A highway settlement intelligent monitoring method

Abstract

The application discloses a kind of highway settlement intelligent monitoring methods, belong to highway engineering technical field.The application is along the sensor of the road section to be monitored, and the data collected by sensor is transmitted to Internet of Things platform to be preprocessed and settlement prediction, and Internet of Things platform is deployed based on the combination model of two different roadbed settlement prediction models, and the preprocessed monitoring data is predicted by combination model, and settlement prediction data is obtained, to realize highway settlement intelligent monitoring.The application can also alarm the abnormal state and data in the settlement process based on the obtained settlement prediction data as needed.The application applies Internet of Things, artificial intelligence and other technologies in the project engineering, which ensures the safe operation of the project work to some extent and improves the safety prevention and control capability of the project.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of highway engineering, and particularly relates to a highway settlement intelligent monitoring method. BACKGROUND

[0002] In soft foundation highway engineering, the post-construction settlement and uneven settlement of the roadbed are the main diseases of the roadbed, so it is very important to monitor the settlement.

[0003] The main problems of the current settlement monitoring of the soft foundation highway engineering are as follows: 1. The informationization degree of the construction site monitoring means is not high, the monitoring method is mainly manual monitoring, and there are disadvantages such as limited application range, low monitoring frequency, great influence of human factors and environment, and high labor cost. 2. The effective analysis of the on-site monitoring data and the post-construction monitoring are not enough, and the monitoring data cannot be used for effective dynamic design. Therefore, how to monitor the settlement of the highway in real time, accurately and intelligently is the core of the problem. SUMMARY

[0004] The purpose of the present application is to overcome the defects of the prior art, and to provide a highway settlement intelligent monitoring method. By using the technology of Internet of Things, big data, artificial intelligence and the like, a settlement intelligent monitoring system is developed, which has the functions of whole-process information collection, data analysis, real-time monitoring, abnormal alarm and the like, and realizes real-time, accurate and intelligent settlement monitoring of the highway.

[0005] In order to achieve the above technical purpose, the present application provides a highway settlement intelligent monitoring method, which comprises the following steps:

[0006] Step S1, sensors are arranged along the to-be-monitored highway section, and the sensors are in communication connection with respective data collectors through RS485 cables;

[0007] Step S2, the data collectors are in wireless communication connection with DTUs; and the monitoring data is wirelessly transmitted to the DTUs;

[0008] Step S3, the DTUs transmit the data to an Internet of Things platform through wireless communication, the Internet of Things platform pre-processes the data and performs visualization and early warning configuration;

[0009] Step S4, a Poisson curve settlement prediction model and a settlement prediction model based on the Markov chain Monte Carlo Bayesian updating method are established; and the two kinds of settlement prediction models are combined to obtain a combined prediction model;

[0010] Step S5, the combined prediction model is embedded into the Internet of Things platform, the monitoring data pre-processed in step S3 is predicted to obtain settlement prediction data, and the highway settlement intelligent monitoring is realized.

[0011] Compared with the prior art, the present application has at least the following advantages:

[0012] (1) Compared with conventional monitoring means, the present application uses Internet of Things technology to monitor highway settlement, has higher informatization level, wider applicable range, higher monitoring frequency, can collect data throughout the whole process of highway engineering, can timely and effectively feedback abnormal data and give alarm, and can effectively ensure safe operation of engineering project work.

[0013] (2) The present application adds an artificial intelligence model, that is, a Poisson curve settlement prediction model is combined with a settlement prediction model based on Markov chain Monte Carlo Bayesian updating method to realize prediction of highway settlement. A single model can only consider the error of a part of the settlement data sequence, and the overall error is not considered enough, so the two kinds of subgrade settlement prediction models are combined to improve the prediction accuracy. Thus, preventive measures can be taken in advance according to the prediction result, potential risks can be effectively avoided, and the safety prevention and control capability is improved.

