Intelligent detection system and method for soil slope construction based on Internet of Things technology
By dividing the detection area in the soil ramp construction area and monitoring the electrical parameters in real time, using the decision tree model and fuzzy logic to analyze the grounding quality and data consistency, and dynamically adjust the detection strategy, the potential difference and common mode interference caused by poor grounding of sensor nodes are solved. The problem has achieved accurate monitoring and early warning of the construction status of soil ramps, and improved construction safety and management level.
Patent Information
- Application Number
- CN202411073799.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-08-07
AI Technical Summary
In soil ramp construction, poor grounding of sensor nodes leads to potential differences and common mode interference, affecting the accuracy and reliability of the detection system. Especially when long-term monitoring data is distorted, the system cannot continuously track changes in construction status. Losing all-round control of the construction environment.
By dividing the detection area in the soil ramp construction area, installing ground monitoring equipment, monitoring electrical parameters in real time, extracting resistance value drift characteristics and potential fluctuation characteristics, calculating the weight assignment of electrical parameters, and analyzing ground quality and data using decision tree model and fuzzy logic Consistency, dynamically adjust detection strategies and resource allocation, divide high, medium and low risk areas, and optimize detection plans.
It improves the accuracy and reliability of the monitoring system, ensures the safety and efficiency of the construction process, provides reliable early warning information, reduces accidents, and improves the overall construction management level.
Smart Images

Figure CN118607885B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of civil engineering and Internet of Things, and in particular to an intelligent detection system and method for soil slope construction based on Internet of Things technology. Background Art
[0002] Intelligent detection of earth ramp construction based on Internet of Things technology refers to the deployment of various sensors (such as temperature, humidity, pressure, displacement, etc.) at the earth ramp construction site to collect data on the construction environment and earth ramp structure in real time, and transmit these data to the central control system through wireless networks. By using data analysis and artificial intelligence algorithms, the stability and safety of the earth ramp are monitored and warned in real time to ensure the safety and efficiency of the construction process. However, if the sensor node is poorly grounded, potential differences will occur between the nodes. This potential difference will propagate through the transmission line, causing common mode interference in the entire network, causing synchronization distortion of multiple sensor nodes. And if the long-term monitoring data is distorted, the detection system may not be able to continuously track changes in the construction status of the earth ramp and lose the ability to fully control the construction environment. Summary of the invention
[0003] The purpose of the present invention is to provide an intelligent detection system and method for soil slope construction based on Internet of Things technology to solve the shortcomings of the background technology.
[0004] In order to achieve the above object, the present invention provides the following technical solution: an intelligent detection method for soil slope construction based on Internet of Things technology, comprising the following steps:
[0005] S1: Divide the earth ramp construction area into several detection areas, determine the number and type of sensors in each detection area, install ground monitoring equipment at each sensor node, and conduct real-time monitoring of the electrical parameters at the connection point;
[0006] S2: After preprocessing the monitored electrical parameters, feature extraction is performed, and the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weight assignment of the electrical parameters of each sensor grounding point;
[0007] S3: Analyze the distortion trend of the earth slope construction status data measured by sensors in each detection area based on the decision tree model to determine the consistency of the earth slope construction status data;
[0008] S4: Analyze the consistency of the overall grounding quality coefficient of the sensor nodes and the soil slope construction status data in each detection area through fuzzy logic, evaluate the risk of each detection area, and divide each detection area into high-risk detection area, medium-risk detection area and low-risk detection area according to the evaluation results;
[0009] S5: Dynamically adjust detection strategies and resource allocation based on the division results, focus on monitoring high-risk areas, and optimize detection plans for medium- and low-risk areas.
[0010] In a preferred embodiment, in S1, a geographic information system is used to obtain terrain data of the earth ramp construction area. According to the terrain characteristics of the earth ramp construction area, the area with complex terrain is divided into small-area high-density detection areas to ensure fine monitoring; the middle of the slope and the area with gentle terrain changes are divided into medium-sized detection areas for routine monitoring; the foot of the slope and the area with stable geological structure are divided into large-area low-density detection areas for basic monitoring; the number and type of sensors are configured according to the specific environment and monitoring requirements of each detection area.
[0011] In a preferred embodiment, in S2, the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, and the resistance drift index is generated by the resistance value drift characteristics of the electrical parameters. The method for obtaining the resistance drift index is:
[0012] Select smoothing parameters α and set the time window size to W, set the initial resistance value S0 and the initial average value M0 of the sliding window; calculate the exponential smoothing value and the sliding window average value, where the specific calculation expression of the exponential smoothing value St is: ; The value range of α is , t is the time, Rt is the data point at time t, and the calculation expression of the sliding window average Mt is: ; In the formula, For the i-th data point, the calculated exponential smoothing value and sliding window average are analyzed to calculate the resistance drift index. The specific calculation expression is: ; In the formula, is the resistance drift index.
