Anchor rod tensioning force simulation test method and system

Abstract

The invention discloses an anchor rod tensioning force simulation test method and system, and the method comprises the steps: obtaining anchor rod material characteristics and application environment data, determining an anchor rod tensioning force qualified range through an intelligent data analysis algorithm, and obtaining tensioning force qualified range data; a high-precision sensor is adopted for collecting real-time data of anchor rod tensioning force, and the real-time monitoring data of the tensioning force are obtained through preprocessing of a data collecting system; the tensioning force qualified range data and the tensioning force real-time monitoring data are compared, whether the anchor rod tensioning force exceeds the qualified range or not is judged, if yes, a tensioning force adjusting mechanism is triggered, and if not, the current tensioning force is maintained; when a tensioning force adjusting mechanism is triggered, the tensioning force adjusting amplitude and speed are determined through an intelligent data analysis algorithm according to anchor rod material characteristics and application environment data, and tensioning force adjusting strategy data are obtained.

Description

Technical Field

[0001] The invention belongs to the field of information technology, and in particular relates to an anchor tension force simulation test method and system. Background Art

[0002] In the simulation test of anchor tension, there is a technical contradiction: how to monitor the tension in real time during the test, and judge and adjust it according to whether the tension is within the qualified range. The core of this contradiction lies in the fact that the material properties and application environment of the anchor are complex and changeable, which makes it difficult to accurately set the qualified range of tension. At the same time, real-time monitoring of tension requires high-precision sensors and data acquisition systems, and judging whether the tension is qualified requires intelligent data analysis algorithms. In addition, during the test, if the tension is found to be unqualified, the tension needs to be adjusted in time, but the adjustment range and speed also need to be determined according to the material properties and application environment of the anchor, otherwise it may cause damage to the anchor or the tension may be unqualified again. This requires integrating the material property database and the application environment database in the tension simulation test system, and dynamically adjusting the qualified range of tension and the tension adjustment strategy based on these data to ensure the accuracy and reliability of the test. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for simulating and testing the tension force of an anchor rod.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A method for simulating the tension force of an anchor bolt, comprising:

[0006] Obtain the anchor material properties and application environment data, determine the qualified range of anchor tension force, and obtain the qualified range data of tension force;

[0007] Collect real-time data of anchor tension force and obtain real-time monitoring data of tension force;

[0008] Compare the qualified range data of tension force with the real-time monitoring data of tension force to determine whether the tension force of the anchor bolt exceeds the qualified range. If it exceeds the qualified range, the tension force adjustment mechanism is triggered; otherwise, the current tension force is maintained;

[0009] When the tension adjustment mechanism is triggered, the tension adjustment amplitude and speed are determined according to the anchor material characteristics and application environment data, and the tension adjustment strategy data is obtained;

[0010] According to the tension adjustment strategy data, the tension adjustment actuator is controlled to apply the adjusted tension to the anchor rod, and the tension data of the adjusted anchor rod is collected to obtain the real-time monitoring data after the tension adjustment;

[0011] The optimized algorithm is applied to subsequent anchor tension force simulation tests through iterative training.

[0012] Preferably, obtaining the qualified range data of the tension force includes:

[0013] Obtain the mechanical properties data of anchor materials and the application environment load data, and establish a correlation model between material properties and environmental factors;

[0014] Train the correlation model between material properties and environmental factors to obtain an optimized prediction model;

[0015] Analyze the stress conditions of different materials in different environments based on the prediction model and determine the applicable scope of the materials;

[0016] Analyze the stress distribution of the anchor rod under different tension forces and determine the ultimate bearing capacity of the anchor rod;

[0017] Based on the applicable scope of the material and the ultimate bearing capacity of the anchor rod, the qualified range of the anchor rod tension force is determined, and the qualified range data of the tension force is obtained.

[0018] Preferably, real-time monitoring data of tension force is obtained, including:

[0019] Preprocess the real-time anchor tension force data, including data cleaning, outlier removal and data normalization, to obtain standardized anchor tension force data;

[0020] Model the pre-processed real-time monitoring data of anchor tension force and extract the time domain and frequency domain characteristics of the data;

[0021] According to the characteristics of anchor tension data, a support vector machine model is trained to determine whether the anchor tension is abnormal.

[0022] If an abnormality is detected, an early warning mechanism is triggered;

[0023] The real-time monitoring data and analysis results of anchor tension force are displayed in the form of curve graphs and bar graphs;

[0024] The real-time monitoring data and analysis results of the anchor tension force are stored in the database to obtain the real-time monitoring data of the tension force.

[0025] Preferably, obtaining the tension adjustment strategy data includes:

[0026] According to the anchor material characteristic data and application environment data, an intelligent adjustment model of anchor tension force is established to determine the optimal tension force adjustment strategy;

[0027] By training historical data, the association rules between anchor material properties, environmental factors and tension adjustment range and speed are obtained;

[0028] Obtain anchor application environment data in real time to determine whether the conditions for triggering the tension adjustment mechanism are met. If so, start the intelligent adjustment process;

[0029] The real-time collected anchor material property data and environmental data are input into the trained machine learning model to predict the optimal tension adjustment amplitude and speed parameters.

[0030] Preferably, obtaining real-time monitoring data after the tension force is adjusted includes:

[0031] According to a pre-established tension adjustment strategy database, obtaining tension adjustment strategy data matching the current anchor state;

[0032] According to the obtained tension adjustment strategy data, the tension adjustment actuator is controlled to apply the adjusted tension to the target anchor bolt;

[0033] Use high-precision sensors to collect the tension data of the adjusted anchor rod in real time, and transmit the collected data to the data acquisition system;

[0034] Preprocessing the collected adjusted anchor tension force data, including data cleaning and data normalization;

[0035] Through data preprocessing, real-time monitoring data after tension adjustment is obtained.

