A tension stringing construction intelligent tensioner control analysis system

Abstract

The application provides a tension stringing construction intelligent tensioner control analysis system, and relates to the fields of power transmission line stringing construction and automatic control.The application uses a data acquisition module to collect multi-dimensional data of the tensioner operation in real time, a data transmission module transmits the multi-dimensional data to a data terminal and a data processing module, the data processing module processes the received data, performs deep feature analysis by using a Markov chain model, and performs fault risk prediction, corresponding early warning operations are performed according to the prediction results, and accurate braking schemes are given, so that the stringing construction progress is not affected due to blind braking or delayed braking, through integration of multiple technologies, combination of real-time data acquisition, data processing and visual fault analysis, comprehensive monitoring and intelligent management of the tensioner construction process are realized, not only the equipment operation state can be monitored in real time, but also potential faults can be early warned, and the construction efficiency and safety are improved.

Description

Technical Field

[0001] This invention relates to the field of power transmission line stringing construction and automatic control, and in particular to an intelligent tensioning machine control and analysis system for tension stringing construction. Background Technology

[0002] Economic development is inseparable from electricity. Electricity is transmitted over long distances from power plants to homes via high-voltage overhead lines, making overhead lines a crucial means of supporting long-distance power transmission. However, actual overhead line construction is complex and involves harsh environments, frequently resulting in construction safety accidents. Traditional intelligent monitoring systems often fail to accurately identify the root causes of tensioning machine malfunctions. Abnormal operation of the tensioning machine can lead to blind braking, or failure to detect potential problems, resulting in serious malfunctions, delays, and hindering construction progress. This patent addresses this issue by establishing a dynamic fault analysis system for tensioning machines based on a Discrete Time Markov Chain (DTMC) model. This system deeply analyzes real-time dynamic data of the tensioning machine, accurately identifies the root causes of malfunctions, and provides precise braking solutions, preventing delays in overhead line construction caused by blind or delayed braking. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent tensioning machine control and analysis system for tensioning construction. Based on the dynamic fault tree analysis of the Markov chain model, a dynamic fault analysis system for the tensioning machine is built. The system deeply analyzes the real-time dynamic data of the tensioning machine and provides a precise braking scheme to avoid blind braking of the tensioning machine, which would greatly delay the construction progress.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: an intelligent tensioning machine control and analysis system for tension stringing construction, comprising:

[0005] The data acquisition module is used to collect multi-dimensional data on the operation of the tensioner in real time. The multi-dimensional data includes data from multiple sensors and relevant data from the traction machine. The multi-sensor data includes speed, tension, force, temperature, and pressure. The relevant data from the traction machine includes tensioner current and traction machine operating status data.

[0006] The data transmission module, connected to the data acquisition module, is used to transmit multi-dimensional data to the data terminal and the data processing module.

[0007] The data processing module, connected to the data transmission module, is used to process the received data.

[0008] The fault analysis module builds a dynamic fault analysis model for the tensioner based on a dynamic fault tree of the Markov chain model, and deeply analyzes the real-time dynamic data of the tensioner.

[0009] The electromagnetic braking module is used to brake according to the braking information transmitted by the fault analysis module. The electromagnetic braking module includes a rotating device and a braking device. The electromagnetic braking module uses the braking force of the braking device to generate resistance, and the resistance restricts the rotational force of the motor in the rotating device to achieve precise braking of the motor.

[0010] The data transmission module, in conjunction with cloud computing technology, tracks and analyzes multi-dimensional data, schedules the monitoring data processed by the data processing module, and transmits it to the fault analysis module. The specific scheduling and planning process is described by the following formula:

[0011] ;

[0012] In the formula, For the current number A scheduling state, For the i-th virtual machine, For the first i One task, The number of instructions executed by the CPU per second. Memory usage is expressed in GB. The unit is the amount of storage used. For the monitoring dataset of the traction machine, k The scheduling strategy used, q For the number of scheduling policies, Number of samples for each strategy The first The task set, the virtual machine's feature vector, and the task allocation state within the virtual machine at each scheduling state. These are the model's predicted values. These are actual monitored values.

[0013] In the preferred embodiment, the multi-sensor data includes a speed sensor to collect the line-laying speed of the traction machine during on-site construction, a tension sensor to collect the traction tension of the traction machine, a force sensor to collect the tail force of the traction machine, a temperature sensor to collect the temperature of each key component of the tensioning machine, and a pressure sensor to collect the pressure status data of the tensioning machine.