[0013] In a preferred embodiment, in S2, the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, and the voltage fluctuation anomaly index is generated by the potential fluctuation characteristics. The method for obtaining the voltage fluctuation anomaly index is:
[0014] Collect voltage time series data to form a data set , where Vt represents the voltage value at time t, , is a positive integer greater than 0. The voltage fluctuation anomaly index is calculated through the long short-term memory network. The voltage time series data is used as the input layer, the hidden layer is the long short-term memory network unit, the output layer is the output predicted voltage value, the input sequence length is Ts, and the mean square error is used as the loss function to minimize the error between the predicted value and the actual value. The calculation expression of the loss function is: ; In the formula, is the mean square error, is the sample size, is the actual voltage value, To predict the voltage value, the real-time voltage data is input into the trained long short-term memory network model to predict the future voltage value, and the error between the actual voltage value and the predicted voltage value is calculated as the voltage fluctuation abnormality index.
[0015] In a preferred embodiment, the resistance drift index and the voltage fluctuation anomaly index are converted into a first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model predicts the weighted assignment label of the electrical parameters of each sensor grounding point with each group of first eigenvectors as the prediction target, and takes minimizing the sum of the prediction errors of the weighted assignment labels of the electrical parameters of all sensor grounding points as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The weighted assignment of the electrical parameters of each sensor grounding point is determined according to the model output results, and the overall grounding quality coefficient of the sensor nodes in each detection area is calculated after the weighted average calculation of the weighted assignment of the electrical parameters of each sensor grounding point is performed.
[0016] In a preferred embodiment, in S3, the distortion trend of the earth ramp construction status data measured by sensors in each detection area is analyzed based on a decision tree model, specifically:
[0017] The construction status data of the earth slope measured by sensors in each detection area are collected, and the data are preprocessed. The processed data set is divided into a training set and a test set. The decision tree model is trained using the training set data to determine the splitting points and splitting criteria of each feature and construct a tree structure.
[0018] Use the trained decision tree model to predict the test set data and determine whether each data point is distorted;
[0019] The model outputs the classification result for each data point;
[0020] The distortion ratio and change trend of the data in each detection area are counted, and the consistency of the data in each detection area is calculated according to the classification results of the decision tree model. The specific calculation expression is: ; In the formula, Consistent Ratio is the data consistency ratio coefficient, is the total number of data points measured by all sensors in the detection area within a certain period of time, Indicates the number of data points that fall within the expected normal range among all data points.
[0021] In a preferred embodiment, the calculated data consistency ratio coefficient is compared with a gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the data consistency ratio coefficient is compared with the first standard threshold and the second standard threshold respectively;
[0022] If the data consistency ratio coefficient is greater than the second standard threshold, the earth ramp construction status data measured by the sensor is classified as high consistency data; if the data consistency ratio coefficient is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the earth ramp construction status data measured by the sensor is classified as medium consistency data; if the data consistency ratio coefficient is less than the first standard threshold, the earth ramp construction status data measured by the sensor is classified as low consistency data.
[0023] In a preferred embodiment, in S4, the consistency of the overall grounding quality coefficient of the sensor nodes in each detection area and the earth ramp construction status data is analyzed by fuzzy logic, specifically:
[0024] The overall grounding quality factor and data consistency ratio factor are used as input items, and the risk classification is used as output item;
[0025] The specific grounding quality coefficient and data consistency degree are converted into membership degree in fuzzy set, and fuzzification is performed using membership function;
[0026] According to the fuzzy rules, the fuzzified data is inferred and the result of each rule is calculated;
[0027] Combining the results of all rules, we get the fuzzy output of each detection area;
[0028] According to the risk level converted from the fuzzy output, the detection areas with high risk levels are divided into high risk detection areas, the detection areas with medium risk levels are divided into medium risk detection areas, and the detection areas with low risk levels are divided into low risk detection areas.
[0029] In a preferred implementation, in S5, the detection strategy and resource allocation are dynamically adjusted according to the division result, specifically:
[0030] Mark the risk level as , x=1, 2, 3, For high-risk detection areas, It is a medium-risk detection area. It is a low-risk detection area; resource allocation is dynamically adjusted according to the risk level, and the amount of allocated resources is expressed as: ; is the amount of resources allocated to detection areas of different risk levels, C is the total amount of resources, and g is a function that calculates the amount of resources allocated to each area based on the risk level and the total amount of resources; the specific calculation expression is: ; where j is the number of the detection area with different risk levels, jϵx, The sum of the risk levels of all detection areas; the detection strategy is adjusted according to the amount of resources allocated;
[0031] Focus on monitoring high-risk areas and optimize the detection schemes for medium- and low-risk areas. The detection strategy is expressed as: ; is the detection strategy of the xth detection area, h is the function that calculates the detection strategy based on the risk level and resource quantity, and and risk level , adjust the detection strategy for each detection area , increase the detection frequency and number of sensors in high-risk areas, and optimize the detection schemes in medium and low-risk areas to improve resource utilization efficiency.
