High liquid limit soil layered filling compaction degree monitoring and adjusting system

Through drones collecting data and using fast Fourier transform and convolutional neural network for real-time analysis and prediction, the shortcomings of traditional compaction monitoring systems in data processing and decision support are solved, and the accuracy of compaction monitoring and adjustment is improved. and efficiency to ensure the stability and safety of the building.

CN118997237APending Publication Date: 2024-11-22HEBEI EXPRESSWAY HANDANG EXPRESSWAY CO LTD +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411260325.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Traditional compaction monitoring systems have shortcomings in real-time data processing and analysis, making it difficult to quickly and accurately analyze compaction and humidity changes, resulting in lag in compaction monitoring results, affecting construction quality control, and being unable to under complex terrain or environmental conditions. Provide sufficient decision support, making it difficult to achieve the expected results of compaction operations.

Method used

The position sensor and humidity sensor equipped with the drone are used for initial data acquisition, combined with fast Fourier transform and convolutional neural network, real-time compaction data analysis and prediction are realized, parameter adjustment scheme is proposed, and the degree of automation of the compaction process is improved. Operational efficiency.

Benefits of technology

It improves data processing efficiency and accuracy, enhances the system's prediction capabilities and decision-making support functions, ensures the accuracy and efficiency of soil compaction monitoring and adjustment, improves the stability and safety of the building, and extends the service life of the building.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118997237A_ABST
    Figure CN118997237A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of compaction monitoring, in particular to a high liquid limit soil layered filling compaction degree monitoring and adjusting system which comprises an initial compaction data acquisition module: based on a position sensor and a humidity sensor carried by an unmanned aerial vehicle, the unmanned aerial vehicle is used for efficiently monitoring and acquiring the specific position and humidity value of an initial soil layer, and the initial compaction data acquisition module is used for acquiring the initial compaction data; and recording operation parameters of the initial compaction machine, and generating a baseline compaction data record. According to the invention, real-time compaction data analysis is carried out through fast Fourier transform, the compaction degree and humidity change of the soil are rapidly analyzed, and the compaction degree analysis result is generated by comparing with the preset compaction standard, so that the data processing efficiency and accuracy are improved, and the real-time performance and reliability of the monitoring data are ensured; the convolutional neural network calculates and predicts the optimal parameters of the compaction operation in the future and estimates the compaction effect by combining the historical compaction data and the real-time analysis result, so that the automation degree and the operation efficiency of the compaction process are improved, and the precision and the efficiency of soil compaction monitoring and adjustment are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of compaction monitoring, and in particular to a system for monitoring and adjusting the compaction degree of layered filling of high liquid limit soil. Background Art

[0002] The field of compaction monitoring technology focuses on ensuring that soil compaction meets the standards required for construction and civil engineering through various detection and adjustment technologies. Compaction monitoring technology involves the use of advanced sensors, instruments and data processing software to measure soil compaction, moisture and other key physical properties to ensure that the stability and bearing capacity of the soil structure meet the design requirements.

[0003] The high liquid limit soil layered filling compaction degree monitoring and adjustment system is used to monitor and adjust the compaction degree of high liquid limit soil used in the construction process. The purpose is to ensure that the soil reaches the optimal compaction standard through real-time monitoring of the compaction degree and corresponding adjustments, improve the stability and safety of the building, prevent uneven foundation settlement or structural failure due to insufficient compaction, ensure project quality and extend the service life of the building.

[0004] Traditional compaction monitoring systems have shortcomings in real-time data processing and analysis. It is difficult to quickly and accurately analyze compaction and humidity changes, resulting in delayed compaction monitoring results, affecting construction quality control. The lack of systematic prediction and optimization methods can easily lead to uneven compaction, which in turn causes problems such as uneven foundation settlement or structural failure. Under certain complex terrain or environmental conditions, traditional systems cannot provide sufficient decision-making support, resulting in compaction operations that are difficult to achieve the expected results, affecting project quality and the service life of buildings, limiting the application scope and effectiveness of traditional compaction monitoring systems, and urgently needing to be improved through innovative technical means. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a system for monitoring and adjusting the compaction degree of layered filling of high liquid limit soil.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A system for monitoring and adjusting the compaction degree of layered filling of high liquid limit soil comprises:

[0007] Initial compaction data collection module: Based on the position sensor and humidity sensor carried by the drone, the drone is used for efficient monitoring to collect the specific position and humidity value of the starting soil layer, and record the operating parameters of the initial compaction machinery to generate baseline compaction data records;