[0014] In the preferred scheme, the scheduling and transmission of monitoring data includes the following steps:

[0015] Acquire the data transmitted through the channel and transmit it according to the preset scheduling path;

[0016] Update the real-time status of the current channel and the preset target channel in real time;

[0017] If the transmitted data is detected to have reached the preset target channel, then one scheduling operation is completed.

[0018] If congestion is detected in the current channel or the target channel, scheduling is stopped, and the execution effect of the transmission task and the congestion status of the channel are quantitatively tested.

[0019] The quantitative test metrics are: bit error rate and data packet loss rate The formula is:

[0020] ;

[0021] ;

[0022] In the formula, and These are the amount of data transmitted into the channel and the amount of data actually received, respectively. To receive erroneous data in the transmitted data.

[0023] The preferred scheme also includes using the data transmission rate as a test indicator of channel transmission efficiency, with the formula as follows:

[0024] ;

[0025] In the formula, and These represent the amount of data transmitted in the channel and the data transmission time, respectively.

[0026] In the preferred embodiment, the data processing module preprocesses the multi-dimensional data, specifically by detecting outliers in the continuous data of the traction machine and filtering out individual outliers in the multi-dimensional data. This includes the following steps:

[0027] Step 1: Let D represent the various monitoring data sets of the tensioning machine. Store m monitoring data points in the set. Based on the isolation tree detection principle, perform anomaly detection for any monitoring data point d in the tensioning machine monitoring data set, using the following formula:

[0028] ;

[0029] In the formula, n is the total number of samples. is a constant in isolated forest, d is the depth of data points, m is the total amount of data, and H represents the Euler constant of isolated tree;

[0030] Step 2: Calculate the anomaly score of the data based on the average path length of the device data in the isolated tree, using the following formula:

[0031] ;

[0032] In the formula, This represents the average length of all data paths in the monitored dataset;

[0033] Step 3: If calculation If the value is greater than the preset anomaly score threshold, it is considered abnormal data and should be deleted; otherwise, a final decision is made using the following formula:

[0034] ;

[0035] ;

[0036] ;

[0037] In the formula, These are the parameters of the radial basis kernel function. Characteristic variable 1 refers to the total number of data samples. Feature variable 2 refers to the total number of data samples. Let i be the feature variable of the data sample to be determined. and Let i be the feature variables of the data sample to be determined and j be the reference data sample. The decision value is for the reference data sample j. The decision value for the data sample to be determined. For support vector coefficients, Let c be the error coefficient, and c be the data sample to be determined. The maximum decision value, when calculated If the value is greater than zero and less than c, the data sample is considered normal; otherwise, it is considered abnormal.

[0038] Step 4: Based on the data anomaly analysis results in Step 3, filter out the abnormal data in the dataset and integrate the remaining normal data as the final result of preprocessing.

[0039] In the preferred scheme, a series of dynamic logic gates are introduced to describe the timing rules and dynamic failure behavior, including four typical dynamic logic gates: priority AND gate, function-dependent gate, sequence-dependent gate, and spare part gate, and the logic gates are transformed into a Markov chain model.

[0040] In the preferred embodiment, the fault analysis module performs in-depth analysis of the real-time dynamic data of the tensioner based on the dynamic fault analysis model, and uses dynamic fault tree analysis to calculate the membership function of the fuzzy failure probability at time t when the tensioner is in an abnormal state, including the following steps:

[0041] 1) Based on the operating status of the tensioning machine and the analysis of various operating data, a fault tree analysis model is established. Then, a Markov chain model is used to transform the dynamic fault tree with n states. In the transformed Markov chain model, fuzzy numbers are used to represent the transition rates between states, thus transforming the model's state transition rate matrix into a fuzzy state transition rate matrix. The formula is:

[0042] ;

[0043] In the formula, The fuzzy state transition rate matrix is... Indicates the failure rate of input events;

[0044] 2) Based on the dynamic fault tree diagram of the tensioning machine, the differential equation of the corresponding Markov chain model is obtained as follows:

[0045] ;

[0046] In the formula, Let be the probability distribution with respect to time. For fuzzy failure rate of input events and It is a probability distribution over time;

[0047] 3) Set initial conditions Applying the Laplace-Stieltjes transformation to the system of equations in 2), we obtain the following linear system of equations:

[0048] ;

[0049] In the formula, Let be the transformation function of the probability distribution over time with respect to s. Let be the s-transform function of the probability distribution with respect to time. The fuzzy failure rate for input events;

[0050] 4) Solving the system of equations in 3) yields information about s. The function is subjected to an inverse Laplace-Stieltjes transform, and the state of the tensioner with respect to time is obtained. The probability distribution is used to obtain the current fault status of the tensioning machine.