[0032] The present invention also provides an intelligent detection system for earth slope construction based on Internet of Things technology, including an electrical parameter acquisition module, a grounding quality analysis module, a construction data trend analysis module, a detection area division module and a dynamic adjustment module;
[0033] Electrical parameter acquisition module: Divide the earth ramp construction area into several detection areas, determine the number and type of sensors in each detection area, install grounding monitoring equipment at each sensor node, and conduct real-time monitoring of the electrical parameters at the connection point;
[0034] Grounding quality analysis module: After pre-processing the monitored electrical parameters, feature extraction is performed. According to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, the weighted values of the electrical parameters of each sensor grounding point are determined. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weighted values of the electrical parameters of each sensor grounding point.
[0035] Construction data trend analysis module: Based on the decision tree model, the distortion trend of the earth slope construction status data measured by sensors in each detection area is analyzed to determine the consistency of the earth slope construction status data;
[0036] Detection area division module: Analyze the overall grounding quality coefficient of sensor nodes in each detection area and the consistency of soil slope construction status data through fuzzy logic, evaluate the risk of each detection area, and divide each detection area into high-risk detection area, medium-risk detection area and low-risk detection area according to the evaluation results;
[0037] Dynamic adjustment module: Dynamically adjust detection strategies and resource allocation according to the division results, focus on monitoring high-risk areas, and optimize detection plans for medium and low-risk areas.
[0038] In the above technical solution, the technical effects and advantages provided by the present invention are:
[0039] 1. The present invention deploys various sensors at the construction site for real-time data collection and monitoring, and uses data analysis and artificial intelligence algorithms to evaluate and warn the stability and safety of the earth slope, thereby ensuring the safety and efficiency of the construction process. The system includes an electrical parameter acquisition module, a grounding quality analysis module, a construction data trend analysis module, a detection area division module, and a dynamic adjustment module, which realizes a comprehensive analysis of grounding quality and data consistency, dynamically adjusts detection strategies and resource allocation, and effectively improves the accuracy and reliability of the monitoring system.
[0040] 2. The present invention analyzes the risk level of each detection area through fuzzy logic and decision tree models, divides the area into high risk, medium risk and low risk, and dynamically adjusts the detection strategy according to the risk level, focusing on monitoring high-risk areas, optimizing the detection schemes for medium and low-risk areas, improving resource utilization efficiency, and ensuring that the system can continuously and accurately monitor the construction status of the earth slope, provide reliable early warning information, thereby ensuring construction safety, reducing accidents, and improving the overall construction management level. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0042] Figure 1 The figure is a flow chart of the method of the present invention.
[0043] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0045] Example 1: Please refer to Figure 1 As shown, the intelligent detection method for soil slope construction based on Internet of Things technology described in this embodiment includes the following steps:
[0046] S1: Divide the earth ramp construction area into several detection areas, determine the number and type of sensors in each detection area, install ground monitoring equipment at each sensor node, and conduct real-time monitoring of the electrical parameters at the connection point;
[0047] S2: After preprocessing the monitored electrical parameters, feature extraction is performed, and the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weight assignment of the electrical parameters of each sensor grounding point;
[0048] S3: Analyze the distortion trend of the earth slope construction status data measured by sensors in each detection area based on the decision tree model to determine the consistency of the earth slope construction status data;
[0049] S4: Analyze the consistency of the overall grounding quality coefficient of the sensor nodes and the soil slope construction status data in each detection area through fuzzy logic, evaluate the risk of each detection area, and divide each detection area into high-risk detection area, medium-risk detection area and low-risk detection area according to the evaluation results;
[0050] S5: Dynamically adjust detection strategies and resource allocation based on the division results, focus on monitoring high-risk areas, and optimize detection plans for medium- and low-risk areas.
[0051] Among them, in S1, the earth slope construction area is divided into several detection areas, the number and type of sensors in each detection area are determined, and grounding monitoring equipment is installed at each sensor node to monitor the electrical parameters of the connection point in real time.
[0052] Ensure that sensor nodes evenly cover the entire earth slope construction area to avoid monitoring blind spots. Use a geographic information system (GIS) to obtain terrain data for the earth slope construction area. According to the terrain characteristics of the earth slope construction area, areas with complex terrain such as the top of the slope, steep slopes, fault zones, and near water sources are divided into small-area high-density detection areas to ensure fine monitoring. The middle of the slope and areas with gentle terrain changes are divided into medium-sized detection areas for routine monitoring. The foot of the slope and areas with stable geological structures are divided into large-area low-density detection areas for basic monitoring.
[0053] Configure the appropriate number and type of sensors according to the specific environment and monitoring needs of each detection area. Use temperature sensors to monitor the temperature changes of the environment and earth ramp materials, and use humidity sensors to monitor the humidity of the construction area, especially in the rainy season or near water sources. Use pressure sensors to monitor the pressure on the earth ramp, especially in heavy loads or high stress areas. Use displacement sensors to monitor the displacement and deformation of the earth ramp, especially at key locations such as the top and foot of the slope.