[0036] As a preference, the optimized algorithm is applied to subsequent anchor tension simulation tests, including:

[0037] Clean and normalize the data based on historical test data;

[0038] According to the pre-processed historical test data, a model for determining the qualified range of anchor tension force is established by setting appropriate kernel functions and penalty factors;

[0039] Based on the pre-processed historical test data, a model for formulating anchor tension adjustment strategies is established by setting appropriate tree depth and number of leaf nodes;

[0040] Evaluate the established anchor tension force qualified range determination model and adjustment strategy formulation model, and determine whether the performance of the model meets the requirements by calculating indicators such as the model's accuracy and recall rate;

[0041] If the model performance does not meet the requirements, the model parameters are optimized and the optimal model parameters are searched by setting different parameter combinations;

[0042] The optimized anchor tension force qualified range judgment model and adjustment strategy formulation model are applied to the subsequent anchor tension force simulation test. According to the simulation test data, it is judged whether the anchor tension force is within the qualified range, and the corresponding adjustment strategy is formulated according to the judgment result.

[0043] In the anchor tension simulation test, the test data is continuously acquired and added to the historical test data to update and optimize the anchor tension qualified range determination model and the adjustment strategy formulation model.

[0044] The present invention also provides an anchor tension force simulation test system, comprising:

[0045] The first processing device is used to obtain the anchor material characteristics and application environment data, determine the qualified range of the anchor tension force, and obtain the qualified range data of the tension force;

[0046] The second processing device is used to collect real-time data of the tension force of the anchor rod and obtain real-time monitoring data of the tension force;

[0047] The third processing device is used to compare the tension force qualified range data with the tension force real-time monitoring data to determine whether the anchor tension force exceeds the qualified range. If it exceeds, the tension force adjustment mechanism is triggered, otherwise the current tension force is maintained;

[0048] The fourth processing device is used to determine the tension adjustment amplitude and speed according to the anchor material characteristics and application environment data when the tension adjustment mechanism is triggered, and obtain the tension adjustment strategy data;

[0049] A fifth processing device is used to control the tension adjustment actuator to apply the adjusted tension to the anchor rod according to the tension adjustment strategy data, and collect the tension data of the anchor rod after adjustment to obtain the real-time monitoring data after the tension adjustment;

[0050] The sixth processing device is used to apply the optimized algorithm to subsequent anchor tension force simulation tests through iterative training.

[0051] Preferably, the first processing device comprises:

[0052] The first processing module is used to obtain the mechanical properties data of the anchor material and the application environment load data, and establish a correlation model between the material properties and the environmental factors;

[0053] The second processing module is used to train the model of correlation between material properties and environmental factors to obtain an optimized prediction model;

[0054] The third processing module is used to analyze the stress conditions of different materials under different environments according to the prediction model and determine the applicable scope of the materials;

[0055] The fourth processing module is used to analyze the stress distribution of the anchor rod under different tension forces and determine the ultimate bearing capacity of the anchor rod;

[0056] The fifth processing module is used to determine the qualified range of the anchor tension force by combining the applicable range of the material and the ultimate bearing capacity of the anchor, and obtain the qualified range data of the tension force.

[0057] Preferably, the second processing device comprises:

[0058] The sixth processing module is used to pre-process the real-time anchor tension force data, including data cleaning, outlier removal and data normalization operations, to obtain standardized anchor tension force data;

[0059] The seventh processing module is used to model the pre-processed real-time monitoring data of anchor tension force and extract the time domain and frequency domain characteristics of the data;

[0060] An eighth processing module is used to train a support vector machine model according to the characteristics of the anchor tension force data, so as to determine whether the anchor tension force is abnormal;

[0061] A ninth processing module, for triggering an early warning mechanism if an abnormality is detected;

[0062] The tenth processing module is used to display the real-time monitoring data and analysis results of the anchor tension force in the form of a curve chart and a bar chart in real time;

[0063] The eleventh processing module is used to store the real-time monitoring data and analysis results of the anchor tension force in a database to obtain the real-time monitoring data of the tension force.

[0064] Preferably, the third processing device comprises:

[0065] The twelfth processing module is used to establish an intelligent adjustment model of the tension force of the anchor rod according to the material characteristic data of the anchor rod and the application environment data, so as to determine the optimal tension force adjustment strategy;

[0066] The thirteenth processing module is used to train the historical data to obtain the association rules between the anchor material characteristics, environmental factors and the tension force adjustment range and speed;

[0067] The fourteenth processing module is used to obtain the anchor application environment data in real time, determine whether the conditions for triggering the tension adjustment mechanism are met, and if so, start the intelligent adjustment process;

[0068] The fifteenth processing module is used to input the real-time collected anchor material property data and environmental data into the trained machine learning model to predict the optimal tension adjustment amplitude and speed parameters.

[0069] Preferably, the fourth processing device comprises:

[0070] A sixteenth processing module is used to obtain tension adjustment strategy data matching the current anchor state according to a pre-established tension adjustment strategy database;

[0071] A seventeenth processing module is used to control the tension adjustment actuator to apply the adjusted tension to the target anchor according to the obtained tension adjustment strategy data;

[0072] An eighteenth processing module is used to collect the tension force data of the adjusted anchor rod in real time using a high-precision sensor, and transmit the collected data to a data acquisition system;

[0073] A nineteenth processing module is used to pre-process the collected adjusted anchor tension force data, including data cleaning and data normalization;

[0074] The twentieth processing module is used to obtain real-time monitoring data after the tension force is adjusted through data preprocessing.

