting and obtaining multi-source monitoring data, and carrying out offset compensation and fluctuation trend analysis. Acquiring a multi-source monitoring deviation value and a multi-source monitoring fluctuation curve; performing road inclination trend prediction through an inclination prediction plug-in; and if the inclination prediction angle exceeds the preset scalar, generating an early warning signal and carrying out construction early warning. The technical problems that in the prior art, a traditional inclination survey safety monitoring method is difficult to comprehensively and accurately monitor the road inclination condition, the inclination trend cannot be predicted in real time, and the real-time performance and accuracy of construction monitoring are low are solved, and the purposes that potential risks are recognized in time through multi-parameter fusion monitoring and real-time inclination trend prediction are achieved. And the real-time performance and the accuracy of construction monitoring are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a multi-parameter fusion inclinometer safety monitoring method. Background Art

[0002] In the field of road construction, monitoring the inclination of roads is crucial to ensuring construction quality and safety. Traditional inclination safety monitoring methods usually rely on a single inclination measuring instrument, such as an inclinometer, to measure the inclination angle of the road. However, this method has many limitations, such as limited measurement accuracy, great interference from environmental factors, and inability to fully reflect the overall inclination of the road, which makes it difficult to detect and deal with the inclination risk during construction in a timely manner. Summary of the invention

[0003] The present application provides a multi-parameter fusion inclinometer safety monitoring method, which is used to solve the technical problems in the prior art that traditional inclinometer safety monitoring methods are difficult to comprehensively and accurately monitor road inclination, cannot predict inclination trends in real time, and have low real-time and accuracy in construction monitoring.

[0004] The present application provides a multi-parameter fusion inclination safety monitoring method, the method comprising: integrating an inclinometer, a displacement sensor, a stress sensor and a temperature sensor to construct a multi-sensor monitoring array; using the multi-sensor monitoring array to monitor the road construction process at a fixed point, collecting and acquiring multi-source monitoring data, and performing offset compensation on the multi-source monitoring data according to real-time environmental parameters to obtain multi-source compensation data; performing synchronous alignment, correlation correction, deviation calculation and fluctuation trend analysis on the multi-source compensation data to obtain a multi-source monitoring deviation value and a multi-source monitoring fluctuation curve; predicting a road inclination trend on the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve through a inclination prediction plug-in to obtain a predicted inclination angle, wherein the inclination prediction plug-in is constructed based on an adversarial network; if the predicted inclination angle exceeds a preset scalar, generating a warning signal and performing a construction warning.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages: The multi-parameter fusion inclination safety monitoring method provided in the present application relates to the field of data processing technology. By integrating an inclinometer, a displacement sensor, a stress sensor and a temperature sensor, a multi-sensor monitoring array is constructed to monitor the road construction process in real time and collect data. The acquired multi-source monitoring data is offset compensated and analyzed to obtain a monitoring deviation value and a fluctuation curve. A tilt prediction plug-in based on an adversarial network is used to predict the road tilt trend and obtain a predicted angle. If the predicted angle exceeds a preset scalar, the system will automatically generate an early warning signal. This solves the technical problems in the prior art that the traditional inclination safety monitoring method is difficult to comprehensively and accurately monitor the road tilt situation, cannot predict the tilt trend in real time, and has low real-time and accuracy of construction monitoring. This achieves the technical effect of timely identifying potential risks and improving the real-time and accuracy of construction monitoring through multi-parameter fusion monitoring and real-time tilt trend prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0007] Figure 1 A schematic flow chart of a multi-parameter fusion inclinometer safety monitoring method provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for performing synchronous alignment, correlation correction, deviation calculation and fluctuation trend analysis of multi-source compensation data in the multi-parameter fusion inclinometer safety monitoring method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0008] The present application provides a multi-parameter fusion inclinometer safety monitoring method, which is used to solve the technical problems in the prior art that traditional inclinometer safety monitoring methods are difficult to comprehensively and accurately monitor road inclination, cannot predict inclination trends in real time, and have low real-time and accuracy in construction monitoring.