[0054] Determine the number of sensors based on the area of each detection area and the number of construction equipment. Install grounding monitoring equipment next to each sensor node to ensure that the grounding device is closely connected to the sensor. Select appropriate grounding methods based on the geological and environmental conditions of the construction site, such as independent grounding, grounding grid, etc. Use corrosion-resistant, highly conductive grounding cables to connect the grounding end of the monitoring equipment to the grounding point. Use the grounding monitoring equipment installed at each sensor node to measure the resistance and potential data of the grounding point in real time. Transmit the measured electrical parameter data to the central control system via a wireless network for real-time monitoring and analysis.
[0055] S2: After preprocessing the monitored electrical parameters, feature extraction is performed. According to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, the weighted values of the electrical parameters of each sensor grounding point are determined. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weighted values of the electrical parameters of each sensor grounding point.
[0056] Collect electrical parameter data from sensor nodes in real time, including resistance and potential data. Use filtering algorithms (such as moving average filtering and Kalman filtering) to remove high-frequency noise in the data to ensure data smoothness. Use statistical methods to identify and remove outliers to ensure data accuracy. Remove abnormal data points that exceed a certain standard deviation range. Normalize electrical parameters of different dimensions so that the data are in the same range (such as 0-1) for subsequent analysis and comparison.
[0057] According to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, the weight assignment of the electrical parameters of each sensor grounding point is determined, and the resistance drift index is generated by the resistance value drift characteristics of the electrical parameters. The method for obtaining the resistance drift index is:
[0058] Select smoothing parameters α and set the time window size to W, set the initial resistance value S0 and the initial average value M0 of the sliding window, usually the first data point R0 is selected as the initial value; calculate the exponential smoothing value and the sliding window average value, where the specific calculation expression of the exponential smoothing value St is: ; The value range of α is , t is the time, Rt is the data point at time t, and the calculation expression of the sliding window average Mt is: ; In the formula, For the i-th data point, the calculated exponential smoothing value and sliding window average are analyzed to calculate the resistance drift index. The specific calculation expression is: ; In the formula, is the resistance drift index.
[0059] The larger the resistance drift index, the faster the grounding resistance value changes and the greater the fluctuation. This indicates that the resistance value of the grounding system is unstable, and there may be problems such as loose grounding electrodes and changes in soil resistivity, which lead to poor grounding quality. The stability of the grounding resistance value is crucial to maintaining a good grounding effect.
[0060] A high resistance drift index means that the grounding resistance value is unstable and cannot effectively suppress electromagnetic interference, which in turn affects the accuracy of sensor data and the reliability of the system. Poor grounding can lead to errors and distortions in the collected data, affecting the monitoring accuracy of the intelligent detection system for earth slope construction, and may also cause unstable operation or failure of electrical equipment.
[0061] Poor grounding quality increases the risk of electric shock, especially when the live housing of electrical equipment cannot be effectively grounded. A high resistance drift index indicates that the grounding system cannot reliably protect personnel safety and there is a risk of poor grounding. In addition, the grounding system plays a key role in lightning protection. A high resistance drift index indicates that the grounding system may not be able to effectively protect against lightning, increasing the risk of lightning damage and overall safety hazards.
[0062] According to the resistance drift characteristics and potential fluctuation characteristics of the electrical parameters, the weight assignment of the electrical parameters of each sensor grounding point is determined, and the voltage fluctuation anomaly index is generated through the potential fluctuation characteristics. The method for obtaining the voltage fluctuation anomaly index is:
[0063] Collect voltage time series data to form a data set , where Vt represents the voltage value at time t, , is a positive integer greater than 0. The voltage fluctuation anomaly index is calculated through the long short-term memory network. The voltage time series data is used as the input layer, the hidden layer is the long short-term memory network unit, the output layer is the output predicted voltage value, the input sequence length is Ts, and the mean square error is used as the loss function to minimize the error between the predicted value and the actual value. The calculation expression of the loss function is: ; In the formula, is the mean square error, is the sample size, is the actual voltage value, To predict the voltage value, the real-time voltage data is input into the trained long short-term memory network model to predict the future voltage value, and the error between the actual voltage value and the predicted voltage value is calculated as the voltage fluctuation abnormality index.
[0064] The larger the voltage fluctuation anomaly index is, the worse the overall grounding quality of the sensor nodes in each detection area is. Specifically, when the voltage fluctuation anomaly index increases, it indicates that the potential difference and voltage instability between sensor nodes are aggravated, which is usually a signal of poor grounding or severe electromagnetic interference. Due to the unstable potential, large noise and errors will appear in the data collection and transmission process of the sensor, resulting in data distortion, affecting the accuracy and reliability of the entire detection system.
[0065] The increase in the voltage fluctuation anomaly index also means that the electrical environment in the earth slope construction area may have large electromagnetic interference or grounding problems. High voltage fluctuations may cause communication signal interference between sensor nodes, increase the system's false alarm rate and missed alarm rate, and thus reduce the monitoring system's sensitivity and response speed to changes in the construction environment. This instability greatly reduces the risk assessment and early warning functions during the construction process, making it difficult to accurately grasp the stability and safety of the earth slope.