[0008] Real-time compaction data analysis module: Based on the baseline compaction data record, the module uses fast Fourier transform to process the data, analyze the specific compaction degree and moisture changes of the soil, and compares the preset compaction standard to generate compaction analysis results;

[0009] Compaction effect prediction module: through the compaction degree analysis results, using convolutional neural network, combined with historical compaction data, calculate and predict the optimal parameters of future compaction operations, and estimate the compaction effect that may be achieved, and generate a compaction effect prediction value;

[0010] Compaction adjustment decision module: based on the compaction effect prediction value, analyzes the deviation between the current compaction operation and the preset standard, proposes a parameter adjustment plan, and directly transmits the adjustment suggestion to the operating system to generate an operation optimization suggestion;

[0011] Feedback execution module: adjusts the operating parameters of the compaction machinery through the operation optimization suggestions, monitors the real-time execution status after the adjustment, and performs real-time recording of the operation data to generate real-time operation feedback;

[0012] Long-term performance monitoring module: Based on the real-time operation feedback, analyze the long-term trend of the data, identify potential performance problems, conduct performance evaluation, and generate a performance evaluation report.

[0013] As a further solution of the present invention, the initial compaction data acquisition module includes a position information acquisition submodule, a humidity detection submodule and an operation parameter recording submodule, wherein:

[0014] Position information acquisition submodule: Use the position sensor on the compaction machinery to capture the specific position information of the soil layer in real time, perform signal processing and calibration, calibrate the position data, record the coordinate points, and generate geographic coordinate information;

[0015] Humidity detection submodule: monitors soil humidity level through humidity sensors, calibrates and standardizes humidity data, summarizes humidity index, and generates humidity data records;

[0016] Operation parameter recording submodule: records the operation parameters of the compaction machinery, including pressure and speed, integrates the geographic coordinate information and moisture data records, integrates the compaction condition information, draws a baseline compaction data chart, and establishes a baseline compaction data archive.

[0017] As a further solution of the present invention, the real-time compaction data analysis module includes a data synchronization submodule, a compaction degree analysis submodule and a standard comparison submodule, wherein:

[0018] Data processing submodule: Based on the baseline compaction data records collected by the drone, perform data cleaning, remove obvious errors and records that do not conform to the format due to environmental interference or sensor errors, process missing data points that may appear in complex terrain or signal interference, and generate optimized data sets;

[0019] Compactness analysis submodule: Based on the optimized data set, compactness analysis is performed, data is smoothed by moving average method, noise in the data collected by the drone is reduced, thresholds are set, significant compactness changes are identified, compactness information is extracted by adaptive filtering and time-frequency analysis techniques, and the reflected wave signal obtained by the drone is processed by fast Fourier transform to obtain compactness analysis data;

[0020] Standard comparison submodule: Based on the compaction analysis data, a difference calculation method is used to compare the difference between the actual compaction data and the preset compaction standard, and the key deviation points are determined through the difference to obtain the standard comparison results.

[0021] As a further solution of the present invention, the fast Fourier transform is according to the formula:

[0022]

[0023] Where: X(k) represents the frequency domain signal, x(n) represents the time domain signal, N is the number of sampling points, k is the frequency index, j is the imaginary unit, and w(n) is the window function.

[0024] As a further solution of the present invention, the compaction effect prediction module includes a parameter calculation submodule, an effect estimation submodule and a prediction value generation submodule, wherein:

[0025] Parameter calculation submodule: based on the historical compaction data and compaction analysis report, the data is corrected and matched, the parameter configuration is iteratively optimized through the convolutional neural network, and the parameter sensitivity analysis is performed, the operation parameters are synchronously updated, and the parameter optimization scheme is generated;

[0026] Effect estimation submodule: Based on the parameter optimization scheme, execute the simulation prediction process, adjust the simulation input variables, estimate the compaction effect under different operating conditions through iterative testing, statistically analyze the simulation output, and generate an effect estimation report;

[0027] Prediction value generation submodule: Based on the effect estimation report, organize and analyze the simulation data, form a compaction effect prediction through data fusion technology, output the prediction result, and establish the compaction effect prediction value.

[0028] As a further solution of the present invention, the convolutional neural network is according to the formula:

[0029] y=f(W·x+b+α·z+β·v)

[0030] Where: y is the output value, f is the activation function, W is the weight matrix, x is the input vector, b is the bias value, α is the adaptive adjustment coefficient, z is the external environment parameter, β is the adjustment coefficient, and v is the time series parameter.