[0009] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0010] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0011] Embodiment 1, as Figure 1 As shown, the present application provides a multi-parameter fusion inclinometer safety monitoring method, the method comprising: P10: Integrate inclinometers, displacement sensors, stress sensors, and temperature sensors to build a multi-sensor monitoring array.

[0012] Specifically, a highly integrated multi-sensor monitoring array is constructed, which integrates inclinometers, displacement sensors, stress sensors and temperature sensors to achieve comprehensive, real-time and accurate monitoring of the road construction process. Among them, the inclinometer, as a precision instrument specially used to measure the tilt angle, can directly reflect the tilt state of the road structure and is a key indicator for evaluating road stability. The displacement sensor is responsible for monitoring the displacement changes of the road during the construction process, and provides important data for the analysis of road deformation by capturing tiny position movements. The stress sensor goes deep into the road structure to monitor its stress distribution state, which is crucial for understanding the stress condition of the road structure and predicting potential structural damage. The temperature sensor is used to monitor the temperature fluctuations in the construction environment to monitor the impact of temperature on material properties and structural stability.

[0013] In order to achieve the effective integration of these sensors, signal acquisition and processing technology can be used to ensure that each sensor can simultaneously and accurately collect data, and transmit the data to the central processing unit through a high-speed data transmission channel. In addition, a special sensor interface and data processing algorithm can be used to achieve unified formatting and standardized processing of different sensor data, laying the foundation for subsequent data fusion and analysis. Through the construction of the multi-sensor array, not only can the comprehensive monitoring of the construction process be achieved, but also a wealth of basic data is provided for subsequent data processing and analysis, which helps to timely discover potential safety hazards and ensure construction safety and quality.

[0014] P20: Use the multi-sensor monitoring array to monitor the road construction process at a fixed point, collect and obtain multi-source monitoring data, and perform offset compensation on the multi-source monitoring data according to real-time environmental parameters to obtain multi-source compensated data.

[0015] Optionally, in the process of implementing a multi-sensor monitoring array, fixed-point monitoring is a key step to ensure construction safety. By arranging a multi-sensor monitoring array at key locations on the construction site, real-time collection of multi-source monitoring data, including tilt information from the inclinometer, displacement from the displacement sensor, stress state from the stress sensor, and ambient temperature from the temperature sensor, can provide a comprehensive view of the construction status.

[0016] During data acquisition, real-time environmental parameters such as temperature and humidity may affect the sensor readings. Therefore, implementing offset compensation is a necessary step. The purpose of offset compensation is to eliminate the interference of environmental factors on the monitoring data, so as to obtain more accurate and reliable multi-source compensated data. Specifically, the characteristics of each sensor need to be analyzed first to identify its sensitivity to environmental influences. For example, changes in temperature may cause the readings of stress sensors to shift, while displacement sensors may have errors under high humidity conditions.

[0017] On this basis, real-time data processing algorithms can be used to compensate for the offset of the data of each sensor. Usually, these algorithms use linear or nonlinear models, combined with real-time data of environmental parameters, to correct the sensor output. Through this compensation, the errors caused by environmental factors can be effectively eliminated to ensure the accuracy of the data. The multi-source compensation data finally obtained will provide a reliable foundation for subsequent synchronous alignment, correlation correction and trend analysis, ensuring more accurate safety monitoring during the construction process.

[0018] P30: Execute synchronous alignment, correlation correction, deviation calculation and fluctuation trend analysis of the multi-source compensation data to obtain a multi-source monitoring deviation value and a multi-source monitoring fluctuation curve.

[0019] Further, such as Figure 2 As shown, step P30 of the embodiment of the present application also includes: P31: Based on the monitoring point coordinates and monitoring timestamps of the multi-source compensation data, time domain alignment and space alignment are performed to obtain multi-source alignment data; P32: An indicator association topology map is constructed, and the indicator association topology map is used to perform association correction on the multi-source alignment data to obtain multi-source correction data; P33: Deviation calculation and fluctuation trend analysis are performed based on the multi-source correction data to obtain the multi-source monitoring deviation value and multi-source monitoring fluctuation curve.

[0020] It should be understood that synchronous alignment, correlation correction, deviation calculation and fluctuation trend analysis are performed on the obtained multi-source compensation data to extract more accurate and practical monitoring information.