[0066] A high long-term voltage fluctuation anomaly index will seriously affect the stable operation of the earth slope construction safety monitoring system. High voltage fluctuations will not only cause continuous distortion of sensor data, but may also cause system failures or data loss, reducing the availability and reliability of the entire monitoring network. Therefore, the level of the voltage fluctuation anomaly index directly reflects the quality of the sensor node grounding. Effective measures must be taken, such as optimizing the grounding design, enhancing electromagnetic shielding, and implementing regular maintenance and calibration to ensure that the detection system can operate continuously and stably to ensure construction safety.
[0067] The resistance drift index and the voltage fluctuation anomaly index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model predicts the weighted label of the electrical parameters of each sensor grounding point with each group of first eigenvectors as the prediction target, and minimizes the sum of the prediction errors of the weighted labels of the electrical parameters of all sensor grounding points as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The weighted assignment of the electrical parameters of each sensor grounding point is determined according to the model output results, and the overall grounding quality coefficient of the sensor nodes in each detection area is calculated after the weighted average calculation of the weighted assignment of the electrical parameters of each sensor grounding point.
[0068] The method for obtaining the weighted values of the electrical parameters of each sensor grounding point is as follows: from the first eigenvector training data of the trained machine learning model, the corresponding function expression is obtained: ; In the formula, is the output function of the model, is the resistance drift index, is the voltage fluctuation abnormality index, Assign weights to the electrical parameters of each sensor ground point.
[0069] S3: Analyze the distortion trend of the earth slope construction status data measured by sensors in each detection area based on the decision tree model to determine the consistency of the earth slope construction status data.
[0070] In intelligent detection of earth slope construction, data measured by sensors is used to monitor the construction status in real time. However, sensor data may be distorted due to electromagnetic interference, poor grounding or other reasons. By analyzing the sensor measurement data in each detection area based on the decision tree model, the distortion trend of these data can be identified and judged, and then the consistency of the data can be evaluated. This is crucial to ensure the reliability and accuracy of the monitoring system.
[0071] The decision tree model is a machine learning algorithm commonly used for classification and regression. It divides data into different categories or predicted values through a tree structure. It can effectively handle nonlinear relationships and noise in data and plays an important role in identifying distortion trends in sensor data.
[0072] The construction status data of the earth slope measured by sensors in each detection area are collected, including temperature, humidity, pressure, displacement, etc.
[0073] Remove outliers and noise, fill in missing data, and ensure data completeness and accuracy.
[0074] Extract key features from the data, such as mean, standard deviation, rate of change, etc., as input to the decision tree model.
[0075] The processed data set is divided into training set and test set, generally with a ratio of 80% and 20%.
[0076] Use the training set data to train the decision tree model, determine the splitting points and splitting criteria of each feature, and build a tree structure.
[0077] Max depth: Limit the maximum depth of the decision tree to prevent overfitting.
[0078] Minimum number of samples (min samples split): The minimum number of samples contained in each node, which controls the splitting of the tree.
[0079] Information gain: A criterion for selecting the best split point, such as the Gini coefficient or entropy.
[0080] Use the trained decision tree model to predict the test set data and determine whether each data point is distorted.
[0081] The model outputs a classification result (normal or distorted) for each data point.
[0082] Count the distortion ratio and change trend of data in each detection area, and identify the pattern and cause of distorted data.
[0083] According to the classification results of the decision tree model, the consistency of the data in each detection area is calculated. The specific calculation expression is: ; In the formula, Consistent Ratio is the data consistency ratio coefficient, is the total number of data points measured by all sensors in the detection area within a certain period of time, Indicates the number of data points that fall within the expected normal range among all data points.
[0084] The calculated data consistency ratio coefficient is compared with the gradient standard threshold, wherein the gradient standard threshold in the present application includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the data consistency ratio coefficient is compared with the first standard threshold and the second standard threshold respectively;
[0085] If the data consistency ratio coefficient is greater than the second standard threshold, the soil ramp construction status data measured by the sensor is classified as high consistency data;
[0086] If the data consistency ratio coefficient is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the earth ramp construction status data measured by the sensor is classified as medium consistency data;
[0087] If the data consistency ratio coefficient is less than the first standard threshold, the earth slope construction status data measured by the sensor is classified as low consistency data.
[0088] S4: The overall grounding quality coefficient of the sensor nodes in each detection area and the consistency of the earth slope construction status data are analyzed through fuzzy logic to evaluate the risk of each detection area. According to the evaluation results, each detection area is divided into high-risk detection area, medium-risk detection area and low-risk detection area.