[0075] The beneficial effects of the present invention are as follows:

[0076] The method of the present invention obtains the material properties of the anchor rod and the application environment data, uses an intelligent data analysis algorithm to determine the qualified range of tension force, and monitors the tension force of the anchor rod in real time. When it is detected that the tension force exceeds the qualified range, the adjustment mechanism is automatically triggered, and the adjustment strategy is intelligently calculated according to the material properties and environmental data to control the actuator to make precise adjustments. The present invention also adopts a machine learning algorithm to continuously optimize the data analysis algorithm through iterative training to improve the accuracy of tension force range determination and adjustment strategy formulation. This intelligent monitoring and adjustment method can realize real-time monitoring and automatic adjustment of the tension force of the anchor rod, ensure that the anchor rod is always in the best working state, improve the safety and reliability of the use of the anchor rod, and is of great significance to engineering construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 The present invention is a flow chart of the anchor tension force simulation test method. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0079] Embodiment 1:

[0080] like Figure 1 As shown, an embodiment of the present invention provides a method for simulating the tension force of an anchor rod, comprising:

[0081] S101. Obtain the anchor material properties and application environment data, determine the acceptable range of anchor tension force through intelligent data analysis algorithm, and obtain the acceptable range data of tension force.

[0082] Obtain the mechanical properties data of anchor materials and the application environment load data, and establish a model for associating material properties with environmental factors. Use machine learning algorithms to train the model for associating material properties with environmental factors to obtain an optimized prediction model. According to the prediction model, analyze the stress conditions of different materials in different environments and determine the applicable scope of the material. Analyze the stress distribution of anchors under different tensions through finite element simulation to determine the ultimate bearing capacity of anchors. Combine the applicable scope of the material and the ultimate bearing capacity of the anchor to determine the qualified range of the anchor tension and obtain the qualified range data of the tension. Use a deep learning algorithm to analyze the qualified range data of the tension and extract key characteristic parameters. Input the extracted key characteristic parameters into the intelligent data analysis model, dynamically optimize the qualified range of the anchor tension, and obtain an accurate tension control threshold.

[0083] Furthermore, the mechanical properties of 100 groups of anchors of different materials were tested to obtain data such as tensile strength and yield strength, and environmental temperature, humidity, pH value and other factors were recorded at the same time. The multivariate linear regression model was used to establish the correlation between material properties and environmental factors, and the gradient descent algorithm was used to train and optimize the model, and the average error was reduced to less than 5%. Based on the trained prediction model, the stress conditions of common materials such as Q345 and HRB400 under different environments were analyzed, and the performance was stable in the temperature range of -40℃~60℃ and humidity range of 30%~95%. Using ANSYS finite element software, different tensile forces of 100kN~1000kN were applied to the Φ28mm anchor, and the stress distribution cloud map was analyzed. When the stress exceeded the yield strength of 500MPa, the anchor was significantly deformed, and the ultimate bearing capacity was determined to be 800kN. Comprehensively considering the applicable scope of the material and the ultimate bearing capacity of the anchor, it was preliminarily determined that the qualified range of tensile force was 100kN~750kN, and 2000 groups of data were generated. The convolutional neural network algorithm is used to extract features from the data, and 10 key parameters such as tension, material properties, and environmental factors are extracted. The key parameters are input into the intelligent data analysis model based on the support vector machine. Through cross-validation and grid search, the model parameters are continuously optimized, and the tension control threshold is finally determined to be 150kN~650kN. When the ambient temperature is 20℃~40℃ and the humidity is 50%~70%, the anchor performance is optimal, and the tension threshold can be relaxed to 200kN~700kN.

[0084] A deep learning algorithm is used to analyze the qualified range data of tension force and extract key characteristic parameters.

[0085] Obtain a set of historical tensile test data, including tensile values ​​and corresponding pass / fail labels; preprocess the tensile data to remove outliers and noise, and perform necessary feature engineering; use convolutional neural networks to model the tensile data and learn the correlation between tensile values ​​and passability; obtain a tensile pass prediction model through training, which can judge whether it is passable based on the input tensile value; extract key feature parameters from the trained model, such as tensile mean, variance and other statistical indicators; store the extracted key feature parameters as tensile pass criteria for subsequent rapid judgment; when new tensile test data is input, extract its key feature parameters and compare them with the criteria to determine passability.

[0086] Furthermore, 10,000 groups of anchor tension test data were randomly extracted from the historical database, each group of data contained the measured tension value and the corresponding manually determined qualified label. The tension data was preprocessed, the 3σ principle was used to remove outliers, and the data was normalized to map the tension value to the interval of 0 to 1. Eight statistical quantities such as tension mean, variance, and kurtosis were extracted as feature vectors. A convolutional neural network model was built, with the tension feature vector as input, and deep features were extracted through three convolutional layers, and then mapped to two output nodes through two fully connected layers, representing the probabilities of qualified and unqualified, respectively. The model was trained using the cross entropy loss function and the Adam optimization algorithm, and the accuracy on the validation set reached 95% after 500 iterations. The kernel parameters of each convolutional layer and the weight matrix of the fully connected layer were extracted from the trained model as the tension qualified criterion. When new tension test data was input, its eight statistical quantity features were first extracted, and then the qualified probability output was obtained through the forward calculation of the convolutional layer and the fully connected layer in turn. If the probability of passing is greater than 95%, it is judged as passing, otherwise it is judged as failing. Through this intelligent judgment, the qualification judgment of a single set of tension data can be completed within 100 milliseconds, greatly improving the judgment efficiency and accuracy.

[0087] S102. Use high-precision sensors to collect real-time data on the tension force of the anchor rods, and obtain real-time monitoring data of the tension force through preprocessing by the data acquisition system.