[0014] (3) The Internet of Things platform of the present application is deployed based on an open source platform, which reduces the development and maintenance cost, is convenient for application in various scenes, and improves the efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 The flowchart of the highway settlement intelligent monitoring method described in the present application example.

[0016] Figure 2 The schematic diagram of the calculation result of the hierarchical summation method described in the present application embodiment.

[0017] Figure 3 The schematic diagram of the data preprocessing process described in the present application embodiment.

[0018] Figure 4 The schematic diagram of the data visualization described in the present application embodiment.

[0019] Figure 5 The step diagram of the highway settlement prediction model based on Markov chain Monte Carlo Bayesian updating method described in the present application embodiment.

[0020] Figure 6 The flowchart of the embedded artificial intelligence data analysis module described in the present application embodiment. DETAILED DESCRIPTION

[0021] The present application will be described in detail below in conjunction with the drawings and specific embodiments. Note that the following description of the embodiments is only substantially illustrative, the present application is not intended to limit the applicable objects or uses thereof, and the present application is not limited to the following embodiments.

[0022] REFERENCE Figure 1As shown, the embodiment of the present application provides a highway settlement intelligent monitoring method step, which comprises:

[0023] Step S1, sensors are arranged along the highway section to be monitored, and the sensors are in communication connection with respective data collectors through RS485 cable;

[0024] The present application is based on a certain highway section reconstruction and expansion project, and the settlement of the highway section is intelligently monitored. The test section of the project is 56 m long, the embankment base is 20 m wide, the embankment height is 3.6 m, and the slope gradient is 1:1.5. The soil layer of the highway section is in turn clay layer, silt layer, clay layer and gravel-containing silty clay layer, with a total thickness of 28 m. Cement mixing piles with a diameter of 0.7 m are arranged every 1.2 m below the embankment base. The construction sequence of the highway section is as follows: first, cement mixing pile construction is carried out, and the construction time is 2 days. After the construction is completed, it is stopped for 30 days. Then embankment filling is carried out, and the embankment is filled in two layers. The lower layer is a crushed stone cushion layer, and the filling height is 0.5 m. The filling time is 15 days, and geogrid is laid at a height of 0.3 m above the base. The upper layer is filled with soil, and the filling height is 3.1 m. The filling time is 95 days. After filling is completed, preloading is carried out for 8 months.

[0025] The deepest soft soil layer of the typical cross section of the project is selected as the research object, and the range of the sensor is determined. The monitoring depth of the multipoint displacement meter is determined according to the depth of the soft soil layer determined by the drilling sampling investigation result. According to the depth of the deepest soft soil layer of the cross section, the monitoring depth of the multipoint displacement meter is selected as 28 m below the top surface of the roadbed, and 4 displacement monitoring points are uniformly arranged along the depth direction to measure the layered compression; the monitoring range of the static leveling instrument is determined according to the calculation result of the layered summation method. It is calculated that the settlement is 0.34 m, and the calculation result is shown in Figure 2 Therefore, the range of the static leveling instrument is selected as the maximum of 0.4 m.

[0026] According to the length of the test section of the project, one settlement monitoring profile is arranged every 14 m along the highway. In each settlement monitoring profile, one multipoint displacement meter and one static leveling instrument are buried on both sides of the roadbed and the roadbed center line (3 multipoint displacement meters and 3 static leveling instruments in each profile), and they are in communication connection with the data collector through RS485 cable. In addition, 4 settlement plates are arranged on both sides of the soil shoulder of the same cross section, and the total road surface settlement is manually observed once a month.

[0027] Step S2, the data collector is in wireless communication connection with the DTU (Data Transfer unit, wireless terminal device), and the monitoring data is wirelessly transmitted to the DTU.

[0028] The data acquisition system consists of a bus and branch lines. A monitoring branch line is laid at each monitoring cross section of the roadbed and protected with PE pipe. Each sensor is connected to its own data collector via the branch line. The serial port parameters are then set in the DTU configuration software based on the sensor type: baud rate 9600, parity bit NONE, and data bits 8. After these settings are set, the data collector is successfully connected to the DTU via the bus through each branch line.