[0021] First, based on the monitoring point coordinates and timestamps of the multi-source compensation data, time domain alignment and spatial alignment are performed. The purpose of time domain alignment is to ensure that the data of all sensors are compared at the same time point, which requires adjusting the readings of each sensor to a unified time format. Spatial alignment is to unify the monitoring data at different locations so that they can be analyzed in the same coordinate system. Through time domain alignment and spatial alignment, multi-source alignment data is obtained, which is consistent in time and space, providing convenience for subsequent analysis.

[0022] Next, we build an indicator correlation topology map. This graphical tool can intuitively display the relationship between various monitoring indicators. For example, the data of the inclinometer may be correlated with the data of the displacement sensor. By analyzing these relationships, we can identify the correlation between the data and potential interference sources. We use this topology map to perform correlation correction on the multi-source alignment data. We use algorithms to adjust the deviations between the data, making the logical relationship between them more reasonable and accurate, eliminating the errors between the data, and obtaining more reliable multi-source correction data.

[0023] Finally, deviation calculation and fluctuation trend analysis are performed based on multi-source correction data. Deviation calculation involves comparing the corrected data with preset standards or historical data to identify abnormal values ​​or deviation levels. Fluctuation trend analysis processes time series data through statistical methods (such as sliding average, standard deviation, etc.) to extract fluctuation curves during the monitoring process. These analysis results include multi-source monitoring deviation values ​​and multi-source monitoring fluctuation curves, which provide a scientific basis for subsequent construction monitoring and early warning.

[0024] Furthermore, step P32 of the embodiment of the present application also includes: P32-1: Through big data collection of sample environmental parameter sets, sample soil property sets and sample indicator association relationship sets, randomly select the first sample indicator association relationship, wherein the first sample indicator association relationship is identified by the first sample environmental parameters and the first sample soil properties; P32-2: respectively perform deviation analysis on the first sample environmental parameters and real-time environmental parameters, the first sample soil properties and construction soil properties, and determine the environmental deviation coefficient and the soil deviation coefficient; P32-3: if the environmental deviation coefficient is less than the environmental deviation scalar and the soil deviation coefficient is less than the soil deviation scalar, add the first sample indicator association relationship to the standard indicator association relationship set; P32-4: perform high-frequency feature extraction on the standard indicator association relationship set, and construct the indicator association topology map based on graph calculation according to the high-frequency indicator association relationship.

[0025] Optionally, in order to further optimize the indicator association correction, big data and machine learning techniques are used to optimize the indicator association topology map construction process and enhance the accuracy and reliability of indicator association.

[0026] First, through the sample environmental parameter set, sample soil property set and sample indicator association relationship set formed by big data collection, the first sample indicator association relationship is randomly selected as the analysis object. This association relationship contains the first sample environmental parameters and the first sample soil properties, providing basic data for subsequent analysis. This random selection process helps to ensure the diversity and representativeness of the samples, thereby improving the universality of the analysis results.

[0027] Next, for the selected first sample, deviation analysis is performed on the first sample environmental parameters and real-time environmental parameters, as well as the first sample soil properties and construction soil properties. This analysis aims to quantify the difference between the sample parameters and the actual monitoring values, and obtain the environmental deviation coefficient and soil deviation coefficient. In this way, the applicability of the sample under the current environmental and soil conditions can be objectively evaluated.

[0028] If the environmental deviation coefficient is less than the preset environmental deviation scalar, and the soil deviation coefficient is less than the preset soil deviation scalar, it is considered that the first sample indicator association relationship has high applicability in the current construction environment, so it is added to the standard indicator association relationship set. This standard set will contain a series of screened indicator association relationships that have high applicability under current construction conditions.

[0029] Finally, high-frequency features are extracted from the standard indicator association relationship set. This process can use graph computing technology to conduct in-depth analysis of the association relationship and identify indicators that show high correlation under different environmental and soil conditions. Based on the extracted high-frequency indicator association relationship, an indicator association topology map is constructed. This topology map can not only clearly show the relationship between the indicators, but also provide an important basis for subsequent data correction and trend analysis.