[0088] According to the technical parameters of the high-precision sensor, the sampling frequency and range of the sensor are determined to meet the accuracy requirements of real-time acquisition of anchor tension. The raw data of anchor tension is collected in real time by the sensor and the data is transmitted to the data acquisition system. In the data acquisition system, the raw data collected by the sensor is preprocessed, including data cleaning, outlier removal, data normalization and other operations to obtain standardized anchor tension data. The time series analysis algorithm is used to model the preprocessed real-time monitoring data of anchor tension, and the time domain and frequency domain characteristics of the data are extracted. According to the characteristics of the anchor tension data, the support vector machine model is trained to determine whether the anchor tension is abnormal. If an abnormality is detected, the early warning mechanism is triggered. The data visualization technology is used to display the real-time monitoring data and analysis results of anchor tension in real time in the form of curves, bar charts, etc., so that engineering personnel can intuitively grasp the stress state of the anchor. The real-time monitoring data and analysis results of anchor tension are stored in the database to form big data of anchor tension, which provides data support for subsequent anchor health assessment and life prediction.

[0089] Furthermore, according to the technical parameters of the high-precision sensor, the sampling frequency of the sensor is determined to be 1000Hz and the range is 0-100kN to meet the accuracy requirements of real-time acquisition of anchor tension. The raw data of anchor tension is collected in real time by the sensor, and the data is transmitted to the data acquisition system with a period of 10ms. In the data acquisition system, the median filter algorithm is used to pre-process the raw data collected by the sensor, and the abnormal values ​​exceeding the range of ±10% are eliminated, and the data is normalized to between 0-1 to obtain the standardized anchor tension data. The wavelet transform algorithm is used to model the pre-processed real-time monitoring data of anchor tension, and the time domain characteristics of the data such as mean, variance and frequency domain characteristics such as wavelet energy spectrum are extracted. According to the characteristics of the anchor tension data, the support vector machine model is trained, and the model parameters are optimized by grid search. When the accuracy of the test set reaches more than 95%, it is used to judge whether the anchor tension is abnormal. If an abnormality is detected, the early warning mechanism is triggered and the relevant personnel are notified by SMS and email. Echarts is used for data visualization, and the real-time monitoring data and analysis results of anchor tension are displayed in real time in the form of line charts, bar charts, etc., and the spatial distribution of anchor tension is displayed in a heat map, which is convenient for engineering personnel to intuitively grasp the stress state of the anchor. The real-time monitoring data and analysis results of anchor tension are stored in the MySQL database to form big data of anchor tension. Hadoop is used to perform distributed storage and calculation of data to provide data support for subsequent anchor health assessment and life prediction.

[0090] The time series analysis algorithm is used to build a model for the preprocessed real-time monitoring data of anchor tension force, and to extract the time domain and frequency domain characteristics of the data.

[0091] Obtain real-time monitoring data of anchor tension, preprocess the data, remove outliers and noise data, and obtain preprocessed monitoring data. According to the preprocessed monitoring data, use the time series analysis algorithm to model the data and build a time series prediction model. Through the time series prediction model, extract the time domain characteristics of the monitoring data, including statistical characteristics such as mean, variance, and kurtosis. Perform frequency domain analysis on the preprocessed monitoring data, and use methods such as Fourier transform to extract the frequency domain characteristics of the data, such as frequency and amplitude. Fuse the extracted time domain features and frequency domain features to construct a feature vector as the input of the algorithm model. Use machine learning algorithms such as support vector machines to train the feature vector and establish an anchor tension prediction model. Input the new real-time monitoring data into the prediction model to determine whether the anchor tension is abnormal. If it exceeds the preset threshold, an early warning is triggered to determine the abnormal situation of the anchor tension.

[0092] Furthermore, in actual engineering applications, the anchor tension data is first collected at a frequency of 10 times per second by high-precision sensors to ensure the real-time and accuracy of the data. The collected raw data is preliminarily processed by the data acquisition system, and the sliding window method is used to remove outliers beyond the range of ±3σ (standard deviation). At the same time, a low-pass filter is applied to remove high-frequency noise to ensure the smoothness and reliability of the data. The autoregressive moving average (ARMA) model is used for time series analysis of the preprocessed data. According to the statistical characteristics of the mean, variance and kurtosis of the historical data, a prediction model is established to capture the time dependence of the data. In addition, the data is subjected to a fast Fourier transform (FFT) to analyze its frequency domain characteristics, such as the main frequency components and the corresponding amplitudes, which helps to identify periodic mechanical behaviors. The time domain and frequency domain features are combined to form a comprehensive feature vector. This feature vector is used as the input of the support vector machine (SVM) model, which has been trained with historical data sets and can effectively predict whether the anchor tension will be abnormal. Once real-time monitoring data shows that the tension force exceeds the preset safety threshold, the system will automatically trigger an early warning and promptly notify maintenance personnel to conduct inspections to ensure the safety of the structure.

[0093] S103, comparing the qualified range data of the tension force with the real-time monitoring data of the tension force to determine whether the tension force of the anchor rod exceeds the qualified range, and if so, triggering the tension force adjustment mechanism, otherwise maintaining the current tension force.

[0094] The real-time monitoring data of the anchor tension force and the preset qualified range data are obtained, and the real-time monitoring data are compared with the qualified range data. If the real-time monitoring data exceeds the qualified range, the tension adjustment mechanism is triggered, and the target value of the adjusted tension is calculated using the PID control algorithm according to the difference value beyond the range. The adjusted target tension value is sent to the tension actuator, and the tension of the anchor is adjusted by adjusting the extension amount of the tension actuator until the tension value returns to the qualified range. If the real-time monitoring data does not exceed the qualified range, it is determined whether the current tension is near the middle value of the qualified range, and the fuzzy control algorithm is used to determine whether fine-tuning is required. If fine-tuning of the tension is required, the fine-tuned tension target value is calculated according to the fuzzy control algorithm, and the fine-tuned target tension value is sent to the tension actuator for fine-tuning. If it is determined that the current tension does not need to be fine-tuned, the current extension amount of the tension actuator is maintained to keep the anchor tension at the current value. Execute the above steps in a continuous cycle to monitor and adjust the anchor tension in real time to ensure that the anchor tension is always within the acceptable range and to ensure the normal stress and safe use of the anchor.

