Wheel-step type equipment working state monitoring method and system for power transmission tower

Abstract

The application provides a wheel-step type equipment working state monitoring method and system for a power transmission tower, which comprises the following steps: collecting current-displacement mapping data of a driving motor to construct a no-load tangential motion reference curve of each wheel leg; sending a corresponding sinusoidal reciprocating motion instruction to the tangential driving joint of the current wheel leg based on the no-load tangential motion reference curve; collecting actual current data and encoder displacement data of the tangential driving joint of the current wheel leg according to the sinusoidal reciprocating motion instruction to generate an actual current-displacement curve, and calculating characteristic parameters of a hysteresis loop area, a nonlinearity coefficient and a peak current deviation between the current-displacement curve and the no-load tangential motion reference curve; comparing the characteristic parameters with preset parameter thresholds, and determining that the current wheel leg is in a virtual leg state when each characteristic parameter is lower than the corresponding parameter threshold. The application can effectively improve the monitoring efficiency of the working state of the wheel-step type equipment.

Description

Technical Field

[0001] This invention relates to the field of power transmission technology, and in particular to a method and system for monitoring the working status of wheeled equipment used in power transmission towers. Background Technology

[0002] Wheel-and-leg hybrid equipment combines the high efficiency of wheeled mobility with the strong obstacle-crossing ability of legged mobility, and has become the mainstream equipment for automated operation of power transmission towers. Its safe operation relies heavily on accurate monitoring of its working status. Among these, monitoring the support contact status between the legs and the ground structure during the contact process between the wheel-and-leg equipment and the ground is a core aspect.

[0003] In existing technologies, the monitoring of support contact status in wheeled walking equipment mainly employs a static measurement scheme using distributed pressure sensors and contact switches. Effective support is determined by detecting the magnitude of the contact force between the leg and the support surface. However, traditional static force measurement methods cannot distinguish between true and false contact, resulting in a false leg identification misjudgment rate as high as 25%-30%, seriously threatening equipment and operational safety.

[0004] The root cause of the above problems lies in the fact that existing technologies are based on the flawed assumption that contact equals force and force equals effectiveness, failing to consider the unique line contact support characteristics and surface anomalies specific to certain ground surfaces. Existing technologies fail to identify the dynamic characteristics of the contact process and lack a multi-dimensional verification mechanism, resulting in an inability to effectively solve the industry-wide problem of high false leg identification misjudgment rates. Summary of the Invention

[0005] Based on this, the purpose of the present invention is to provide a method and system for monitoring the working status of wheeled equipment for power transmission towers, so as to solve the problem that the existing static force measurement method cannot distinguish between true contact and false contact, resulting in a high misjudgment rate of false retreat identification.

[0006] The first aspect of the present invention proposes: A method for monitoring the working status of wheeled equipment for power transmission towers, wherein the method includes: When each wheel leg of the wheel-stepping equipment is in a completely suspended state, the end effector of the current wheel leg is controlled to perform a reciprocating motion with a constant amplitude along a preset tangential direction, and the current-displacement mapping data of the drive motor is collected accordingly to construct the no-load tangential motion reference curve of each wheel leg. After any wheel leg performs a ground contact action and reaches the preset target position, the normal joint output torque of the current wheel leg is controlled to be zero, and a corresponding sinusoidal reciprocating motion command is sent to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve. According to the sinusoidal reciprocating motion command, the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg are collected to generate the actual current-displacement curve, and the characteristic parameters of hysteresis loop area, nonlinearity coefficient and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve are calculated. The feature parameters are compared with preset parameter thresholds, and when each feature parameter is lower than its corresponding parameter threshold, the current wheel leg is determined to be in a virtual leg state.

[0007] The beneficial effects of this invention are as follows: This technical solution fundamentally solves the core problem of high false leg identification misjudgment rate caused by the inability of traditional static force measurement methods to distinguish between true and false contact by pre-calibrating the no-load tangential motion reference curve and comparing the tangential mechanical characteristics after ground contact in multiple dimensions. The solution utilizes the essential difference in the mechanical response of tangential motion under true and false contact states of the wheel and leg, eliminates normal load interference by reducing the normal torque to zero after ground contact, and performs fusion judgment based solely on three characteristic parameters extracted from existing drive motor current and encoder data: hysteresis loop area, nonlinearity coefficient, and peak current deviation. No additional force sensor is required, resulting in low hardware cost and strong real-time performance. Simultaneously, each wheel and leg independently constructs a no-load reference curve, effectively eliminating identification deviations caused by individual differences in wheel and leg, installation errors, and long-term wear, significantly improving the accuracy and robustness of false leg identification, thereby ensuring the gait stability and walking safety of wheeled walking equipment in complex terrain.