[0031] As a further solution of the present invention, the compaction adjustment decision module includes a compaction effect analysis submodule, a parameter adjustment submodule and an operation optimization suggestion submodule, wherein:

[0032] Compaction effect analysis submodule: based on the compaction effect prediction value, compare the real-time compaction operation data, analyze the deviation size and direction, subdivide various deviations, identify key deviation areas, summarize the deviation causes, and generate a deviation degree assessment;

[0033] Parameter adjustment submodule: Based on the deviation degree assessment, by screening the range and value of the adjustment parameters, adopting the optimization model adjustment strategy, determining the specific adjustment value, evaluating the potential adjustment effect, and forming a complete parameter adjustment plan, the adjustment strategy details are obtained;

[0034] Operation optimization suggestion submodule: Based on the adjustment strategy details, formulate operation suggestions, format them into executable operation instructions, and transmit them directly to the operating system through the interface to obtain operation optimization suggestions.

[0035] As a further solution of the present invention, the feedback execution module includes a parameter adjustment submodule, a real-time monitoring submodule and an operation recording submodule, wherein:

[0036] Parameter adjustment submodule: based on the operation optimization suggestion, adjust the operation parameters of the compaction machinery, by adjusting the power, speed and compaction depth of the compaction machinery, and recording the adjusted parameter settings to generate optimized operation parameters;

[0037] Real-time monitoring submodule: Based on the optimized operating parameters, real-time monitoring is performed, and the operating data of the compaction machinery, including temperature, vibration frequency, compaction effect, etc., is collected through sensors, and the change trend is analyzed to generate real-time monitoring data;

[0038] Operation record submodule: Based on the real-time monitoring data, real-time recording of operation data is performed, recording operation data for each time period, including equipment operation status, parameter adjustment records and compaction effect records, and generating real-time operation feedback.

[0039] As a further solution of the present invention, the long-term performance monitoring module includes a data analysis submodule, a problem identification submodule and a performance evaluation submodule, wherein:

[0040] Data analysis submodule: Based on the real-time operation feedback, long-term trend analysis of data is performed, by summarizing the operation data collected over a long period of time, generating trend charts, analyzing the changing trends of data, identifying potential abnormalities, and generating long-term trend data;

[0041] Problem identification submodule: Based on the long-term trend data, potential performance problems are identified, deviations and anomalies are detected by comparing actual performance with expected standards, performance problem areas are marked, and a list of potential problems is generated;

[0042] Performance evaluation submodule: Based on the potential problem list, a performance evaluation is conducted to confirm the severity and scope of the problem through detailed inspection and testing, record specific performance indicators and evaluation results, and generate a performance evaluation report.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are:

[0044] 1. In the present invention, real-time compaction data analysis is performed through fast Fourier transform, soil compaction and moisture changes are quickly analyzed, and compaction analysis results are generated by comparing with preset compaction standards, thereby improving data processing efficiency and accuracy, and ensuring the real-time and reliability of monitoring data.

[0045] 2. In the present invention, a convolutional neural network is used to calculate and predict the optimal parameters of future compaction operations and estimate the compaction effect by combining historical compaction data and real-time analysis results, thereby enhancing the prediction ability and decision support function of the system and making the compaction operation more accurate;

[0046] 3. In the present invention, by analyzing the deviation between the current compaction operation and the preset standard, a parameter adjustment plan is proposed, and the adjustment suggestion is directly transmitted to the operating system, thereby improving the automation level and operating efficiency of the compaction process. Innovations in data processing, predictive analysis and decision support have improved the accuracy and efficiency of soil compaction monitoring and adjustment, improved the stability and safety of buildings, ensured project quality and extended the service life of buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a system flow chart of the present invention;

[0048] Figure 2 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] Embodiment 1

[0051] See also Figure 1 The present invention provides a technical solution: a high liquid limit soil layer filling compaction degree monitoring and adjustment system comprises:

[0052] Initial compaction data collection module: Based on the position sensor and humidity sensor carried by the drone, the drone is used for efficient monitoring to collect the specific position and humidity value of the starting soil layer, and record the operating parameters of the initial compaction machinery to generate baseline compaction data records;

[0053] Real-time compaction data analysis module: Based on the baseline compaction data records, the module uses fast Fourier transform to process the data, analyze the specific compaction degree and moisture changes of the soil, and compares it with the preset compaction standards to generate compaction analysis results;

[0054] Compaction effect prediction module: Based on the compaction analysis results, a convolutional neural network is used in combination with historical compaction data to calculate and predict the optimal parameters for future compaction operations, estimate the compaction effects that may be achieved, and generate compaction effect prediction values;

[0055] Compaction adjustment decision module: Based on the compaction effect prediction value, it analyzes the deviation between the current compaction operation and the preset standard, proposes parameter adjustment plans, and directly transmits the adjustment suggestions to the operating system to generate operation optimization suggestions;

[0056] Feedback execution module: adjust the operating parameters of the compaction machinery through operation optimization suggestions, monitor the real-time execution status after adjustment, and record the operation data in real time to generate real-time operation feedback;

[0057] Long-term performance monitoring module: Based on real-time operation feedback, analyze the long-term trend of data, identify potential performance problems, conduct performance evaluation, and generate performance evaluation reports.