[0095] Furthermore, in the process of real-time monitoring of the anchor tension, the anchor tension data is first obtained through a high-precision sensor. Assuming that the sampling frequency of the sensor is 100Hz, the real-time and accuracy of the data can be ensured. When the tension data read by the sensor is 1500kN, it is compared with the preset qualified range (for example, 1400kN to 1600kN). At this time, the data is within the qualified range. The system uses the fuzzy control algorithm to analyze whether the current tension value is close to the middle value of 1500kN. If the current value is 1498kN, the fuzzy control algorithm determines that no adjustment is required. If the real-time monitoring data shows 1700kN, which exceeds the qualified range, the system triggers the tension adjustment mechanism. The PID control algorithm is used to calculate the target tension value that needs to be adjusted. Assume that the target value calculated by the PID algorithm is 1580kN. This target value will be sent to the tension actuator, which reduces the tension by adjusting the expansion and contraction amount until the tension value returns to the qualified range. During the entire process, the system continuously monitors the anchor tension force in real time to ensure that each data collection and processing is accurate and correct, thus guaranteeing the safety of the anchor and the stability of its functions.

[0096] S104. When the tension adjustment mechanism is triggered, the tension adjustment amplitude and speed are determined according to the anchor material characteristics and application environment data through an intelligent data analysis algorithm to obtain tension adjustment strategy data.

[0097] According to the anchor material characteristic data and application environment data, an intelligent adjustment model of anchor tension force is established to determine the optimal tension force adjustment strategy. The historical data is trained by machine learning algorithm to obtain the association rules between anchor material characteristics, environmental factors and tension force adjustment amplitude and speed. The anchor application environment data is obtained in real time to determine whether the conditions for triggering the tension force adjustment mechanism are met. If so, the intelligent adjustment process is started. The real-time collected anchor material characteristic data and environmental data are input into the trained machine learning model to predict the optimal tension force adjustment amplitude and speed parameters. According to the adjustment parameters output by the machine learning model, the tension force adjustment actuator is controlled to realize the intelligent adaptive adjustment of the anchor tension force. During the tension force adjustment process, the anchor force state data is continuously collected, the adjustment effect is evaluated in real time, and the parameters are fine-tuned when necessary to achieve the optimal control performance. The data records of each intelligent tension force adjustment are saved in the knowledge base to optimize the machine learning model and continuously improve the accuracy and reliability of tension force adjustment.

[0098] Furthermore, in order to establish an intelligent adjustment model for anchor tension, the elastic modulus, yield strength and other characteristic parameters of the anchor material, as well as the temperature, humidity, vibration frequency and other data of the application environment were first collected, totaling 1,000 sets of historical data. The support vector machine algorithm was used to train the data, and the average accuracy of the model reached 95% through 5-fold cross validation. In actual applications, the temperature, humidity and other data of the anchor environment are collected in real time through sensors. If the temperature exceeds 40°C for 5 consecutive minutes or the humidity exceeds 90% for 10 consecutive minutes, the tension adjustment mechanism is triggered. The collected anchor material characteristic data and environmental data are input into the trained support vector machine model. The model predicts that under the current environment, the optimal adjustment range of the anchor tension is 10N and the adjustment speed is 5N / s. According to the adjustment parameters output by the model, the control system drives the motor of the tension actuator to increase the tension at a speed of 5N / s until it stops after increasing by 10N. During the adjustment process, the sensor collects anchor force data at a cycle of 100ms. If the force value of three consecutive cycles deviates from the target value by less than 1%, the adjustment is considered complete; if the force value of three consecutive cycles deviates from the target value by more than 5%, the parameter fine-tuning is triggered, and 2N is increased or decreased based on the current adjustment range according to the direction and size of the deviation. The environmental parameters, material parameters, adjustment parameters, adjustment effects and other data of this adjustment process are recorded in the knowledge base in a fixed format for subsequent optimization of the machine learning model to improve the accuracy and reliability of intelligent adjustment.

[0099] S105. According to the tension adjustment strategy data, the tension adjustment actuator is controlled to apply the adjusted tension to the anchor rod, and a high-precision sensor is used to collect the adjusted anchor tension data. The data acquisition system pre-processes the data to obtain the real-time monitoring data after the fourth tension adjustment.

[0100] According to the pre-established tension adjustment strategy database, the tension adjustment strategy data matching the current anchor state is obtained; according to the obtained tension adjustment strategy data, the tension adjustment actuator is controlled to apply the adjusted tension to the target anchor; a high-precision sensor is used to collect the tension data of the adjusted anchor in real time, and the collected data is transmitted to the data acquisition system; in the data acquisition system, the collected adjusted anchor tension data is preprocessed, including data cleaning, data normalization, etc.; through data preprocessing, the real-time monitoring data after the tension adjustment is obtained; the real-time monitoring data after the tension adjustment is compared with the preset threshold value to determine whether the adjusted anchor tension meets the requirements; if the adjusted anchor tension does not meet the requirements, return to the second step, re-acquire the tension adjustment strategy data and perform adjustments until the adjusted anchor tension meets the requirements.