[0051] This invention provides an intelligent control and analysis system for tensioning machines used in tension line construction. It utilizes a data acquisition module to collect multi-dimensional data on the tensioning machine's operation, a data transmission module to transmit this multi-dimensional data, a data processing module to process the received data, and a fault analysis module to perform in-depth analysis and fault risk prediction. This system accurately identifies the root causes of tensioning machine faults and provides precise braking solutions, avoiding delays in the construction progress caused by blind or delayed braking. By integrating multiple technologies and combining real-time data acquisition, data processing, and fault analysis, it achieves comprehensive monitoring and intelligent management of the tensioning machine's construction process, real-time monitoring of equipment operating status, and early warning of potential faults, thereby improving construction efficiency and safety. Attached Figure Description

[0052] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0053] Figure 1 This is a schematic diagram of the system modules of the present invention;

[0054] Figure 2 This is a schematic diagram of the installation positions of the multiple sensors in this invention;

[0055] Figure 3 This is the logic diagram of the tensioner fault analysis system of the present invention;

[0056] Figure 4 This is a flowchart of the data processing for monitoring in this invention;

[0057] Figure 5 This is a diagram of a braking device provided by the present invention;

[0058] Figure 6 This is a schematic diagram of the dynamic fault analysis of the tensioning machine according to the present invention;

[0059] Figure 7 This is a schematic diagram illustrating the transformation of the fault tree model into a fuzzy Markov chain model according to the present invention;

[0060] Figure 8 This is a membership function graph of the real-time fuzzy failure probability of the traction machine of the present invention;

[0061] Figure 9 This is the state diagram of the cold spare parts conversion Markov model of the present invention;

[0062] Figure 10 This is the state diagram of the Markov model for converting the temperature-controlled spare parts of this invention;

[0063] Figure 11 This is the state diagram of the Markov model for converting hot spare parts according to the present invention.

[0064] In the diagram: 1. Pressure sensor; 2. Tension sensor; 3. Monitoring system; 4. Braking system; 5. Data transmission system; 6. Speed ​​sensor; 7. Tension sensor; 8. Temperature sensor; 9. Signal receiving device; 10. Brake pad; 11. Brake disc; 12. Mounting screw; 13. Brake housing; 14. Brake stator. Detailed Implementation

[0065] Example 1

[0066] like Figure 1-9 As shown, an intelligent tensioning machine control and analysis system for tension stringing construction includes:

[0067] The data acquisition module is used to collect multi-dimensional data on the operation of the tensioner in real time. The multi-dimensional data includes data from multiple sensors and relevant data from the traction machine. The multi-sensor data includes speed, tension, force, temperature, and pressure. The relevant data from the traction machine includes tensioner current and traction machine operating status data.

[0068] The data transmission module, connected to the data acquisition module, is used to transmit multi-dimensional data to the data terminal and the data processing module.

[0069] The data processing module, connected to the data transmission module, is used to process the received data.

[0070] The fault analysis module builds a dynamic fault analysis model for the tensioning machine based on the dynamic fault tree of the Markov chain model. It deeply analyzes the real-time dynamic data of the tensioning machine, accurately grasps the root cause of the tensioning machine fault, and provides a precise braking scheme to avoid affecting the progress of the stringing construction due to blind braking or delayed braking.

[0071] The electromagnetic braking module is used to brake based on the braking information transmitted by the fault analysis module. The electromagnetic braking module includes a rotating device and a braking device. The electromagnetic braking module uses the braking force of the braking device to generate resistance, and the resistance restricts the rotational force of the motor in the rotating device to achieve precise braking of the motor.

[0072] like Figure 5 As shown, an electromagnetic braking module is provided with a braking device, which consists of a signal receiver 9, a brake pad 10, a brake disc 11, mounting screws 12, a brake housing 13, and a brake stator 14.

[0073] The automatic fault analysis system sends a braking signal to the signal receiver 9. The control system of the braking device precisely controls the brake disc 11 according to the braking scheme given by the fault analysis system, i.e., the control signal received by the braking device. The brake body on the brake disc 11 receives the braking force and generates resistance. This resistance restricts the rotational force of the motor, thereby achieving precise braking of the motor.