[0030] Furthermore, step P32-4 of the embodiment of the present application also includes: P32-41: Randomly select a first standard indicator association relationship from the standard indicator association relationship set, wherein the first standard indicator association relationship includes a first tilt mapping relationship, a first displacement mapping relationship, a first stress mapping relationship and a first temperature mapping relationship; P32-42: Perform a similarity comparison between the first tilt mapping relationship and other tilt mapping relationships in the standard indicator association relationship set, and count the amount of data with a similarity greater than a similarity threshold, which is set as a first correlation degree; P32-43: If the first correlation degree is greater than a predetermined correlation scalar, perform a similarity comparison between the first displacement mapping relationship and other displacement mapping relationships in the standard indicator association relationship set to obtain a second correlation degree; P32-44: If the second correlation degree is greater than a predetermined correlation scalar, calculate and obtain a third correlation degree of the first stress mapping relationship, and if the third correlation degree is greater than a predetermined correlation scalar, calculate and obtain a fourth correlation degree of the first temperature mapping relationship; P32-45: If the fourth correlation degree is greater than a predetermined correlation scalar, add the first standard indicator association relationship to a high-frequency association data set, and perform mode extraction on the high-frequency association data set to obtain the high-frequency indicator association relationship.

[0031] In a possible embodiment of the present application, in order to further enhance the accuracy and effectiveness of the indicator association, first, a first standard indicator association relationship is randomly selected from the standard indicator association relationship set. This relationship includes a first tilt mapping relationship, a first displacement mapping relationship, a first stress mapping relationship, and a first temperature mapping relationship, providing basic data for subsequent similarity comparison.

[0032] Next, the selected first tilted mapping relationship is compared with other tilted mapping relationships in the standard indicator association relationship set for similarity. This comparison calculates the similarity of the mapping relationships, counts the amount of data with similarity greater than a preset similarity threshold, and sets it as the first association degree. This process can identify other relationships similar to the current tilted mapping relationship, thereby laying the foundation for further analysis.

[0033] If the first correlation is greater than the predetermined correlation scalar, the selected first displacement mapping relationship is further compared with other displacement mapping relationships in the standard indicator correlation relationship set to obtain a second correlation. This step-by-step similarity comparison method helps to screen out highly correlated data layer by layer and ensure the reliability of the final result.

[0034] If the second correlation is also greater than the predetermined correlation scalar, the third correlation of the first stress mapping relationship is calculated. If the third correlation exceeds the predetermined scalar, the fourth correlation of the first temperature mapping relationship is calculated. In this way, the correlation between different indicators can be comprehensively evaluated to ensure the effectiveness of each mapping relationship.

[0035] Finally, the first temperature mapping relationship is analyzed and the fourth correlation degree is calculated. Only when the correlation degrees of the four mapping relationships are greater than the predetermined correlation scalar, the first standard indicator correlation relationship is considered to be highly representative as a whole, so it is added to the high-frequency correlation data set. The high-frequency correlation data set contains all standard indicator correlation relationships that are highly representative in many aspects. Thereafter, the high-frequency correlation data set is subjected to mode extraction, that is, the indicator correlation relationship with the highest frequency in the data set is found to obtain the high-frequency indicator correlation relationship. These high-frequency relationships will be used to construct the final indicator correlation topology map to ensure the reliability and accuracy of the map.

[0036] Furthermore, step P33 of the embodiment of the present application also includes: P33-1: Based on the standard value of the multi-source indicator, the deviation of the multi-source correction data is calculated respectively to obtain the multi-source monitoring deviation value; P33-2: Multiple historical correction data sets under multiple monitoring nodes in a preset historical time zone are retrieved, and multiple monitoring data sequences are constructed in combination with the multi-source correction data; P33-3: The multiple monitoring data sequences are distributed in a two-dimensional coordinate system, and are serially fitted in chronological order to obtain the multi-source monitoring fluctuation curve.