[0058] See also Figure 2 The initial compaction data acquisition module includes a position information acquisition submodule, a humidity detection submodule and an operation parameter recording submodule, wherein:

[0059] Position information acquisition submodule: Use the position sensor on the compaction machinery to capture the specific position information of the soil layer in real time, perform signal processing and calibration, calibrate the position data, record the coordinate points, and generate geographic coordinate information;

[0060] Humidity detection submodule: monitors soil humidity level through humidity sensors, calibrates and standardizes humidity data, summarizes humidity index, and generates humidity data records;

[0061] Operation parameter recording submodule: records the operation parameters of compaction machinery, including pressure and speed, integrates geographic coordinate information and moisture data records, integrates compaction condition information, draws baseline compaction data charts, and establishes baseline compaction data archives;

[0062] In the position information acquisition submodule, based on the position sensor on the compaction machinery, the position information of the soil layer is captured in real time, the signal is converted into the frequency domain using the Fourier transform algorithm, and noise is suppressed through a high-pass filter. The filter cutoff frequency is set to 500Hz and the gain is set to 20dB. The position data is calibrated, the coordinate points are recorded, and the geographic coordinate information is generated;

[0063] In the humidity detection submodule, based on the humidity sensor, linear regression analysis is used to monitor the soil humidity level, and the regression model parameters are set, including a slope of 0.01 and an intercept of 5. The least squares method is used to optimize the parameters, and the humidity data is standardized. The standard deviation and mean are used to normalize the data, and the humidity index is summarized to generate humidity data records;

[0064] In the operating parameter recording submodule, based on the sensor data of the compaction machinery, data integration technology is used to record the operating parameters of the compaction machinery, including pressure and speed. The geographic coordinate information and humidity data records are merged using the data fusion algorithm. The data fusion threshold is set to 0.05, and the weighted average method is used for information integration. A baseline compaction data chart is drawn and a baseline compaction data archive is established.

[0065] See also Figure 2 The real-time compaction data analysis module includes a data synchronization submodule, a compaction degree analysis submodule and a standard comparison submodule, among which:

[0066] Data processing submodule: Based on the baseline compaction data records collected by the drone, perform data cleaning, remove obvious errors and records that do not conform to the format due to environmental interference or sensor errors, process missing data points that may appear in complex terrain or signal interference, and generate optimized data sets;

[0067] Compactness analysis submodule: Based on the optimized data set, compaction analysis is performed, the data is smoothed by the moving average method to reduce the noise in the data collected by the drone, and thresholds are set to identify significant compaction changes. Adaptive filtering and time-frequency analysis techniques are used to extract compaction information, and fast Fourier transform is used to process the reflected wave signal obtained by the drone to obtain compaction analysis data;

[0068] Standard comparison submodule: Based on the compaction analysis data, the difference calculation method is used to compare the difference between the actual compaction data and the preset compaction standard, and the key deviation points are determined through the difference to obtain the standard comparison results;

[0069] In the data processing submodule, based on the baseline compaction data collected by the drone, data cleaning technology is used to remove format error records caused by environmental interference or sensor errors, and data loss caused by the drone in complex terrain or signal interference is processed. The missing data points are filled using the linear interpolation method, and the logistic regression model is used to identify and eliminate abnormal data to generate an optimized data set;

[0070] In the compaction analysis submodule, the moving average method is used to smooth the data using the optimized data set, the window size is set to 5 to reduce the noise in the data, the threshold analysis is used to identify significant compaction changes, the threshold is set to the data average plus two times the standard deviation, the adaptive filtering technology is used to perform time-frequency analysis on the compaction, and the fast Fourier transform algorithm is used to process the reflected wave signal collected by the drone to generate compaction analysis data;

[0071] In the standard comparison submodule, based on the compaction analysis data, the difference calculation method is used to compare the difference between the actual compaction data and the preset compaction standard. The total deviation value is calculated using the square sum algorithm, the key deviation points are identified, the deviation threshold is set to 10%, and the standard comparison results are generated.