[0089] In the intelligent detection system for earth slope construction, the overall grounding quality coefficient of the sensor nodes in each detection area and the consistency of the earth slope construction status data are analyzed through fuzzy logic, which can more accurately evaluate the risk level of each area. According to the analysis results, each detection area is divided into high-risk, medium-risk and low-risk areas, so that corresponding monitoring and preventive measures can be taken to ensure the safety and efficiency of the construction process.
[0090] The overall grounding quality factor and data consistency ratio factor are used as input items, and the risk classification is used as output item;
[0091] The overall grounding quality coefficient is defined as three fuzzy sets of “high”, “medium” and “low” to describe different levels of grounding quality.
[0092] The data consistency ratio coefficient is defined as three fuzzy sets of "high", "medium" and "low" to describe different levels of data consistency.
[0093] Combining ground quality and data consistency, fuzzy logic rules are defined to evaluate the risk level. If the ground quality is "low" and the data consistency is "low", the risk level is "high risk". If the ground quality is "medium" and the data consistency is "medium", the risk level is "medium risk". If the ground quality is "high" and the data consistency is "high", the risk level is "low risk".
[0094] The specific grounding quality coefficient and data consistency degree are converted into membership degree in fuzzy sets.
[0095] Use membership functions (such as triangular and trapezoidal functions) for fuzzification.
[0096] According to the fuzzy rules, the fuzzified data is inferred and the result of each rule is calculated.
[0097] The results of all rules are combined to obtain the fuzzy output of each detection area.
[0098] According to the risk level converted from the fuzzy output, the detection areas with high risk levels are divided into high risk detection areas, the detection areas with medium risk levels are divided into medium risk detection areas, and the detection areas with low risk levels are divided into low risk detection areas.
[0099] S5: Dynamically adjust detection strategies and resource allocation based on the division results, focus on monitoring high-risk areas, and optimize detection plans for medium- and low-risk areas.
[0100] Mark the risk level as , x=1, 2, 3, For high-risk detection areas, It is a medium-risk detection area. It is a low-risk detection area; resource allocation is dynamically adjusted according to the risk level, and the amount of allocated resources is expressed as: ; is the amount of resources allocated to detection areas of different risk levels, C is the total amount of resources, and g is a function that calculates the amount of resources allocated to each area based on the risk level and the total amount of resources; the specific calculation expression is: ; where j is the number of the detection area with different risk levels, jϵx, The sum of the risk levels of all detection areas; the detection strategy is adjusted according to the amount of resources allocated;
[0101] Focus on monitoring high-risk areas and optimize the detection schemes for medium- and low-risk areas. The detection strategy is expressed as: ; is the detection strategy of the xth detection area, h is the function that calculates the detection strategy based on the risk level and resource quantity, and and risk level , adjust the detection strategy for each detection area , increase the detection frequency and number of sensors in high-risk areas, and optimize the detection schemes in medium and low-risk areas to improve resource utilization efficiency.
[0102] In this embodiment, the earth ramp construction area is divided into several detection areas, the number and type of sensors in each detection area are determined, and a grounding monitoring device is installed at each sensor node to monitor the electrical parameters of the docking location in real time; the monitored electrical parameters are pre-processed and feature extraction is performed, and the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, and the overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average according to the weight assignment of the electrical parameters of each sensor grounding point; based on the decision tree model, the distortion trend of the earth ramp construction status data measured by the sensors in each detection area is analyzed to determine the consistency of the earth ramp construction status data; the overall grounding quality coefficient of the sensor nodes in each detection area and the consistency of the earth ramp construction status data are analyzed by fuzzy logic, and the risk of each detection area is evaluated. According to the evaluation results, each detection area is divided into a high-risk detection area, a medium-risk detection area and a low-risk detection area; the detection strategy and resource allocation are dynamically adjusted according to the division results, and the high-risk area is monitored, and the detection scheme of the medium and low-risk areas is optimized. Thereby improving the accuracy and reliability of the detection system, ensuring the safety and efficiency of the construction process, and fully controlling changes in the construction environment.
[0103] Example 2: Please refer to Figure 2 As shown, the soil slope construction intelligent detection system based on the Internet of Things technology described in this embodiment includes an electrical parameter acquisition module, a grounding quality analysis module, a construction data trend analysis module, a detection area division module and a dynamic adjustment module;
[0104] Electrical parameter acquisition module: Divide the earth ramp construction area into several detection areas, determine the number and type of sensors in each detection area, install grounding monitoring equipment at each sensor node, and conduct real-time monitoring of the electrical parameters at the connection point;
[0105] Grounding quality analysis module: After pre-processing the monitored electrical parameters, feature extraction is performed. According to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, the weighted values of the electrical parameters of each sensor grounding point are determined. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weighted values of the electrical parameters of each sensor grounding point.
[0106] Construction data trend analysis module: Based on the decision tree model, the distortion trend of the earth slope construction status data measured by sensors in each detection area is analyzed to determine the consistency of the earth slope construction status data;
[0107] Detection area division module: Analyze the overall grounding quality coefficient of sensor nodes in each detection area and the consistency of soil slope construction status data through fuzzy logic, evaluate the risk of each detection area, and divide each detection area into high-risk detection area, medium-risk detection area and low-risk detection area according to the evaluation results;
[0108] Dynamic adjustment module: Dynamically adjust detection strategies and resource allocation according to the division results, focus on monitoring high-risk areas, and optimize detection plans for medium and low-risk areas.