[0074] like Figure 1 As shown, this embodiment utilizes a data acquisition module to collect multi-dimensional data on the operation of the tensioning machine in real time. The data transmission module transmits the multi-dimensional data to the data terminal and the data processing module. The data processing module processes the received data, performs deep feature analysis using a Markov chain model, and predicts fault risks. Based on the prediction results, it performs corresponding early warning operations and provides precise braking schemes to avoid affecting the progress of the stringing construction due to blind or delayed braking. By integrating multiple technologies and combining real-time data acquisition, data processing, and visualized fault analysis, comprehensive monitoring and intelligent management of the tensioning machine construction process are achieved. The system monitors the equipment's operating status in real time, provides early warnings of potential faults, and greatly improves construction efficiency and safety.

[0075] In the preferred embodiment, the multi-sensor data includes a speed sensor to collect the line-laying speed of the traction machine during on-site construction, a tension sensor to collect the traction tension of the traction machine, a force sensor to collect the tail force of the traction machine, a temperature sensor to collect the temperature of each key component of the tensioning machine, and a pressure sensor to collect the pressure status data of the tensioning machine.

[0076] Among the multi-sensor data, the speed sensor measures the laying speed of the traction machine during construction to ensure the accuracy of the construction progress; the tension sensor measures the traction tension of the traction machine to ensure that the tension is within a safe range; the force sensor monitors the tension at the tail of the traction machine to prevent overload; the temperature sensor monitors the temperature of each key component of the tensioning machine to prevent overheating and malfunctions; and the pressure sensor records the pressure status of the tensioning machine to ensure the normal operation of the system.

[0077] Multiple sensors need to be positioned at their respective optimal detection locations to ensure the accuracy of the acquired data. In this embodiment, the actual reference positioning positions are given as follows: Figure 2 As shown: Pressure sensor 1 is installed in the hydraulic system of the tensioner to determine the pressure status of the tensioner; tension sensor 2 is installed at the end of the tensioner to monitor the tail tension experienced by the traction machine during on-site construction and line laying; speed sensor 6 is installed on the tensioner axle to monitor the line laying speed of the traction machine during on-site construction; tension sensor 7 is installed on the tensioner axle to monitor the traction tension of the traction machine; and temperature sensor 8 is installed near the tensioner engine to monitor the engine temperature of the traction machine.

[0078] According to such Figure 3 and 4The data collected by various sensors of the tensioning machine for the overhead line construction is transmitted remotely to the data terminal using a data transmission module built with cloud computing. The data terminal then transmits the data to the dynamic fault analysis module to detect potential problems, provide a precise braking solution, and transmit the braking signal to the electromagnetic braking module, thereby realizing the autonomous detection and control of potential faults of the tensioning machine.

[0079] like Figure 4 As shown, the data push component and the real-time fault analysis component communicate via WebSocket to transmit data. After receiving the data, the component analyzes the current data in real time, performs fault analysis, stores the data, and feeds it back to the terminal of the analysis system for display or to issue corresponding warnings.

[0080] Detailed description: Taking overhead line construction monitoring data as an example, the fault analysis module is used to analyze the real-time data of the tensioning machine, specifically divided into data such as the temperature of key equipment components, tail tension, wire laying speed, and pressure status, etc. Figure 6 As shown, X1, X2...X8 represent the following abnormalities: abnormal traction speed, abnormal tail tension of the traction machine, abnormal traction tension of the traction machine, abnormal engine temperature of the tensioning machine, abnormal hydraulic system of the tensioning machine, oil circuit failure of the tensioning machine, and abnormal internal pressure of the tensioning machine, respectively; M1 represents the abnormal external construction status of the tensioning machine, and M2 represents the abnormal internal system status of the tensioning machine.

[0081] like Figure 7 As shown, the fault tree model is then transformed into a fuzzy Markov chain model, which includes five states: S1-S5.

[0082] S1: Normal working state of the tensioner; S2: Partial malfunction caused by external construction abnormalities of the tensioner; S3: Partial malfunction caused by internal abnormalities of the traction machine; S4: Partial malfunction caused by internal abnormalities of the tensioner; S5: Complete malfunction.

[0083] The probability of state S5 as a function of time t is calculated, and the fuzzy failure probability function is given by the following formula:

[0084] ;

[0085] Given a task time t, the upper and lower limits of the fuzzy probability of the system being in state S5 are calculated, which are the membership functions of the fuzzy failure probability of the system at that time t. The severity of the fault is set to three levels: A, B, and C. When the failure probability of the tensioner is below 0.1, it is level C. In this case, the electromagnetic braking system can temporarily remain inactive while monitoring continues. When the failure probability of the tensioner is above 0.1 but below 0.5, it is level B. In this case, the electromagnetic braking system needs to activate to reduce the tensioner's laying speed or take other measures to reduce the risk. When the failure probability of the tensioner is above 0.5, it is level A. In this case, the tensioner must be stopped immediately, and construction personnel must take measures to troubleshoot the fault.