[0037] Specifically, in order to better understand the dynamic changes in the construction process, a more accurate and comprehensive method can be used to calculate the multi-source monitoring deviation value and draw the multi-source monitoring fluctuation curve. First, based on the standard value of the multi-source indicator, the deviation of the obtained multi-source correction data is calculated to quantify the difference between each correction data and its corresponding standard value, and finally obtain the multi-source monitoring deviation value. This calculation of the deviation value provides basic data for subsequent analysis and helps identify possible abnormal situations or unstable factors.

[0038] Next, multiple historical correction data sets under multiple monitoring nodes in the preset historical time zone are retrieved. These historical data sets contain the monitoring data of each monitoring node in the past period of time, and have high accuracy and reliability after correction. These historical correction data sets are combined with the current multi-source correction data to construct multiple monitoring data series. Each data series contains data points of the same monitoring indicator at different time points, which will be used to draw the fluctuation curve.

[0039] Next, the constructed multiple monitoring data sequences are distributed in a two-dimensional coordinate system and fitted in series according to the time sequence. In the coordinate system, the horizontal axis represents time and the vertical axis represents the value of the monitoring indicator. The data points in each data sequence are fitted in series according to the time sequence. The fitting process can use a smoothing algorithm, such as linear interpolation, polynomial fitting or smooth curve fitting, to ensure that the curve can accurately reflect the changing trend of the data. Through fitting, multi-source monitoring fluctuation curves are obtained, which intuitively show the changes of various monitoring indicators over time during the road construction process.

[0040] Multi-source monitoring fluctuation curves can not only be used to identify abnormal fluctuations and potential risks during the construction process, but also provide important reference information for construction decision-making and risk management. For example, when the fluctuation curve of a monitoring indicator suddenly rises or falls, it may mean that there are problems or potential safety hazards in the construction process, and timely measures need to be taken to intervene and adjust. It can ensure that the deviation degree and change trend of monitoring indicators during road construction can be accurately evaluated, which not only improves the efficiency and accuracy of data processing, but also provides more comprehensive and in-depth information support for subsequent construction monitoring and early warning.

[0041] P40: Predicting the road inclination trend of the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve through a tilt prediction plug-in to obtain a tilt prediction angle, wherein the tilt prediction plug-in is constructed based on an adversarial network.

[0042] Further, a tilt prediction plug-in is constructed, and step P40 of the embodiment of the present application further includes: P41: Construct a tilt prediction plug-in based on an adversarial network, wherein the tilt prediction plug-in includes an angle generator and an angle discriminator, the input data of the angle generator is the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve, and the output data is the predicted tilt angle, the input data of the angle discriminator is the predicted tilt angle output by the angle generator, and the output data is the angle credibility; P42: Call the construction monitoring log to collect sample multi-source monitoring deviation values, sample multi-source monitoring fluctuation curves and sample tilt angles; P43: Use the sample multi-source monitoring deviation values, sample multi-source monitoring fluctuation curves and sample tilt angles to iteratively train the angle generator and the angle discriminator until the expected convergence conditions are met to obtain the tilt prediction plug-in.

[0043] It should be understood that by constructing a tilt prediction plug-in to predict the road tilt trend of the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve, the tilt prediction angle is obtained.

[0044] First, a tilt prediction plug-in is built based on the adversarial network. The plug-in consists of two main parts: the angle generator and the angle discriminator. The angle generator is responsible for receiving multi-source monitoring deviation values ​​and multi-source monitoring fluctuation curves as input data, which contain real-time information and historical change trends of various monitoring indicators during road construction. Through a complex calculation process, the angle generator can output a predicted tilt angle, which reflects the possible future tilt trend of the road.

[0045] At the same time, the angle discriminator receives the predicted tilt angle output by the angle generator as input data, evaluates it, and outputs an angle credibility. This credibility reflects the accuracy and reliability of the predicted tilt angle. The existence of the angle discriminator allows the angle generator to be continuously challenged and improved during the training process, thereby improving the accuracy of the prediction.

[0046] In order to build the tilt prediction plug-in, a large amount of sample data needs to be collected first. These sample data include sample multi-source monitoring deviation values, sample multi-source monitoring fluctuation curves and sample tilt angles, which are collected and sorted from past road construction projects through construction monitoring logs. These sample data will serve as training data sets to provide a basis for the learning of adversarial networks.