[0109] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0110] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0111] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.
[0112] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0113] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0114] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.
Claims
1. An intelligent detection method for soil slope construction based on Internet of Things technology, characterized by: The steps include: S1: Divide the earth ramp construction area into several detection areas, determine the number and type of sensors in each detection area, install ground monitoring equipment at each sensor node, and conduct real-time monitoring of the electrical parameters at the connection point; S2: After preprocessing the monitored electrical parameters, feature extraction is performed, and the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weight assignment of the electrical parameters of each sensor grounding point; The step S2 specifically includes: determining the weight assignment of the electrical parameters of each sensor grounding point according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, and generating a resistance drift index according to the resistance value drift characteristics of the electrical parameters. The method for obtaining the resistance drift index is: Select smoothing parameter α and set the time window size to W, set the initial resistance value S0 and the initial average value M0 of the sliding window; calculate the exponential smoothing value and the sliding window average value, where the specific calculation expression of the exponential smoothing value St is: St = α*Rt+(1-α)*S(t-1); the value range of α is [0, 1], t is the time, Rt is the data point at time t, and the calculation expression of the sliding window average value Mt is: In the formula, R i For the i-th data point, the calculated exponential smoothing value and sliding window average are analyzed to calculate the resistance drift index. The specific calculation expression is: Where KP is the resistance drift index; S3: Analyze the distortion trend of the earth slope construction status data measured by sensors in each detection area based on the decision tree model to determine the consistency of the earth slope construction status data; S4: Analyze the consistency of the overall grounding quality coefficient of the sensor nodes and the soil slope construction status data in each detection area through fuzzy logic, evaluate the risk of each detection area, and divide each detection area into high-risk detection area, medium-risk detection area and low-risk detection area according to the evaluation results; S5: Dynamically adjust detection strategies and resource allocation based on the division results, focus on monitoring high-risk areas, and optimize detection plans for medium- and low-risk areas.
2. According to claim 1, the intelligent detection method for soil slope construction based on Internet of Things technology is characterized in that: In S1, the geographic information system is used to obtain the terrain data of the earth slope construction area. According to the terrain characteristics of the earth slope construction area, the area with complex terrain is divided into small-area high-density detection areas to ensure fine monitoring; the middle of the slope and the area with gentle terrain changes are divided into medium-sized detection areas for routine monitoring; the foot of the slope and the area with stable geological structure are divided into large-area low-density detection areas for basic monitoring; the number and type of sensors are configured according to the specific environment and monitoring needs of each detection area.
3. The intelligent detection method for soil slope construction based on Internet of Things technology according to claim 2 is characterized in that: In S2, the weight assignment of the electrical parameters of each sensor grounding point is determined according to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, and the voltage fluctuation anomaly index is generated through the potential fluctuation characteristics. The method for obtaining the voltage fluctuation anomaly index is: The voltage time series data is collected to form a data set V = {V1, V2, ..., Vt, ..., VT}, where Vt represents the voltage value at time t, t = 1, 2, ..., T, T is a positive integer greater than 0, and the voltage fluctuation anomaly index is calculated through the long short-term memory network. The voltage time series data is used as the input layer, the hidden layer is the long short-term memory network unit, and the output layer is the output predicted voltage value. The input sequence length is Ts, and the mean square error is used as the loss function to minimize the error between the predicted value and the actual value. The calculation expression of the loss function is: In the formula, MSE is the mean square error, N is the number of samples, Vi is the actual voltage value, and FVi is the predicted voltage value. The real-time voltage data is input into the trained long short-term memory network model to predict the future voltage value, and the error between the actual voltage value and the predicted voltage value is calculated as the voltage fluctuation abnormality index.
4. The intelligent detection method for soil slope construction based on Internet of Things technology according to claim 3 is characterized in that: The resistance drift index and the voltage fluctuation anomaly index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model predicts the weighted label of the electrical parameters of each sensor grounding point with each group of first eigenvectors as the prediction target, and minimizes the sum of the prediction errors of the weighted labels of the electrical parameters of all sensor grounding points as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The weighted assignment of the electrical parameters of each sensor grounding point is determined according to the model output results, and the overall grounding quality coefficient of the sensor nodes in each detection area is calculated after the weighted average calculation of the weighted assignment of the electrical parameters of each sensor grounding point.