[0086] like Figure 8 As shown, when the traction machine's wire-laying speed is abnormal, the membership function of the fuzzy failure probability of the tensioning machine at t=50h is calculated.

[0087] Depend on Figure 8 The data shows that the median of the fuzzy probability is 0.166. This indicates that after 50 hours of operation, the maximum possible failure probability is 0.166. When the maximum possible failure probability of the tensioner reaches 0.166, the electromagnetic braking module needs to activate to reduce the tensioner's wire release speed or take other measures to reduce the risk.

[0088] In the preferred scheme, the data transmission module, combined with cloud computing technology, tracks and analyzes multi-dimensional data, schedules the monitoring data processed by the data processing module, and transmits it to the fault analysis module. The specific scheduling and planning process is formulated as follows:

[0089] ;

[0090] In the formula, For the current number A scheduling state, For the i-th virtual machine, For the first i One task, The number of instructions executed by the CPU per second. Memory usage is expressed in GB. The unit is the amount of storage used. For the monitoring dataset of the traction machine, k The scheduling strategy used, q For the number of scheduling policies, Number of samples for each strategy The first The task set, the virtual machine's feature vector, and the task allocation state within the virtual machine at each scheduling state. These are the model's predicted values. These are actual monitored values.

[0091] Furthermore, it can be set To initialize the zero matrix, To initialize with an empty graph.

[0092] Employing cloud computing technology to schedule channel transmission during data transmission can effectively reduce channel congestion and improve the monitoring efficiency of the data transmission system. The data transmitted by the data transmission system is preprocessed to detect anomalies in the continuous data of the traction machine, filtering out individual abnormal data from the entire transmission. The data set is denoted by D, and n device data points are stored in the set. Data segmentation points and cutting planes are selected to spatially segment the data set. Based on the principle of isolated tree detection, anomaly detection is performed on any device data d in the data set, and a decision model is used to make a final decision on data with potential risks.

[0093] In this embodiment, data transmission effectively utilizes cloud computing resources, improving the efficiency of data processing and transmission.

[0094] Compared to traditional data transmission, cloud-based data transmission does not require a large amount of complex hardware infrastructure, thus reducing energy waste and costs. Furthermore, due to cloud computing's advanced data encryption technology, it is more secure.

[0095] In the preferred scheme, the scheduling and transmission of monitoring data includes the following steps:

[0096] Acquire the data transmitted through the channel and transmit it according to the preset scheduling path.

[0097] Update the real-time status of the current channel and the preset target channel.

[0098] If the transmitted data is detected to have reached the preset target channel, then a scheduling operation is completed.

[0099] If congestion is detected in the current channel or the target channel, scheduling is stopped, and the execution effect of the transmission task and the congestion status of the channel are quantitatively tested.

[0100] The quantitative test metrics are: bit error rate and data packet loss rate The formula is:

[0101] ;

[0102] ;

[0103] In the formula, and These are the amount of data transmitted into the channel and the amount of data actually received, respectively. To receive erroneous data in the transmitted data.

[0104] The monitoring data is scheduled and transmitted. Its data acquisition and preliminary processing include acquiring data transmitted through the channel and transmitting it according to the preset scheduling path, so as to transmit the data through the specified channel according to the set scheduling strategy.

[0105] Update the current channel status in real time, including monitoring and updating the real-time status of the current channel.

[0106] In the preferred scheme, the data transmission rate is used as the test index for channel transmission efficiency, and the formula is:

[0107] ;

[0108] In the formula, and These represent the amount of data transmitted in the channel and the data transmission time, respectively.

[0109] In the preferred embodiment, the data processing module preprocesses the multi-dimensional data, specifically by detecting outliers in the continuous data of the traction machine and filtering out individual outliers in the multi-dimensional data. This includes the following steps:

[0110] Step 1: Let D represent the various monitoring data sets of the tensioning machine. Store m monitoring data points in the set. Based on the isolation tree detection principle, perform anomaly detection for any monitoring data point d in the tensioning machine monitoring data set, using the following formula:

[0111] ;

[0112] In the formula, n is the total number of samples. is a constant in isolated forest, d is the depth of the data points, m is the total amount of data, and H represents the Euler constant of the isolated tree.

[0113] Step 2: Calculate the anomaly score of the data based on the average path length of the device data in the isolated tree, using the following formula:

[0114] ;

[0115] In the formula, This represents the average length of all data paths in the monitored dataset.