[0047] Next, the angle generator and angle discriminator are iteratively trained using the collected sample multi-source monitoring deviation values, sample multi-source monitoring fluctuation curves, and sample tilt angles. In this process, the model will continuously adjust its parameters to minimize the gap between the predicted tilt angle and the actual tilt angle until the expected convergence conditions are met, and the tilt prediction plug-in is obtained. This iterative training method is the core of the adversarial network and can effectively improve the prediction ability of the model.

[0048] The application of the generated tilt prediction plug-in will greatly improve the efficiency and accuracy of road construction monitoring. It can predict the tilt trend of the road in real time and provide important reference information for construction decision-making and risk management. At the same time, because the plug-in is built based on the adversarial network, it has strong adaptive ability and robustness, and can adapt to changes in different road construction environments and conditions.

[0049] Further, to obtain the predicted tilt angle, step P40 in the embodiment of the present application further includes: P44: Acquire multiple actual tilt angles and multiple predicted tilt angles within a preset time period, perform prediction deviation calculations on the multiple actual tilt angles and the multiple predicted tilt angles, and determine multiple prediction angle deviations; P45: Eliminate abnormalities on the multiple predicted angle deviations, obtain multiple standard prediction deviations, perform mean calculations on the multiple standard prediction deviations, determine the mean of the prediction deviations, and compensate the predicted tilt angle according to the mean of the prediction deviations.

[0050] Optionally, in order to obtain a more accurate tilt prediction angle, a prediction deviation calculation and compensation mechanism is introduced to ensure the accuracy and reliability of the tilt prediction angle.

[0051] First, multiple actual tilt angles and multiple predicted tilt angles within a preset time period are obtained. This preset time period is usually determined based on the specific conditions of road construction and monitoring requirements, in order to ensure that there is sufficient data for accurate deviation analysis. Next. By performing prediction deviation calculations on the multiple actual tilt angles and multiple predicted tilt angles, the difference between the predicted angle and the actual angle is identified, thereby determining multiple predicted angle deviations. This calculation process usually uses a simple difference method, such as the formula: Deviation = actual tilt angle - predicted tilt angle. This deviation value provides the necessary data support for subsequent abnormality elimination and compensation.

[0052] Furthermore, the obtained multiple prediction angle deviations are subjected to abnormal elimination to exclude extreme values ​​caused by unexpected events or data noise. These abnormal values ​​are identified and eliminated by statistical methods (such as Z-score or IQR method) to obtain multiple standard prediction deviations. After that, the mean of these standard prediction deviations is calculated to obtain the mean of the prediction deviations. This mean reflects the overall performance of the prediction model over the entire preset time period.

[0053] Finally, the predicted tilt angle is compensated according to the calculated mean value of the predicted deviation. The purpose of compensation is to adjust the predicted angle to make it closer to the actual angle, thereby improving the accuracy of the prediction. The specific method of compensation can be a simple addition and subtraction operation, or an adjustment based on a complex algorithm. Through compensation, a corrected predicted tilt angle can be obtained, which is closer to the actual tilt of the road, providing more reliable information support for construction decision-making and risk management.

[0054] P50: If the predicted inclination angle exceeds a preset scalar, a warning signal is generated and a construction warning is issued.

[0055] Specifically, a safety warning mechanism is implemented to ensure that the tilt situation during road construction can be monitored in a timely and effective manner. First, when the calculated predicted tilt angle is judged to exceed the preset scalar, the system will immediately identify this abnormal state. This preset scalar is usually a safety threshold set based on historical data and engineering standards, which is used to evaluate the safety and stability of the road construction process. A predicted tilt angle exceeding this scalar indicates that there may be potential risks in the construction environment, such as unstable foundation, soil loss or equipment failure, which requires high attention.

[0056] Next, based on the situation of exceeding the scalar value, an early warning signal is generated. This signal may be conveyed to relevant personnel in a variety of ways, such as alarm prompts on the real-time monitoring platform, SMS notifications, or email reminders, to ensure that managers at the construction site can quickly obtain information and respond. The generation of early warning signals is not only a timely response to the construction status, but also an effective protection measure for the safety of construction personnel.