[0072] Fast Fourier transform, according to the formula:

[0073]

[0074] Where: X(k) represents the frequency domain signal, x(n) represents the time domain signal, N is the number of sampling points, k is the frequency index, j is the imaginary unit, and w(n) is the window function;

[0075] Execution process: First, the time domain signal x(n) is obtained from the layered filling process of high liquid limit soil. The sensor detects the compaction reflection wave signal, which contains the information of the compaction degree of the soil layer. The collected time domain signal x(n) is processed by the window function w(n). Commonly used window functions include Hanning window, Hamming window and Blackman window. The selection of window function depends on the characteristics of the signal and the analysis requirements. Through the processing of window function, the spectrum leakage effect can be reduced and the accuracy of frequency domain analysis can be improved. Hanning window Hamming Window Blackman Window The processed signal is subjected to fast Fourier transform calculation to obtain the frequency domain signal X(k), where k is the frequency index ranging from 0 to N-1. The frequency domain signal X(k) contains the amplitude and phase information of each frequency component. The frequency domain signal X(k) is calculated to analyze the compaction degree of high liquid limit soil, identify anomalies or deficiencies in the compaction process, and provide corresponding adjustment suggestions to improve the accuracy and reliability of the compaction analysis data.

[0076] See also Figure 2The compaction effect prediction module includes a parameter calculation submodule, an effect estimation submodule and a prediction value generation submodule, wherein:

[0077] Parameter calculation submodule: Based on historical compaction data and compaction analysis reports, the module corrects and matches the data, optimizes the parameter configuration iteratively through convolutional neural networks, performs parameter sensitivity analysis, synchronously updates the operating parameters, and generates parameter optimization solutions;

[0078] Effect estimation submodule: Based on the parameter optimization scheme, execute the simulation prediction process, adjust the simulation input variables, estimate the compaction effect under different operating conditions through iterative testing, statistically analyze the simulation output, and generate an effect estimation report;

[0079] Prediction value generation submodule: Based on the effect estimation report, organize and analyze the simulation data, form a compaction effect prediction through data fusion technology, output the prediction results, and establish the compaction effect prediction value;

[0080] In the parameter calculation submodule, based on historical compaction data and compaction analysis reports, data matching technology is used to iteratively optimize parameter configuration through convolutional neural networks, and parameter sensitivity analysis is performed using multi-layer perceptrons in the network. The operating parameters are updated synchronously, the number of iterations is set to 100 times, the learning rate is set to 0.01, and the ReLU activation function is used to optimize nonlinear data expression to generate parameter optimization solutions;

[0081] In the effect estimation submodule, based on the parameter optimization scheme, the simulation prediction process is executed, the simulation input variables are adjusted, the compaction effect under different operating conditions is estimated through iterative testing, the Monte Carlo simulation method is used for multi-scenario analysis, the random seed is set to 42 to keep the results reproducible, the simulation output is statistically analyzed, and the effect estimation report is generated;

[0082] In the prediction value generation submodule, based on the effect estimation report, the simulation data is sorted and analyzed, and the compaction effect prediction is formed through data fusion technology. The weighted average algorithm is used to fuse different data sources, and the weight ratio is set to automatically adjust according to the data reliability. The prediction result is output and the compaction effect prediction value is established.

[0083] Convolutional neural network, according to the formula:

[0084] y=f(W·x+b+α·z+β·v)

[0085] Where: y is the output value, f is the activation function, W is the weight matrix, x is the input vector, b is the bias value, α is the adaptive adjustment coefficient, z is the external environment parameter, β is the adjustment coefficient, and v is the time series parameter;

[0086] Execution process: The corrected and matched data is used as the input vector x, and the activation function f (such as ReLU function: f(z) = max(0,z)) is used to perform nonlinear transformation on the input data. The weight matrix W and the bias value b are initialized and updated according to the training process of the convolutional neural network. The adaptive adjustment coefficient α and the external environment parameter z are combined with the current environment to correct and adjust the model. The adjustment coefficient β and the time series parameter v are used to consider the impact of time on the model output and ensure the sensitivity of the model to time changes. The corrected and matched data x is input into the convolutional neural network, the weight matrix W and the bias value b are initialized, and updated through the back propagation algorithm. The external environment parameter z is measured according to the current environment, and the model is corrected by the adaptive adjustment coefficient α. Considering the time series parameter v, the time factor is introduced into the model by the adjustment coefficient β, and the output value of the model is adjusted. When determining the weight matrix W and the adjustment coefficient β, the cross-validation method and the grid search algorithm are used. The cross-validation method is used to divide the training set and the validation set, evaluate the performance of the model on different data sets, select the optimal parameter configuration, ensure the generalization ability and stability of the model, generate a parameter optimization plan, and ensure the robustness and accuracy of the convolutional neural network under complex environments and time series changes.