[0029] Step S3: The DTU transmits the data to the IoT platform via wireless communication, and the IoT platform pre-processes and visualizes the data;

[0030] S301, the IoT platform deploys a monitoring solution based on the ThingsBoard platform and receives raw monitoring data through the MQTT protocol;

[0031] Deploy a ThingsBoard-based IoT platform locally. Import the DTU device information into the IoT platform, select the MQTT protocol for connection, and set the device credentials: client ID, username, and password. Then, select the MQTT protocol connection method in the MQTT node in Node-RED, enter the parameters: Server: 127.0.0.1, Port: 8080, enter the device credentials, and select the data upload format: JSON with IMEI. After completing the settings, the MQTT node will successfully establish a publish-subscribe connection with the IoT platform via the MQTT protocol, upload the monitoring data to the IoT platform, and store it in the database as raw monitoring data.

[0032] In S302, Node-RED is used as an edge computing node to pre-process the original monitoring data by removing abnormal data and processing noise data. The process is as follows: Figure 3 First, establish a TCP connection with the IoT platform database through the TCP node, set the address and port, and select the TCP connection method. Once the connection is successful, the raw monitoring data stored in the database is transferred to the TCP node.

[0033] Then write the abnormal data removal code in the Function1 node: Set the settlement value to be independent x1, x2..., x n , calculate its arithmetic mean x and residual error v i =x i -x(i=1,2,...,n), and calculate the standard deviation σ according to Bessel's formula. If a certain sedimentation value x b The residual error v b (1<=b<=n), satisfying the following formula |v b |=|x b -x|>3σ, then x b It is a bad value containing gross error and should be eliminated.

[0034] Then connect the TCP node with the Function1 node to get the monitoring data with abnormal data removed. Then access the Function2 node and write the noise data processing code in the Function2 node: moving average = (data point 1 + data point 2 + data point 3 +... + data point n) / n, where n is the time window size of the moving average, indicating that the first n data points are taken for average calculation. Starting from the first data point of the data sequence, the moving average is calculated in turn. For the first moving average, the first n data points are used for calculation; for the subsequent moving averages, one data point is slid back each time and the average is recalculated. The calculated moving average is recorded as the smoothed data sequence.

[0035] Finally, the preprocessed monitoring data is connected to the Internet of Things platform through the MQTT node and stored as effective monitoring data in the database.

[0036] S303, visualize and configure the preprocessed effective monitoring data, as shown in Figure 4 The duration curve of the selected time period is generated according to the change of the monitoring point data, facilitating backtracking and analysis. At the same time, the differential settlement curve is generated according to the settlement of different measuring points. The differential settlement at different times can be displayed in the same page, facilitating comparison and analysis, and being updated in real time. The warning module is set up to set the range of settlement. According to the characteristics of the project, three alarm thresholds are set. The first alarm is the most serious, the second alarm is the second, and the third alarm is the least serious. The second alarm and the third alarm are displayed in the automatically pushed daily report, weekly report and monthly report. Once the first alarm is reached, not only is it displayed in the report, but also an alarm information is pushed to the user in time. The platform can export EXCEL and WORD reports. The report has functions such as measuring point settlement statistics, curve, alarm level and alarm number statistics, and abnormal data type statistics. The data is comprehensively processed and analyzed to better guide the construction.

[0037] Step S4, two kinds of roadbed settlement prediction artificial intelligence models are established, and the two kinds of roadbed prediction artificial intelligence models are combined.

[0038] The Poisson curve change trend is more in line with the law of roadbed settlement in the later period, so it has good accuracy in predicting roadbed settlement. First, a Poisson curve settlement prediction model is established, and the expression of the Poisson curve settlement prediction model is:

[0039]

[0040] In the formula: S tis the settlement of the subgrade at a certain predicted position at time t; k, a, b are undetermined coefficients (where a > 0, b > 0).