[0116] Step 3: If calculation If the value is greater than the preset anomaly score threshold, it is considered abnormal data and should be deleted; otherwise, a final decision is made using the following formula:

[0117] ;

[0118] ;

[0119] ;

[0120] In the formula, For the radial basis function parameters, in this embodiment... The value is a constant 1. Characteristic variable 1 refers to the total number of data samples. Feature variable 2 refers to the total number of data samples. Let i be the feature variable of the data sample to be determined. and Let i be the feature variables of the data sample to be determined and j be the reference data sample. The decision value is for the reference data sample j. The decision value for the data sample to be determined. For support vector coefficients, Let c be the error coefficient, and c be the data sample to be determined. The maximum decision value, when calculated If the value is greater than zero and less than c, the data sample is considered normal; otherwise, it is considered abnormal.

[0121] Step 4: Based on the data anomaly analysis results in Step 3, filter out the abnormal data in the dataset and integrate the remaining normal data as the final result of preprocessing.

[0122] In the preferred scheme, a series of dynamic logic gates are introduced into the fault analysis module to describe the system's timing rules and dynamic failure behavior. These mainly include four typical dynamic logic gates: priority AND gates, function-dependent gates, sequence-dependent gates, and spare parts gates. These logic gates are then transformed into a Markov chain model to calculate the fuzzy failure rate of the tensioning machine, enabling dynamic fault analysis of the tensioning machine's real-time data, timely detection of tensioning machine faults, and ensuring the smooth progress of overhead line construction.

[0123] In the preferred embodiment, the fault analysis module performs in-depth analysis of the real-time dynamic data of the tensioner based on the dynamic fault analysis model, and uses dynamic fault tree analysis to calculate the membership function of the fuzzy failure probability at time t when the tensioner is in an abnormal state, including the following steps:

[0124] 1) Based on the operating status of the tensioning machine and the analysis of various operating data, a fault tree analysis model is established. Then, a Markov chain model is used to transform the dynamic fault tree with n states. In the transformed Markov chain model, fuzzy numbers are used to represent the transition rates between states, thus transforming the model's state transition rate matrix into a fuzzy state transition rate matrix. The formula is:

[0125] ;

[0126] In the formula, The fuzzy state transition rate matrix is... Indicates the failure rate of the input event.

[0127] 2) Based on the dynamic fault tree diagram of the tensioning machine, the differential equation of the corresponding Markov chain model is obtained as follows:

[0128] ;

[0129] In the formula, Let be the probability distribution with respect to time. For fuzzy failure rate of input events and It is a probability distribution over time.

[0130] 3) Set initial conditions Applying the Laplace-Stieltjes transformation to the system of equations in 2), we obtain the following linear system of equations:

[0131] ;

[0132] In the formula, Let be the transformation function of the probability distribution over time with respect to s. Let be the s-transform function of the probability distribution with respect to time. The fuzzy failure rate of input events.

[0133] 4) Solving the system of equations in 3) yields information about s. The function is subjected to an inverse Laplace-Stieltjes transform, and the state of the tensioner with respect to time is obtained. The probability distribution is used to obtain the current fault status of the tensioning machine.

[0134] In this embodiment, the TREE-NEURAL NETWORK (TNN) technology is specifically described as follows: A series of dynamic logic gates are introduced to describe the system's timing rules and dynamic failure behavior. These mainly include four typical dynamic logic gates: Priority-AND Gate (PAND), Functional Dependency Gate (FDEP), Sequence Enforcing Gate (SEQ), and Spare Gate (SP). These logic gates are then transformed into a Markov chain model. The four typical dynamic logic gates are all commonly used forms and will not be elaborated upon here.

[0135] like Figure 9As shown, based on the working mechanism of the above three spare parts doors, assuming the failure rate of the component is λ, when used as a spare part, its failure rate can be described as βλ.

[0136] Analysis yields the following: Figure 9 The diagram shows the state diagram for converting a cold spare part into a Markov model. When β=0, the part is a cold spare. The state diagram consists of three state nodes (00, 10, Fail) and state transition lines. A represents the basic component, and S represents the spare part. Starting from node 00, a state line points to node 10, indicating that the condition is met. When the condition is met, the status changes from 00 to 10, indicating that component A has failed; similarly, if the condition is met... When the status changes from 10 to Fail, it indicates that the entire system has failed.

[0137] I. State Definition:

[0138] 00: Indicates a system state where both basic component A and spare component S are operational.

[0139] 10: Indicates the system state where component A has failed.

[0140] Fail: Indicates a system failure status.

[0141] condition This indicates the failure probability of component A when it is in operation.