[0087] See also Figure 2 The compaction adjustment decision module includes a compaction effect analysis submodule, a parameter adjustment submodule and an operation optimization suggestion submodule, among which:

[0088] Compaction effect analysis submodule: Based on the compaction effect prediction value, compare the real-time compaction operation data, analyze the deviation size and direction, subdivide various deviations, identify key deviation areas, summarize the deviation causes, and generate a deviation degree assessment;

[0089] Parameter adjustment submodule: Based on the deviation degree assessment, by screening the range and value of the adjustment parameters, using the optimization model adjustment strategy, determining the specific adjustment value, evaluating the potential adjustment effect, and forming a complete parameter adjustment plan, the adjustment strategy details are obtained;

[0090] Operation optimization suggestion submodule: Based on the adjustment strategy details, formulate operation suggestions, format them into executable operation instructions, and directly transmit them to the operating system through the interface to obtain operation optimization suggestions;

[0091] In the compaction effect analysis submodule, based on the compaction effect prediction value, the real-time compaction operation data is compared, and the deviation size and direction are analyzed using statistical analysis methods. The variance analysis technology is used to subdivide various deviations. The clustering algorithm is used to identify the key deviation areas. The decision tree model is used to summarize the causes of the deviations. The number of clusters is set to 5, and the Gini coefficient is used as the splitting standard to generate a deviation degree assessment.

[0092] In the parameter adjustment submodule, based on the deviation degree assessment, by screening the range and value of the adjustment parameters, the genetic algorithm is used to optimize the model adjustment strategy, and the specific adjustment value is determined. The population size is set to 50, the number of iterations is set to 100, the crossover rate is set to 0.7, and the mutation rate is set to 0.01. The potential adjustment effect is evaluated, a complete set of parameter adjustment plans is formed, and the adjustment strategy details are obtained;

[0093] In the operation optimization suggestion submodule, operation suggestions are formulated based on the adjustment strategy details, formatted into executable operation instructions using the rule engine method, data encapsulation is performed, interface parameters are set including instruction type, execution priority, and expected execution time, and instructions are serialized using the JSON format. The security of data transmission is ensured through the HTTPS protocol and directly transmitted to the operating system to obtain operation optimization suggestions.

[0094] See also Figure 2 The feedback execution module includes a parameter adjustment submodule, a real-time monitoring submodule and an operation recording submodule, wherein:

[0095] Parameter adjustment submodule: Based on the operation optimization suggestions, the operation parameters of the compaction machinery are adjusted. By adjusting the power, speed and compaction depth of the compaction machinery and recording the adjusted parameter settings, the optimized operation parameters are generated.

[0096] Real-time monitoring submodule: Based on the optimized operating parameters, real-time monitoring is carried out. The operating data of the compaction machinery, including temperature, vibration frequency, compaction effect, etc., are collected through sensors, and the changing trends are analyzed to generate real-time monitoring data.

[0097] Operation record submodule: Based on real-time monitoring data, it records the operation data in real time, records the operation data of each time period, including equipment operation status, parameter adjustment records and compaction effect records, and generates real-time operation feedback;

[0098] In the parameter adjustment submodule, based on the operation optimization suggestions, the operation parameters of the compaction machinery are adjusted. The power, speed and compaction depth of the compaction machinery are set by adjusting the control algorithm. The feedback loop is adjusted using the PID controller. The PID parameters are set to include a proportional coefficient of 0.02, an integral coefficient of 0.01, and a differential coefficient of 0.005. The adjusted parameter settings are recorded to generate optimized operation parameters.

[0099] In the real-time monitoring submodule, real-time monitoring is carried out based on the optimization of operating parameters. Multi-sensor data acquisition technology is used to collect the operation data of the compaction machinery, including temperature, vibration frequency, compaction effect, etc. The time series analysis method is used to analyze the change trend. The moving average and exponential smoothing methods are used to process the data. The time window is set to 30 seconds to generate real-time monitoring data.

[0100] In the operation record submodule, real-time recording of operation data is performed based on real-time monitoring data. Database transaction processing technology is used to record the operation data of each time period, involving equipment operating status, parameter adjustment records and compaction effect records. Structured query language is used to insert and update commands to ensure data consistency and generate real-time operation feedback.