[0101] Furthermore, according to the pre-established tension adjustment strategy database, the tension adjustment strategy data matching the current anchor state is obtained through association rule mining algorithms, such as the Apriori algorithm. For example, when the current anchor tension is 80kN, the ambient temperature is 20℃, and the humidity is 60%, the optimal adjustment strategy obtained through association rule mining is to increase the tension by 10kN and the adjustment speed is 1kN / min. According to the obtained strategy data, the tension adjustment actuator is controlled, and the PID control algorithm is used to apply the adjusted tension to the target anchor. A tension sensor with an accuracy of 1% FS is used to collect the tension data of the adjusted anchor in real time, and transmit it to the data acquisition system through the RS485 bus. In the data acquisition system, the collected data is cleaned to remove dirty data that exceeds the measurement range and fluctuates abnormally, and then the minimum-maximum normalization method is used to normalize the data, and the data is mapped to the [0,1] interval to obtain the normalized real-time monitoring data after tension adjustment. The monitoring data is compared with the preset safety threshold (such as 90kN), and the logistic regression algorithm is used to determine whether the adjusted anchor tension meets the requirements. If it does not meet the requirements, the feedback control mechanism returns to the strategy data acquisition step, re-acquires the tension adjustment strategy data and performs adjustments until the tension meets the requirements. The entire adjustment process is automated and intelligently adjusted through closed-loop control to ensure that the anchor tension is always within the safety range, thereby improving the reliability and stability of anchor support.

[0102] S106. Based on the saved test data, the machine learning algorithm is used to optimize the intelligent data analysis algorithm. Through iterative training, the accuracy of determining the qualified range of tension force and formulating the adjustment strategy is improved. The optimized algorithm is applied to subsequent anchor tension force simulation tests to ensure the reliability and efficiency of the test.

[0103] According to the historical test data, data preprocessing technology is used to clean and normalize the data to eliminate noise and outliers in the data and improve data quality. For the preprocessed historical test data, the support vector machine algorithm is used to establish the anchor tension force qualified range judgment model. By setting appropriate kernel functions and penalty factors, the generalization ability and classification accuracy of the model are improved. For the preprocessed historical test data, the decision tree algorithm is used to establish the anchor tension force adjustment strategy formulation model. By setting appropriate tree depth and leaf node number, the model fitting ability and rationality of strategy formulation are improved. The cross-validation method is used to evaluate the established anchor tension force qualified range judgment model and adjustment strategy formulation model. By calculating the model's accuracy, recall rate and other indicators, it is judged whether the model performance meets the requirements. If the model performance does not meet the requirements, the grid search method is used to optimize the model's hyperparameters. By setting different parameter combinations, the optimal model parameters are searched to improve the model performance. The optimized anchor tension force qualified range judgment model and adjustment strategy formulation model are applied to the subsequent anchor tension force simulation test. According to the simulation test data, it is judged whether the anchor tension force is within the qualified range, and the corresponding adjustment strategy is formulated according to the judgment result. In the subsequent anchor tension force simulation test, the test data is continuously collected and added to the historical test data. The anchor tension force qualified range judgment model and adjustment strategy formulation model are updated and optimized by the incremental learning method to adapt to the new test data and improve the adaptability and robustness of the model.

[0104] Furthermore, in the data preprocessing stage, the Z-score normalization method is used to normalize the historical test data, and the data is mapped to a standard normal distribution with a mean of 0 and a variance of 1 to eliminate the dimensional influence of the data. At the same time, the box plot method is used to clean the data, and the data points that exceed the upper and lower quartiles by 5 times the interquartile range are regarded as outliers and removed to improve the data quality. When establishing the anchor tension qualified range judgment model, the radial basis kernel function is used as the kernel function of the support vector machine, and the penalty factor C=10 is set. The grid search method is used to search for the optimal gamma parameter, and the model has the highest accuracy and recall rate when gamma=1, reaching 95% and 93% respectively. When establishing the anchor tension adjustment strategy formulation model, the CART decision tree algorithm is used, and the maximum depth of the tree is set to 5, and the minimum number of leaf node samples is 10. The model performance is evaluated by the 5-fold cross-validation method, and the average accuracy of the model is 92%, and the average recall rate is 90%. The optimized model is applied to the anchor tension simulation test. The anchor tension is judged according to the test data whether it is within the qualified range. When the tension is less than 100kN, it is judged that the tension is insufficient and needs to be increased. When the tension is greater than 200kN, it is judged that the tension is too large and needs to be reduced. In the subsequent simulation test, the test data is continuously collected, and the AdaBoost algorithm is used to perform incremental learning on the model. When the amount of new test data reaches 1000, the model is optimized again to improve the adaptability and robustness of the model.

[0105] Embodiment 2:

[0106] The embodiment of the present invention provides an anchor tension simulation test system, comprising:

[0107] The first processing device is used to obtain the anchor material characteristics and application environment data, determine the qualified range of the anchor tension force, and obtain the qualified range data of the tension force;

[0108] The second processing device is used to collect real-time data of the tension force of the anchor rod and obtain real-time monitoring data of the tension force;

[0109] The third processing device is used to compare the tension force qualified range data with the tension force real-time monitoring data to determine whether the anchor tension force exceeds the qualified range. If it exceeds, the tension force adjustment mechanism is triggered, otherwise the current tension force is maintained;

[0110] The fourth processing device is used to determine the tension adjustment amplitude and speed according to the anchor material characteristics and application environment data when the tension adjustment mechanism is triggered, and obtain the tension adjustment strategy data;

[0111] A fifth processing device is used to control the tension adjustment actuator to apply the adjusted tension to the anchor rod according to the tension adjustment strategy data, and collect the tension data of the anchor rod after adjustment to obtain the real-time monitoring data after the tension adjustment;

[0112] The sixth processing device is used to apply the optimized algorithm to subsequent anchor tension force simulation tests through iterative training.