[0041] The steps of predicting the subgrade settlement using the Poisson curve settlement prediction model are as follows:

[0042] The sensor monitoring data in a set time range before the current time is obtained, and the settlement measured values with equal monitoring time intervals are taken therefrom, so that the subgrade settlement measured values corresponding to each time point t = 1, 2, 3, …, n are S1, S2, S3, …, Sn respectively. n ;

[0043] The time range is equally divided into three monitoring periods, and i = n / 3; then there are t = 1, 2, 3, …, i in the first monitoring period, t = i + 1, i + 2, i + 3, …, 2i in the second monitoring period, and t = 2i + 1, 2i + 2, 2i + 3, …, 3i in the third monitoring period; the subgrade settlement characteristic parameters corresponding to the three monitoring periods are y1, y2, y3.

[0044]

[0045]

[0046] The undetermined coefficients in the expression of the Poisson curve settlement prediction model are updated according to the subgrade settlement characteristic parameters, wherein:

[0047]

[0048] The three coefficients a, b, and k obtained are substituted into the Poisson curve settlement prediction model, and the time t tends to infinity to obtain the final subgrade settlement S.

[0049] After the Poisson curve model is established, a highway settlement prediction model based on the Bayesian updating method of Markov chain Monte Carlo is established, and the steps are as shown in Figure 5 The finite element software is used to establish a finite element model. The soil parameters are regarded as random variables, 2000 groups of parameter samples are generated by Latin hypercube sampling and input into the finite element model, and the corresponding settlement calculation results are obtained. Based on the parameter samples, a subgrade settlement data set is constructed, and then a random forest method is used to construct a proxy model. The subgrade settlement data set constructed above is randomly divided into two parts, of which 80% is used as the training set and the remaining 20% is used as the test set. The training set is used to train the proxy model, and then the soil parameters in the test set are input into the proxy model and the subgrade settlement calculation results are output. The proxy model calculation results are compared with the original finite element model calculation results, and the fitting effect meets the requirements, and the proxy model is used to replace the original finite element model.

[0050] According to the properties of soft soil, a soil parameter dataset is generated; the soil parameter dataset is input into the proxy model to calculate the settlement, and according to the influence degree of each parameter on the soft soil subgrade settlement value, the high sensitivity soil parameters are determined as random variables, and the other soil parameters are regarded as constants.

[0051] According to the geological exploration report, existing literature data and engineering experience, the prior information of the high sensitivity soil parameters is obtained to determine the prior distribution thereof. Based on the proxy model, the Markov Chain Monte Carlo Bayesian updating algorithm (MCMC) is adopted to obtain the posterior samples of the soil parameters based on the monitoring data; the posterior samples of the updated soil parameters are substituted into the proxy model to obtain the soft foundation settlement prediction value.

[0052] After the two kinds of settlement prediction models are established, the error reciprocal method is used to determine the weights of different models to form a combined prediction model:

[0053] Let y represent the actual settlement monitoring value, the prediction value of the i-th model at time t, and y(t) represent the prediction value of the combined model at time t;

[0054] The absolute error e i (t) of the i-th model at time t, the sum d i (t) of the absolute error reciprocals, the weight wi of the i-th model at time t, and then the prediction value y(t) of the combined model at time t is obtained, and the calculation formula is as follows:

[0055]

[0056] Step S5 embeds the combined prediction model into the Internet of Things platform, predicts the monitoring data preprocessed in step S3 to obtain settlement prediction data, and realizes intelligent monitoring of highway settlement.

[0057] The above artificial intelligence model is embedded with the help of Node-RED, and the process is shown in FIG. 6. First, the preprocessed effective monitoring data in the database is called through the TCP node, the code of the developed artificial intelligence analysis module is written into the Function node, the monitoring data is connected to the Function node, the prediction data (predicted settlement) is generated after the calculation and processing of the Function node code, and then the prediction data is connected to the Internet of Things platform through the MQTT node. According to the actual situation of the project, the safety range of the prediction data is set, and if it exceeds the safety range, a warning is generated, and the warning message is fed back to the remote terminal of the design, establishment, construction, and owner units through SMS and other methods. The management personnel judge whether there is potential risk, and adjust the construction method according to it.