[0057] On this basis, a construction warning is carried out. The construction warning is a more detailed report or notification, which contains the specific value of the predicted tilt angle, the degree of exceeding the preset scalar, possible safety risks, and recommended countermeasures. It helps the construction team quickly understand the current tilt situation, assess the risks, and make corresponding decisions, such as suspending construction, strengthening monitoring, and taking reinforcement measures. Improving the safety of construction and providing timely decision-making support for the construction team will help reduce potential safety risks and losses.

[0058] In summary, the embodiments of the present application have at least the following technical effects: This application integrates inclinometers, displacement sensors, stress sensors, and temperature sensors to build a multi-sensor monitoring array to monitor the road construction process in real time and collect data. Offset compensation and analysis are performed on the acquired multi-source monitoring data to obtain monitoring deviation values ​​and fluctuation curves. The tilt prediction plug-in based on the adversarial network is used to predict the road tilt trend and obtain the predicted angle. If the predicted angle exceeds the preset scalar, the system will automatically generate a warning signal to ensure construction safety.

[0059] The technical effect of timely identifying potential risks and improving the real-time and accuracy of construction monitoring has been achieved through multi-parameter fusion monitoring and real-time tilt trend prediction.

[0060] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0062] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A multi-parameter fusion inclinometer safety monitoring method, characterized in that: Methods include: Integrate inclinometers, displacement sensors, stress sensors, and temperature sensors to build a multi-sensor monitoring array; The multi-sensor monitoring array is used to monitor the road construction process at a fixed point, collect and obtain multi-source monitoring data, and offset compensate the multi-source monitoring data according to real-time environmental parameters to obtain multi-source compensated data; Performing synchronous alignment, correlation correction, deviation calculation and fluctuation trend analysis of the multi-source compensation data to obtain a multi-source monitoring deviation value and a multi-source monitoring fluctuation curve; Predicting the road inclination trend on the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve through an inclination prediction plug-in to obtain an inclination prediction angle, wherein the inclination prediction plug-in is constructed based on an adversarial network; If the predicted inclination angle exceeds a preset scalar, a warning signal is generated and a construction warning is issued.

2. The multi-parameter fusion inclinometer safety monitoring method according to claim 1 is characterized in that: Performing offset compensation on the multi-source monitoring data according to the real-time environmental parameters to obtain multi-source compensated data includes: Obtaining the properties of the construction soil, wherein the properties of the construction soil include soil physical properties, soil chemical properties and soil stress properties; Taking the soil physical properties, soil chemical properties and soil stress properties as search constraints and guided by multi-source monitoring indicators, a sample environmental parameter set and a sample multi-source monitoring data deviation set are collected, wherein the environmental parameters at least include ambient temperature and ambient humidity; Using the sample environmental parameter set and the sample multi-source monitoring data deviation set to supervise the training of the environmental offset compensator until a convergence state is reached; The real-time environmental parameters are analyzed by the environmental offset compensator, and the multi-source monitoring data deviation is output to perform offset compensation on the multi-source monitoring data to obtain the multi-source compensated data.

3. The multi-parameter fusion inclinometer safety monitoring method according to claim 2 is characterized in that: Performing synchronous alignment, correlation correction, deviation calculation and fluctuation trend analysis of the multi-source compensation data to obtain a multi-source monitoring deviation value and a multi-source monitoring fluctuation curve includes: Based on the monitoring point coordinates and monitoring timestamps of the multi-source compensation data, performing time domain alignment and space alignment to obtain multi-source alignment data; Constructing an indicator association topology map, and using the indicator association topology map to perform association correction on the multi-source alignment data to obtain multi-source correction data; Deviation calculation and fluctuation trend analysis are performed based on the multi-source correction data to obtain the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve.