[0142] condition This represents the failure probability of spare part S when it is in working condition.

[0143] II. State Transition:

[0144] 1. 00→10: This indicates that two components have changed from a normal operating state to a failure state for component A. Since spare part S can still work, the entire system has not failed.

[0145] 2. 10 → Fail: This indicates a change from the failure state of component A to the failure state of both components. The entire system fails.

[0146] like Figure 10 and 11 The figures show the state diagrams of the warm and hot spare parts converted into Markov models, respectively. When β=1, the part is a hot spare part; when 0<β<1, the part is a warm spare part.

[0147] It consists of four state nodes (00, 01, 10, 00) and state transition lines.

[0148] I. State Definition:

[0149] 00: Indicates that both the warm / hot spare and input component A are in normal working condition. In this state, the spare is on standby and does not actually participate in the operation, but remains ready to take over from the main unit at any time.

[0150] 10: Input component fails, spare part works normally. At this time, the spare part changes from standby state to working state, while input component A is in a failed state.

[0151] 01: Warm / Hot Spare Part Failure. After taking over the work of input component A, the spare part may fail for various reasons, leaving the system in a state without a working component.

[0152] Fail: This indicates that both spare part S and component A have failed, resulting in a complete system failure.

[0153] condition This indicates the failure probability of component A when it is in operation.

[0154] condition This represents the failure probability of spare part S when it is in working condition.

[0155] II. State Transition:

[0156] 1. 00→01: This indicates that two components have changed from a normal operating state to a failure state of spare component S. Since component A can still work, the entire system has not failed.

[0157] 2. 00→10: This indicates that two components have changed from a normal operating state to a failure state for component A. Since spare part S can still work, the entire system has not failed.

[0158] 3. 10 / 01 → Fail: This indicates a change from the failure state of component A / spatial component S to a state where both components have failed. The entire system has failed.

[0159] In Markov models, these intermediate states are typically simplified to a single transition process.

[0160] This embodiment uses a deep analysis virtual model to accurately identify the root cause of tensioning machine failures based on the depth characteristics of the data, and provides precise braking solutions to avoid affecting the progress of stringing construction due to blind or delayed braking. By integrating multiple technologies and combining real-time data acquisition, data processing, and visualized fault analysis, it achieves comprehensive monitoring and intelligent management of the tensioning machine construction process.

[0161] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. An intelligent tensioning machine control and analysis system for tension stringing construction, characterized in that, include: The data acquisition module is used to collect multi-dimensional data on the operation of the tensioner in real time. The multi-dimensional data includes data from multiple sensors and relevant data from the traction machine. The multi-sensor data includes speed, tension, force, temperature, and pressure. The relevant data from the traction machine includes tensioner current and traction machine operating status data. The data transmission module, connected to the data acquisition module, is used to transmit multi-dimensional data to the data terminal and the data processing module. The data processing module, connected to the data transmission module, is used to process the received data. The fault analysis module builds a dynamic fault analysis model for the tensioner based on a dynamic fault tree of the Markov chain model, and deeply analyzes the real-time dynamic data of the tensioner. The electromagnetic braking module is used to brake according to the braking information transmitted by the fault analysis module. The electromagnetic braking module includes a rotating device and a braking device. The electromagnetic braking module uses the braking force of the braking device to generate resistance, and the resistance restricts the rotational force of the motor in the rotating device to achieve precise braking of the motor. The data transmission module, in conjunction with cloud computing technology, tracks and analyzes multi-dimensional data, schedules the monitoring data processed by the data processing module, and transmits it to the fault analysis module. The specific scheduling and planning process is described by the following formula: ； In the formula, For the current number A scheduling state, For the i-th virtual machine, For the first i One task, The number of instructions executed by the CPU per second. Memory usage is expressed in GB. The unit is the amount of storage used. For the monitoring dataset of the traction machine, k The scheduling strategy used, q For the number of scheduling policies, Number of samples for each strategy The first The task set, the virtual machine's feature vector, and the task allocation state within the virtual machine at each scheduling state. These are the model's predicted values. These are actual monitored values.

2. The intelligent tensioning machine control and analysis system for tension stringing construction according to claim 1, characterized in that, Among the multi-sensor data, a speed sensor is used to collect the laying speed of the traction machine during on-site construction, a tension sensor is used to collect the traction tension of the traction machine, a force sensor is used to collect the tail force of the traction machine, a temperature sensor is used to collect the temperature of each key component of the tensioning machine, and a pressure sensor is used to collect the pressure status data of the tensioning machine.