5. The intelligent detection method for soil slope construction based on Internet of Things technology according to claim 4 is characterized in that: In S3, the distortion trend of the earth slope construction status data measured by sensors in each detection area is analyzed based on the decision tree model, specifically: The construction status data of the earth slope measured by sensors in each detection area are collected, and the data are preprocessed. The processed data set is divided into a training set and a test set. The decision tree model is trained using the training set data to determine the splitting points and splitting criteria of each feature and construct a tree structure. Use the trained decision tree model to predict the test set data and determine whether each data point is distorted; The model outputs the classification result for each data point; The distortion ratio and change trend of the data in each detection area are counted, and the consistency of the data in each detection area is calculated according to the classification results of the decision tree model. The specific calculation expression is: Where Consistent Ratio is the data consistency ratio coefficient, Total Number of Dat Points is the total number of data points measured by all sensors in the detection area within a certain period of time, and Number of Consistent Data Points indicates the number of data points that meet the expected normal range among all data points.
6. The intelligent detection method for soil slope construction based on Internet of Things technology according to claim 5 is characterized in that: The calculated data consistency ratio coefficient is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the data consistency ratio coefficient is compared with the first standard threshold and the second standard threshold respectively; If the data consistency ratio coefficient is greater than the second standard threshold, the earth ramp construction status data measured by the sensor is classified as high consistency data; if the data consistency ratio coefficient is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the earth ramp construction status data measured by the sensor is classified as medium consistency data; if the data consistency ratio coefficient is less than the first standard threshold, the earth ramp construction status data measured by the sensor is classified as low consistency data.
7. The intelligent detection method for soil slope construction based on Internet of Things technology according to claim 6 is characterized in that: In S4, the consistency of the overall grounding quality coefficient of the sensor nodes in each detection area and the soil slope construction status data is analyzed by fuzzy logic, specifically: The overall grounding quality factor and data consistency ratio factor are used as input items, and the risk classification is used as output item; The specific grounding quality coefficient and data consistency degree are converted into membership degree in fuzzy set, and fuzzification is performed using membership function; According to the fuzzy rules, the fuzzified data is inferred and the result of each rule is calculated; Combining the results of all rules, we get the fuzzy output of each detection area; According to the risk level converted from the fuzzy output, the detection areas with high risk levels are divided into high risk detection areas, the detection areas with medium risk levels are divided into medium risk detection areas, and the detection areas with low risk levels are divided into low risk detection areas.
8. The intelligent detection method for soil slope construction based on Internet of Things technology according to claim 7 is characterized in that: In S5, the detection strategy and resource allocation are dynamically adjusted according to the division results, specifically: Mark the risk level as A x , x = 1, 2, 3, A1 is the high-risk detection area, A2 is the medium-risk detection area, and A3 is the low-risk detection area; resource allocation is dynamically adjusted according to the risk level, and the amount of allocated resources is expressed as: Q x =g(A x , C); Q x is the amount of resources allocated to detection areas of different risk levels, C is the total amount of resources, and g is a function that calculates the amount of resources allocated to each area based on the risk level and the total amount of resources; the specific calculation expression is: Where j is the label of the detection area with different risk levels, j∈x, The sum of the risk levels of all detection areas; the detection strategy is adjusted according to the amount of resources allocated; Focus on monitoring high-risk areas and optimize the detection schemes for medium- and low-risk areas. The detection strategy is expressed as: S x =h(A x , Q x );S x is the detection strategy of the xth detection area, h is the function for calculating the detection strategy based on the risk level and resource quantity, and Q is the allocated resource quantity. x and risk level A x , adjust the detection strategy S for each detection area x , increase the detection frequency and number of sensors in high-risk areas, and optimize the detection schemes in medium and low-risk areas to improve resource utilization efficiency.
9. An intelligent detection system for earth slope construction based on Internet of Things technology, used to implement the intelligent detection method for earth slope construction based on Internet of Things technology according to any one of claims 1 to 8, characterized in that: It includes electrical parameter acquisition module, grounding quality analysis module, construction data trend analysis module, detection area division module and dynamic adjustment module; Electrical parameter acquisition module: Divide the earth ramp construction area into several detection areas, determine the number and type of sensors in each detection area, install grounding monitoring equipment at each sensor node, and conduct real-time monitoring of the electrical parameters at the connection point; Grounding quality analysis module: After pre-processing the monitored electrical parameters, feature extraction is performed. According to the resistance value drift characteristics and potential fluctuation characteristics of the electrical parameters, the weighted values of the electrical parameters of each sensor grounding point are determined. The overall grounding quality coefficient of the sensor nodes in each detection area is calculated by weighted average based on the weighted values of the electrical parameters of each sensor grounding point. Construction data trend analysis module: Based on the decision tree model, the distortion trend of the earth slope construction status data measured by sensors in each detection area is analyzed to determine the consistency of the earth slope construction status data; Detection area division module: Analyze the overall grounding quality coefficient of sensor nodes in each detection area and the consistency of soil slope construction status data through fuzzy logic, evaluate the risk of each detection area, and divide each detection area into high-risk detection area, medium-risk detection area and low-risk detection area according to the evaluation results; Dynamic adjustment module: Dynamically adjust detection strategies and resource allocation according to the division results, focus on monitoring high-risk areas, and optimize detection plans for medium and low-risk areas.