[0101] See also Figure 2 ,The long-term performance monitoring module includes a data analysis submodule, a problem identification submodule, and a performance evaluation submodule, where:

[0102] Data analysis submodule: Based on real-time operation feedback, it conducts long-term trend analysis of data. By summarizing the operation data collected over a long period of time, it generates trend charts, analyzes the changing trends of data, identifies potential abnormalities, and generates long-term trend data.

[0103] Problem identification submodule: Identifies potential performance problems based on long-term trend data, detects deviations and anomalies by comparing actual performance with expected standards, marks performance problem areas, and generates a list of potential problems;

[0104] Performance evaluation submodule: Based on the potential problem list, conduct performance evaluation, confirm the severity and impact of the problem through detailed inspection and testing, record specific performance indicators and evaluation results, and generate a performance evaluation report;

[0105] In the data analysis submodule, based on real-time operation feedback, long-term trend analysis of data is performed. The time series analysis method is used to summarize the operation data collected over a long period of time, and the autoregressive moving average model is applied to process the data points. The model parameters are set to autoregressive term p = 2 and moving average term q = 2. The trend of data changes is analyzed, and the outlier detection algorithm is used to identify potential anomalies to generate long-term trend data.

[0106] In the problem identification submodule, potential performance problems are identified based on long-term trend data. The variance analysis method is used to compare the actual performance with the expected standard. The F statistic is calculated and the significance level is set to 0.05 to detect statistically significant deviations in the performance data. At the same time, multiple regression analysis is used to determine the key factors affecting performance, mark the performance problem areas that deviate from expectations, and generate a list of potential problems.

[0107] In the performance evaluation submodule, performance evaluation is carried out based on the list of potential problems. The failure mode and effect analysis method is used to check and test each potential problem in detail. The problem is scored according to severity, probability of occurrence and difficulty of detection. The risk priority number is calculated to confirm the severity and scope of impact of the problem, record specific performance indicators and evaluation results, and generate a performance evaluation report.

[0108] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A system for monitoring and adjusting the compaction degree of layered filling of high liquid limit soil. It is characterized in that The system comprises: Initial compaction data collection module: Based on the position sensor and humidity sensor carried by the drone, the drone is used for efficient monitoring to collect the specific position and humidity value of the starting soil layer, and record the operating parameters of the initial compaction machinery to generate baseline compaction data records; Real-time compaction data analysis module: Based on the baseline compaction data record, the module uses fast Fourier transform to process the data, analyze the specific compaction degree and moisture changes of the soil, and compares the preset compaction standard to generate compaction analysis results; Compaction effect prediction module: through the compaction degree analysis results, using convolutional neural network, combined with historical compaction data, calculate and predict the optimal parameters of future compaction operations, and estimate the compaction effect that may be achieved, and generate a compaction effect prediction value; Compaction adjustment decision module: based on the compaction effect prediction value, analyzes the deviation between the current compaction operation and the preset standard, proposes a parameter adjustment plan, and directly transmits the adjustment suggestion to the operating system to generate an operation optimization suggestion; Feedback execution module: adjusts the operating parameters of the compaction machinery through the operation optimization suggestions, monitors the real-time execution status after the adjustment, and performs real-time recording of the operation data to generate real-time operation feedback; Long-term performance monitoring module: Based on the real-time operation feedback, analyze the long-term trend of the data, identify potential performance problems, conduct performance evaluation, and generate a performance evaluation report.

2. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The initial compaction data acquisition module includes a position information acquisition submodule, a humidity detection submodule and an operation parameter recording submodule, wherein: Position information acquisition submodule: Use the position sensor on the compaction machinery to capture the specific position information of the soil layer in real time, perform signal processing and calibration, calibrate the position data, record the coordinate points, and generate geographic coordinate information; Humidity detection submodule: monitors soil humidity level through humidity sensors, calibrates and standardizes humidity data, summarizes humidity index, and generates humidity data records; Operation parameter recording submodule: records the operation parameters of the compaction machinery, including pressure and speed, integrates the geographic coordinate information and moisture data records, integrates the compaction condition information, draws a baseline compaction data chart, and establishes a baseline compaction data archive.

3. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The real-time compaction data analysis module includes a data synchronization submodule, a compaction degree analysis submodule and a standard comparison submodule, wherein: Data processing submodule: Based on the baseline compaction data records collected by the drone, perform data cleaning, remove obvious errors and records that do not conform to the format due to environmental interference or sensor errors, process missing data points that may appear in complex terrain or signal interference, and generate optimized data sets; Compactness analysis submodule: Based on the optimized data set, compactness analysis is performed, data is smoothed by moving average method, noise in the data collected by the drone is reduced, thresholds are set, significant compactness changes are identified, compactness information is extracted by adaptive filtering and time-frequency analysis techniques, and the reflected wave signal obtained by the drone is processed by fast Fourier transform to obtain compactness analysis data; Standard comparison submodule: Based on the compaction analysis data, a difference calculation method is used to compare the difference between the actual compaction data and the preset compaction standard, and the key deviation points are determined through the difference to obtain the standard comparison results.

4. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The fast Fourier transform is according to the formula: Where: X(k) represents the frequency domain signal, x(n) represents the time domain signal, N is the number of sampling points, k is the frequency index, j is the imaginary unit, and w(n) is the window function.

5. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The compaction effect prediction module includes a parameter calculation submodule, an effect estimation submodule and a prediction value generation submodule, wherein: Parameter calculation submodule: based on the historical compaction data and compaction analysis report, the data is corrected and matched, the parameter configuration is iteratively optimized through the convolutional neural network, and the parameter sensitivity analysis is performed, the operation parameters are synchronously updated, and the parameter optimization scheme is generated; Effect estimation submodule: Based on the parameter optimization scheme, execute the simulation prediction process, adjust the simulation input variables, estimate the compaction effect under different operating conditions through iterative testing, statistically analyze the simulation output, and generate an effect estimation report; Prediction value generation submodule: Based on the effect estimation report, organize and analyze the simulation data, form a compaction effect prediction through data fusion technology, output the prediction result, and establish the compaction effect prediction value.

6. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The convolutional neural network, according to the formula: y=f(W·x+b+α·z+β·v) Where: y is the output value, f is the activation function, W is the weight matrix, x is the input vector, b is the bias value, α is the adaptive adjustment coefficient, z is the external environment parameter, β is the adjustment coefficient, and v is the time series parameter.

7. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The compaction adjustment decision module includes a compaction effect analysis submodule, a parameter adjustment submodule and an operation optimization suggestion submodule, wherein: Compaction effect analysis submodule: based on the compaction effect prediction value, compare the real-time compaction operation data, analyze the deviation size and direction, subdivide various deviations, identify key deviation areas, summarize the deviation causes, and generate a deviation degree assessment; Parameter adjustment submodule: Based on the deviation degree assessment, by screening the range and value of the adjustment parameters, adopting the optimization model adjustment strategy, determining the specific adjustment value, evaluating the potential adjustment effect, and forming a complete parameter adjustment plan, the adjustment strategy details are obtained; Operation optimization suggestion submodule: Based on the adjustment strategy details, formulate operation suggestions, format them into executable operation instructions, and transmit them directly to the operating system through the interface to obtain operation optimization suggestions.

8. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The feedback execution module includes a parameter adjustment submodule, a real-time monitoring submodule and an operation recording submodule, wherein: Parameter adjustment submodule: based on the operation optimization suggestion, adjust the operation parameters of the compaction machinery, by adjusting the power, speed and compaction depth of the compaction machinery, and recording the adjusted parameter settings to generate optimized operation parameters; Real-time monitoring submodule: Based on the optimized operating parameters, real-time monitoring is performed, and the operating data of the compaction machinery, including temperature, vibration frequency, compaction effect, etc., is collected through sensors, and the change trend is analyzed to generate real-time monitoring data; Operation record submodule: Based on the real-time monitoring data, real-time recording of operation data is performed, recording operation data for each time period, including equipment operation status, parameter adjustment records and compaction effect records, and generating real-time operation feedback.

9. The high liquid limit soil layer filling compaction degree monitoring and adjustment system according to claim 1, It is characterized in that The long-term performance monitoring module includes a data analysis submodule, a problem identification submodule and a performance evaluation submodule, wherein: Data analysis submodule: Based on the real-time operation feedback, long-term trend analysis of data is performed, by summarizing the operation data collected over a long period of time, generating trend charts, analyzing the changing trends of data, identifying potential abnormalities, and generating long-term trend data; Problem identification submodule: Based on the long-term trend data, potential performance problems are identified, deviations and anomalies are detected by comparing actual performance with expected standards, performance problem areas are marked, and a list of potential problems is generated; Performance evaluation submodule: Based on the potential problem list, a performance evaluation is conducted to confirm the severity and scope of the problem through detailed inspection and testing, record specific performance indicators and evaluation results, and generate a performance evaluation report.

Citation Information

Cited By

  • Multi-layer soft soil foundation settlement prediction method

    CN121834991A