3. The intelligent tensioning machine control and analysis system for tension stringing construction according to claim 2, characterized in that, The scheduling and transmission of monitoring data includes the following steps: Acquire the data transmitted through the channel and transmit it according to the preset scheduling path; Update the real-time status of the current channel and the preset target channel in real time; If the transmitted data is detected to have reached the preset target channel, then one scheduling operation is completed. If congestion is detected in the current channel or the target channel, scheduling is stopped, and the execution effect of the transmission task and the congestion status of the channel are quantitatively tested. The quantitative test metrics are: bit error rate and data packet loss rate The formula is: ； ； In the formula, and These are the amount of data transmitted into the channel and the amount of data actually received, respectively. To receive erroneous data in the transmitted data.

4. The intelligent tensioning machine control and analysis system for tension stringing construction according to claim 3, characterized in that, It also includes using the data transmission rate as a test indicator of channel transmission efficiency, with the formula: ； In the formula, and These represent the amount of data transmitted in the channel and the data transmission time, respectively.

5. The intelligent tensioning machine control and analysis system for tension stringing construction according to claim 1, characterized in that, The data processing module preprocesses the multi-dimensional data, specifically by detecting outliers in the continuous traction data and filtering out individual outliers from the multi-dimensional data. This includes the following steps: Step 1: Let D represent the various monitoring data sets of the tensioning machine. Store m monitoring data points in the set. Based on the isolation tree detection principle, perform anomaly detection for any monitoring data point d in the tensioning machine monitoring data set, using the following formula: ； In the formula, n is the total number of samples. is a constant in isolated forest, d is the depth of data points, m is the total amount of data, and H represents the Euler constant of isolated tree; Step 2: Calculate the anomaly score of the data based on the average path length of the device data in the isolated tree, using the following formula: ； In the formula, This represents the average length of all data paths in the monitored dataset; Step 3: If calculation If the value is greater than the preset anomaly score threshold, it is considered abnormal data and should be deleted; otherwise, a final decision is made using the following formula: ； ； ； In the formula, These are the parameters of the radial basis kernel function. Characteristic variable 1 refers to the total number of data samples. Feature variable 2 refers to the total number of data samples. Let i be the feature variable of the data sample to be determined. and Let i be the feature variables of the data sample to be determined and j be the reference data sample. The decision value is for the reference data sample j. The decision value for the data sample to be determined. For support vector coefficients, Let c be the error coefficient, and c be the data sample to be determined. The maximum decision value, when calculated If the value is greater than zero and less than c, the data sample is considered normal; otherwise, it is considered abnormal. Step 4: Based on the data anomaly analysis results in Step 3, filter out the abnormal data in the dataset and integrate the remaining normal data as the final result of preprocessing.

6. The intelligent tensioning machine control and analysis system for tension stringing construction according to claim 1, characterized in that, By introducing a series of dynamic logic gates to describe timing rules and dynamic failure behavior, including four typical dynamic logic gates: priority AND gate, function-dependent gate, sequence-dependent gate, and spare part gate, the logic gates are transformed into a Markov chain model.

7. The intelligent tensioning machine control and analysis system for tension stringing construction according to claim 6, characterized in that, The fault analysis module, based on the dynamic fault analysis model, deeply analyzes the real-time dynamic data of the tensioner, and uses dynamic fault tree analysis to calculate the membership function of the fuzzy failure probability at time t when the tensioner is in an abnormal state. This includes the following steps: 1) Based on the operating status of the tensioning machine and the analysis of various operating data, a fault tree analysis model is established. Then, a Markov chain model is used to transform the dynamic fault tree with n states. In the transformed Markov chain model, fuzzy numbers are used to represent the transition rates between states, thus transforming the model's state transition rate matrix into a fuzzy state transition rate matrix. The formula is: ； In the formula, The fuzzy state transition rate matrix is... Indicates the failure rate of input events; 2) Based on the dynamic fault tree diagram of the tensioning machine, the differential equation of the corresponding Markov chain model is obtained as follows: ； In the formula, Let be the probability distribution with respect to time. For fuzzy failure rate of input events and It is a probability distribution over time; 3) Set initial conditions Applying the Laplace-Stieltjes transformation to the system of equations in 2), we obtain the following linear system of equations: ； In the formula, Let be the transformation function of the probability distribution over time with respect to s. Let be the s-transform function of the probability distribution with respect to time. The fuzzy failure rate for input events; 4) Solving the system of equations in 3) yields information about s. The function is subjected to an inverse Laplace-Stieltjes transform, and the state of the tensioner with respect to time is obtained. The probability distribution is used to obtain the current fault status of the tensioning machine.

Images