[0113] As an implementation of an embodiment of the present invention, the first processing device includes:

[0114] The first processing module is used to obtain the mechanical properties data of the anchor material and the application environment load data, and establish a correlation model between the material properties and the environmental factors;

[0115] The second processing module is used to train the model of correlation between material properties and environmental factors to obtain an optimized prediction model;

[0116] The third processing module is used to analyze the stress conditions of different materials under different environments according to the prediction model and determine the applicable scope of the materials;

[0117] The fourth processing module is used to analyze the stress distribution of the anchor rod under different tension forces and determine the ultimate bearing capacity of the anchor rod;

[0118] The fifth processing module is used to determine the qualified range of the anchor tension force by combining the applicable range of the material and the ultimate bearing capacity of the anchor, and obtain the qualified range data of the tension force.

[0119] As an implementation of an embodiment of the present invention, the second processing device includes:

[0120] The sixth processing module is used to pre-process the real-time anchor tension force data, including data cleaning, outlier removal and data normalization operations, to obtain standardized anchor tension force data;

[0121] The seventh processing module is used to model the pre-processed real-time monitoring data of anchor tension force and extract the time domain and frequency domain characteristics of the data;

[0122] An eighth processing module is used to train a support vector machine model according to the characteristics of the anchor tension force data, so as to determine whether the anchor tension force is abnormal;

[0123] A ninth processing module, for triggering an early warning mechanism if an abnormality is detected;

[0124] The tenth processing module is used to display the real-time monitoring data and analysis results of the anchor tension force in the form of a curve chart and a bar chart in real time;

[0125] The eleventh processing module is used to store the real-time monitoring data and analysis results of the anchor tension force in a database to obtain the real-time monitoring data of the tension force.

[0126] As an implementation of the embodiment of the present invention, the third processing device includes:

[0127] The twelfth processing module is used to establish an intelligent adjustment model of the tension force of the anchor rod according to the material characteristic data of the anchor rod and the application environment data, so as to determine the optimal tension force adjustment strategy;

[0128] The thirteenth processing module is used to train the historical data to obtain the association rules between the anchor material characteristics, environmental factors and the tension force adjustment range and speed;

[0129] The fourteenth processing module is used to obtain the anchor application environment data in real time, determine whether the conditions for triggering the tension adjustment mechanism are met, and if so, start the intelligent adjustment process;

[0130] The fifteenth processing module is used to input the real-time collected anchor material property data and environmental data into the trained machine learning model to predict the optimal tension adjustment amplitude and speed parameters.

[0131] As an implementation manner of the embodiment of the present invention, the fourth processing device includes:

[0132] A sixteenth processing module is used to obtain tension adjustment strategy data matching the current anchor state according to a pre-established tension adjustment strategy database;

[0133] A seventeenth processing module is used to control the tension adjustment actuator to apply the adjusted tension to the target anchor according to the obtained tension adjustment strategy data;

[0134] An eighteenth processing module is used to collect the tension force data of the adjusted anchor rod in real time using a high-precision sensor, and transmit the collected data to a data acquisition system;

[0135] A nineteenth processing module is used to pre-process the collected adjusted anchor tension force data, including data cleaning and data normalization;

[0136] The twentieth processing module is used to obtain real-time monitoring data after the tension force is adjusted through data preprocessing.

[0137] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for simulating the tension force of an anchor rod, characterized in that: include: Obtain the anchor material properties and application environment data, determine the qualified range of anchor tension force, and obtain the qualified range data of tension force; Collect real-time data of anchor tension force and obtain real-time monitoring data of tension force; Compare the qualified range data of tension force with the real-time monitoring data of tension force to determine whether the tension force of the anchor bolt exceeds the qualified range. If it exceeds the qualified range, the tension force adjustment mechanism is triggered; otherwise, the current tension force is maintained; When the tension adjustment mechanism is triggered, the tension adjustment amplitude and speed are determined according to the anchor material characteristics and application environment data, and the tension adjustment strategy data is obtained; According to the tension adjustment strategy data, the tension adjustment actuator is controlled to apply the adjusted tension to the anchor rod, and the tension data of the adjusted anchor rod is collected to obtain the real-time monitoring data after the tension adjustment; The optimized algorithm is applied to subsequent anchor tension force simulation tests through iterative training.

2. The anchor tension simulation test method according to claim 1, characterized in that: Get the qualified range data of tension force, including: Obtain the mechanical properties data of anchor materials and the application environment load data, and establish a correlation model between material properties and environmental factors; Train the correlation model between material properties and environmental factors to obtain an optimized prediction model; Analyze the stress conditions of different materials in different environments based on the prediction model and determine the applicable scope of the materials; Analyze the stress distribution of the anchor rod under different tension forces and determine the ultimate bearing capacity of the anchor rod; Based on the applicable scope of the material and the ultimate bearing capacity of the anchor rod, the qualified range of the anchor rod tension force is determined, and the qualified range data of the tension force is obtained.

3. The anchor tension simulation test method according to claim 2, characterized in that: Get real-time monitoring data of tension, including: Preprocess the real-time anchor tension force data, including data cleaning, outlier removal and data normalization, to obtain standardized anchor tension force data; Model the pre-processed real-time monitoring data of anchor tension force and extract the time domain and frequency domain characteristics of the data; According to the characteristics of anchor tension data, a support vector machine model is trained to determine whether the anchor tension is abnormal. If an abnormality is detected, an early warning mechanism is triggered; The real-time monitoring data and analysis results of anchor tension force are displayed in the form of curve graphs and bar graphs; The real-time monitoring data and analysis results of the anchor tension force are stored in the database to obtain the real-time monitoring data of the tension force.