[0058] Through the above steps, a highway monitoring intelligent system is established, and a schematic diagram of the monitoring intelligent system is shown in FIG. Figure 6 .

[0059] While the application has been described in terms of several preferred embodiments, the skilled person will appreciate that other embodiments can be created within the scope and spirit of the application as described above. Additionally, it should be noted that the language used in the specification has been principally directed to enabling an understanding of the present application rather than to exhaustively describe all possible combinations meant to be within the scope of the application. Accordingly, many modifications and variations will be apparent to practitioners skilled in the art. The scope of the application is intended to be illustrative rather than restrictive. The actual scope of the application is defined by the appended claims.

[0060] The above merely represents embodiments of the application, and it is emphasized that those skilled in the art can still make some improvements and adjustments on the basis of the principles of the application. These improvements and adjustments should also be considered as within the scope of protection of the application.

Claims

1. A method for intelligent monitoring of highway settlement, characterized in that: The steps include: Step S1: Deploy sensors along the road section to be monitored, wherein the sensors are connected to respective data collectors via RS485 cables; Step S2: The data collector is connected to the DTU via wireless communication; the monitoring data is wirelessly transmitted to the DTU; Step S3: The DTU transmits the data to the IoT platform via wireless communication. The IoT platform pre-processes the data and performs visualization and early warning configuration. Step S4, establishing a Poisson curve settlement prediction model and a settlement prediction model based on the Bayesian update method of Markov Chain Monte Carlo; combining the two settlement prediction models to obtain a combined prediction model; The expression of the Poisson curve settlement prediction model is: Where: S t is the settlement of the roadbed at a certain predicted position at time t; k, a, b are unknown coefficients; The steps for predicting roadbed settlement using the Poisson curve settlement prediction model are as follows: Obtain the sensor monitoring data within the set time range before the current moment, and take the settlement measured values ​​with equal monitoring time intervals. Let the roadbed settlement measured values ​​corresponding to each time point t = 1, 2, 3, ..., n be S1, S2, S3, ..., S n ; Divide the time range into three monitoring periods, let i = n / 3; then, the time points in the first monitoring period are t = 1, 2, 3, ..., i, the time points in the second monitoring period are t = i+1, i+2, i+3, ..., 2i, and the time points in the third monitoring period are t = 2i+1, 2i+2, 2i+3, ..., 3i; the roadbed settlement characteristic parameters corresponding to the three monitoring time periods are y1, y2, y3; Update the undetermined coefficients in the Poisson curve settlement prediction model expression according to the roadbed settlement characteristic parameters, where: Substitute the three coefficients a, b, and k obtained into the Poisson curve settlement prediction model, let time t tend to infinity, and obtain the final roadbed settlement S; The establishment of a settlement prediction model based on the Bayesian update method of Markov Chain Monte Carlo includes: Finite element software was used to establish a finite element model based on highway section survey data. Soil parameters were treated as random variables, and multiple sets of parameter samples were generated through Latin hypercube sampling. These samples were then input into the finite element model to calculate the corresponding settlement results. Based on the parameter samples, a roadbed settlement dataset was constructed. A proxy model was then constructed using the random forest method. The constructed roadbed settlement dataset was randomly divided into a training set and a test set. The proxy model was trained using the training set. The soil parameters in the test set were then input into the proxy model, and the subgrade settlement calculation results were output. The proxy model results were compared with those of the original finite element model. If the comparison results met the requirements, the proxy model was used to replace the original finite element model. Based on the properties of soft soil, a soil parameter data set is generated. This data set is input into the proxy model to calculate settlement. Based on the degree of influence of each parameter on the settlement of the soft soil roadbed, the highly sensitive soil parameters are determined and treated as random variables, while other soil parameters are treated as constants. In step S5, the combined prediction model is embedded in the Internet of Things platform to predict the monitoring data preprocessed in step S3 to obtain settlement prediction data, thereby realizing intelligent monitoring of highway settlement.