4. The multi-parameter fusion inclinometer safety monitoring method according to claim 3 is characterized in that: Construct an indicator association topology diagram, including: Collecting a sample environmental parameter set, a sample soil property set and a sample indicator association relationship set through big data, and randomly selecting a first sample indicator association relationship, wherein the first sample indicator association relationship is identified by a first sample environmental parameter and a first sample soil property; Performing deviation analysis on the first sample environmental parameters and real-time environmental parameters, the first sample soil properties and construction soil properties respectively, and determining an environmental deviation coefficient and a soil deviation coefficient; If the environmental deviation coefficient is less than the environmental deviation scalar and the soil deviation coefficient is less than the soil deviation scalar, adding the first sample indicator association relationship to the standard indicator association relationship set; The high-frequency features of the standard indicator association relationship set are extracted, and the indicator association topology graph is constructed based on the high-frequency indicator association relationship based on graph calculation.

5. The multi-parameter fusion inclinometer safety monitoring method according to claim 4 is characterized in that: The high-frequency feature extraction is performed on the standard indicator association relationship set, including: Randomly selecting a first standard indicator association relationship from the standard indicator association relationship set, wherein the first standard indicator association relationship includes a first tilt mapping relationship, a first displacement mapping relationship, a first stress mapping relationship, and a first temperature mapping relationship; Performing a similarity comparison between the first tilted mapping relationship and other tilted mapping relationships in the standard indicator association relationship set, and counting the amount of data whose similarity is greater than a similarity threshold, and setting it as a first association degree; If the first correlation degree is greater than a predetermined correlation scalar, performing a similarity comparison between the first displacement mapping relationship and other displacement mapping relationships in the standard indicator correlation relationship set to obtain a second correlation degree; If the second correlation degree is greater than a predetermined correlation scalar, a third correlation degree of the first stress mapping relationship is calculated and obtained; if the third correlation degree is greater than a predetermined correlation scalar, a fourth correlation degree of the first temperature mapping relationship is calculated and obtained; If the fourth correlation degree is greater than a predetermined correlation scalar, the first standard indicator correlation relationship is added to the high-frequency correlation data set, and mode extraction is performed on the high-frequency correlation data set to obtain the high-frequency indicator correlation relationship.

6. The multi-parameter fusion inclinometer safety monitoring method according to claim 3 is characterized in that: Deviation calculation and fluctuation trend analysis are performed according to the multi-source correction data to obtain the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve, including: Based on the multi-source indicator standard value, respectively calculate the deviation of the multi-source correction data to obtain the multi-source monitoring deviation value; Retrieving multiple historical calibration data sets under multiple monitoring nodes in a preset historical time zone, and constructing multiple monitoring data sequences in combination with the multi-source calibration data; The multiple monitoring data sequences are distributed in a two-dimensional coordinate system, and are serially fitted in chronological order to obtain the multi-source monitoring fluctuation curve.

7. The multi-parameter fusion inclinometer safety monitoring method according to claim 1, characterized in that: Build the tilt prediction plugin, including: A tilt prediction plug-in is constructed based on an adversarial network, wherein the tilt prediction plug-in includes an angle generator and an angle discriminator. The input data of the angle generator is the multi-source monitoring deviation value and the multi-source monitoring fluctuation curve, and the output data is the predicted tilt angle. The input data of the angle discriminator is the predicted tilt angle output by the angle generator, and the output data is the angle credibility. Call the construction monitoring log to collect sample multi-source monitoring deviation values, sample multi-source monitoring fluctuation curves and sample tilt angles; The angle generator and the angle discriminator are iteratively trained using the sample multi-source monitoring deviation value, the sample multi-source monitoring fluctuation curve and the sample tilt angle until the expected convergence conditions are met, thereby obtaining the tilt prediction plug-in.

8. The multi-parameter fusion inclinometer safety monitoring method according to claim 7 is characterized in that: Get the predicted tilt angle, including: Acquire a plurality of actual tilt angles and a plurality of predicted tilt angles within a preset time period, perform prediction deviation calculations on the plurality of actual tilt angles and the plurality of predicted tilt angles, and determine a plurality of predicted angle deviations; Abnormalities are eliminated for the multiple predicted angle deviations, multiple standard predicted deviations are obtained, and the mean of the multiple standard predicted deviations is calculated to determine the mean of the predicted deviations, and the tilt prediction angle is compensated according to the mean of the predicted deviations.

Images