4. The anchor tension simulation test method according to claim 3, characterized in that: Get real-time monitoring data after tension adjustment, including: According to a pre-established tension adjustment strategy database, obtaining tension adjustment strategy data matching the current anchor state; According to the obtained tension adjustment strategy data, the tension adjustment actuator is controlled to apply the adjusted tension to the target anchor bolt; Use high-precision sensors to collect the tension data of the adjusted anchor rod in real time, and transmit the collected data to the data acquisition system; Preprocessing the collected adjusted anchor tension force data, including data cleaning and data normalization; Through data preprocessing, real-time monitoring data after tension adjustment is obtained.

5. The anchor tension simulation test method according to claim 4, characterized in that: The optimized algorithm is applied to subsequent anchor tension simulation tests, including: Clean and normalize the data based on historical test data; According to the pre-processed historical test data, a model for determining the qualified range of anchor tension force is established by setting appropriate kernel functions and penalty factors; Based on the pre-processed historical test data, a model for formulating anchor tension adjustment strategies is established by setting appropriate tree depth and number of leaf nodes; Evaluate the established anchor tension force qualified range determination model and adjustment strategy formulation model, and determine whether the performance of the model meets the requirements by calculating the model's accuracy and recall rate indicators; If the model performance does not meet the requirements, the model's hyperparameters are optimized and the optimal model parameters are searched by setting different parameter combinations; The optimized anchor tension force qualified range judgment model and adjustment strategy formulation model are applied to the subsequent anchor tension force simulation test. According to the simulation test data, it is judged whether the anchor tension force is within the qualified range, and the corresponding adjustment strategy is formulated according to the judgment result. In the anchor tension simulation test, the test data is continuously acquired and added to the historical test data to update and optimize the anchor tension qualified range determination model and the adjustment strategy formulation model.

6. An anchor tension simulation test system, characterized in that: include: The first processing device is used to obtain the anchor material characteristics and application environment data, determine the qualified range of the anchor tension force, and obtain the qualified range data of the tension force; The second processing device is used to collect real-time data of the tension force of the anchor rod and obtain real-time monitoring data of the tension force; The third processing device is used to compare the tension force qualified range data with the tension force real-time monitoring data to determine whether the anchor tension force exceeds the qualified range. If it exceeds, the tension force adjustment mechanism is triggered, otherwise the current tension force is maintained; The fourth processing device is used to determine the tension adjustment amplitude and speed according to the anchor material characteristics and application environment data when the tension adjustment mechanism is triggered, and obtain the tension adjustment strategy data; A fifth processing device is used to control the tension adjustment actuator to apply the adjusted tension to the anchor rod according to the tension adjustment strategy data, and collect the tension data of the anchor rod after adjustment to obtain the real-time monitoring data after the tension adjustment; The sixth processing device is used to apply the optimized algorithm to subsequent anchor tension force simulation tests through iterative training.

7. The anchor tension simulation test system according to claim 6, characterized in that: The first processing device comprises: The first processing module is used to obtain the mechanical properties data of the anchor material and the application environment load data, and establish a correlation model between the material properties and the environmental factors; The second processing module is used to train the model of correlation between material properties and environmental factors to obtain an optimized prediction model; The third processing module is used to analyze the stress conditions of different materials under different environments according to the prediction model and determine the applicable scope of the materials; The fourth processing module is used to analyze the stress distribution of the anchor rod under different tension forces and determine the ultimate bearing capacity of the anchor rod; The fifth processing module is used to determine the qualified range of the anchor tension force by combining the applicable range of the material and the ultimate bearing capacity of the anchor, and obtain the qualified range data of the tension force.

8. The anchor tension simulation test system according to claim 7, characterized in that: The second processing device comprises: The sixth processing module is used to pre-process the real-time anchor tension force data, including data cleaning, outlier removal and data normalization operations, to obtain standardized anchor tension force data; The seventh processing module is used to model the pre-processed real-time monitoring data of anchor tension force and extract the time domain and frequency domain characteristics of the data; An eighth processing module is used to train a support vector machine model according to the characteristics of the anchor tension force data, so as to determine whether the anchor tension force is abnormal; A ninth processing module, for triggering an early warning mechanism if an abnormality is detected; The tenth processing module is used to display the real-time monitoring data and analysis results of the anchor tension force in the form of a curve chart and a bar chart in real time; The eleventh processing module is used to store the real-time monitoring data and analysis results of the anchor tension force in a database to obtain the real-time monitoring data of the tension force.

9. The anchor tension simulation test system according to claim 8, characterized in that: The third processing device comprises: The twelfth processing module is used to establish an intelligent adjustment model of the tension force of the anchor rod according to the material characteristic data of the anchor rod and the application environment data, so as to determine the optimal tension force adjustment strategy; The thirteenth processing module is used to train the historical data to obtain the association rules between the anchor material characteristics, environmental factors and the tension force adjustment range and speed; The fourteenth processing module is used to obtain the anchor application environment data in real time, determine whether the conditions for triggering the tension adjustment mechanism are met, and if so, start the intelligent adjustment process; The fifteenth processing module is used to input the real-time collected anchor material property data and environmental data into the trained machine learning model to predict the optimal tension adjustment amplitude and speed parameters.

10. The anchor tension simulation test system according to claim 9, characterized in that: The fourth processing device comprises: A sixteenth processing module is used to obtain tension adjustment strategy data matching the current anchor state according to a pre-established tension adjustment strategy database; A seventeenth processing module is used to control the tension adjustment actuator to apply the adjusted tension to the target anchor according to the obtained tension adjustment strategy data; An eighteenth processing module is used to collect the tension force data of the adjusted anchor rod in real time using a high-precision sensor, and transmit the collected data to a data acquisition system; A nineteenth processing module is used to pre-process the collected adjusted anchor tension force data, including data cleaning and data normalization; The twentieth processing module is used to obtain real-time monitoring data after the tension force is adjusted through data preprocessing.

Images