2. A method for intelligent monitoring of road subsidence according to claim 1, characterized in that: The sensors in step S1 are a static level and a multi-point displacement meter. In step S1, the settlement of the road section to be monitored is estimated based on geological survey reports, literature and engineering experience, and the range and accuracy of the sensor are selected based on the estimation results.

3. The method for intelligent monitoring of road subsidence according to claim 1, characterized in that: In step S1, a settlement monitoring profile is selected at a set distance along the highway section to be monitored, and a multi-point displacement meter and a static level are buried on both sides of the roadbed and in the center line of the roadbed of each settlement monitoring profile. The static level is used to obtain the vertical displacement value at the roadbed surface, and the multi-point displacement meter is arranged below the static level to obtain the settlement value at different depths.

4. The method for intelligent monitoring of road subsidence according to claim 1, characterized in that: The DTU in step S2 is selected as a 4G DTU with low bandwidth, low power consumption, long distance, and support for multiple connections, and the DTU parameters are configured according to the sensor type to achieve the reception of monitoring data.

5. The method for intelligent monitoring of road subsidence according to claim 1, characterized in that: Step S3 specifically includes the following steps: S301, the IoT platform deploys a monitoring solution based on the ThingsBoard platform and receives raw monitoring data through the MQTT protocol; S302, using Node-RED as an edge computing node, pre-processing the original monitoring data to remove abnormal data and process noise data; S303: Visualize the effective monitoring data after preprocessing and configure early warning.

6. The method for intelligent monitoring of road subsidence according to claim 5, characterized in that: The abnormal data removal is as follows: the mean μ and standard deviation σ of the single-day monitoring data are calculated. If the deviation of a single measurement data exceeds 3σ, the measurement data is considered abnormal data and is removed. Then, the mean, standard deviation and single measurement data deviation of the remaining data are calculated, and the data removal process is repeated until the deviation of each data is less than 3 times the standard deviation. The noise data processing adopts a data smoothing method, that is, for the data after the abnormal data is eliminated, the mean of n items of data in the window is calculated respectively by moving the time window, where n is the size of the moving time window, and the mean is used to replace the original value to eliminate random fluctuations.

7. The method for intelligent monitoring of road subsidence according to claim 1, characterized in that: In step S4, a highway settlement prediction model based on the Bayesian update method of Markov Chain Monte Carlo is established, which further includes: Prior information of highly sensitive soil parameters is obtained and their prior distribution is determined. Based on the surrogate model and combined with the actual settlement monitoring data of the sensor, the Markov Chain Monte Carlo Bayesian update algorithm is used to obtain the posterior samples of the soil parameters. The posterior samples of the updated soil parameters are substituted into the surrogate model to obtain the predicted value of soft foundation settlement.

8. The method for intelligent monitoring of road subsidence according to claim 1, characterized in that: In step S4, the two settlement prediction models are combined to obtain a combined prediction model, which specifically includes: After the two settlement prediction models are established, the weights of different models are determined using the inverse error method to form a combined prediction model; Let y represent the actual settlement monitoring value, represents the predicted value of the i-th model at time t, and y(t) represents the predicted value of the combined model at time t; Calculate the absolute error e of the i-th model at time t i (t), the sum of the reciprocal of the absolute error d i (t), the weight w of the i-th model at time t i , and then get the combined model prediction value at time t, the calculation formula is as follows: 。 9. The method for intelligent monitoring of road subsidence according to claim 1, characterized in that: Step S5 also includes: feeding back the settlement prediction results to the user's remote terminal through cloud services to issue early warnings for excessive settlement and stability risks.

Images