[0008] Furthermore, the step of acquiring the current-displacement mapping data of the corresponding drive motor to construct the no-load tangential motion reference curve of each wheel leg includes: When the wheel leg is completely suspended in the air and the fuselage remains stationary, a composite excitation signal is sent to the tangential drive joint, and the corresponding current, displacement and acceleration data are collected to separate the effective current component with the same frequency as the excitation signal through cross-correlation analysis. The complete motion cycle of the wheel-walking equipment is divided into several motion stages, and the slope and intercept of the current change with displacement in each motion stage are calculated to detect the viscous friction coefficient and equivalent damping coefficient of the tangential joint under different motion states, and a phased no-load dynamic model is constructed accordingly. Based on the no-load dynamic model, a theoretical no-load current-displacement curve is generated. The actual collected effective current data is then compared with the theoretical no-load current-displacement curve using residual analysis. Target data is selected accordingly, and the least squares method is used to fit the target data to construct the no-load tangential motion reference curve.

[0009] Furthermore, the step of generating a theoretical no-load current-displacement curve based on the no-load dynamic model, and performing residual analysis between the actual collected effective current data and the theoretical no-load current-displacement curve to filter out the target data includes: Granger causality tests were performed on the current-displacement time series data for each motion stage to verify whether the current change was driven unidirectionally by the displacement change, and data segments with invalid causal relationships were removed to generate intermediate data. Within the intermediate data, residual data of all wheel legs in the same motion stage and displacement range are detected to construct a multi-wheel leg residual matrix. The Pearson correlation coefficient between each wheel leg residual and all other wheel-rail residuals is calculated to mark several data points to be verified. Based on the rated current, stall current, maximum output torque of the joint, and maximum motion acceleration of the end effector of the tangential drive motor, the physical upper and lower limits of the current for each motion stage are constructed, and it is determined whether the current value of each data point to be verified is within the physical upper and lower limits of the current, so as to filter out the target data accordingly.

[0010] Furthermore, the step of sending a corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve includes: The no-load tangential motion reference curve is piecewise fitted and decoupled from physical parameters to detect the natural angular frequency, equivalent viscous damping coefficient and joint transmission clearance compensation value of the current wheel leg under no-load conditions, and a reference motion parameter set containing the individual mechanical characteristics of the wheel leg is generated accordingly. Based on the reference motion parameter set, with the natural angular frequency as the fundamental frequency, corresponding harmonics and harmonic components are superimposed to construct a composite sinusoidal excitation signal. The amplitude of each harmonic component is weighted and corrected according to the equivalent viscous damping coefficient. The phase of the composite sinusoidal excitation signal is pre-offset based on the joint transmission clearance compensation value to generate an initial motion command. The initial motion command is dynamically modified to generate the sinusoidal reciprocating motion command.

[0011] Furthermore, the step of dynamically correcting the initial motion command to generate the sinusoidal reciprocating motion command includes: Structural vibration spectrum data of transmission towers under environmental wind loads are collected to detect the first-order and second-order natural vibration frequencies of the tower body. A corresponding environmental vibration interference feature library of the transmission tower is constructed, and the vibration acceleration amplitude and phase angle at the current moment are calculated based on the environmental vibration interference feature library. The time-domain signal of the initial motion command is converted into a frequency-domain signal to filter out harmonic components that overlap with the inherent vibration frequency of the transmission tower. The remaining harmonic components are then phase-inversely canceled according to the vibration acceleration amplitude and the phase angle to generate an anti-interference motion command. During the tangential motion confidence process, the tangential joint current fluctuation data of the other stable ground-contact wheel legs are collected simultaneously to calculate the cross-correlation coefficient of the current fluctuation of the multiple wheel legs. When the cross-correlation coefficient is lower than the preset coefficient threshold, the anti-interference motion command is set to the sinusoidal reciprocating motion command.

[0012] Furthermore, the step of acquiring the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command to generate an actual current-displacement curve includes: Two equally spaced zero-velocity dwell segments are inserted within each sinusoidal motion cycle, and current data within each zero-velocity dwell segment are collected to calculate the average baseline current. The corresponding dynamic current baseline for the entire cycle is then generated using piecewise linear interpolation. Based on the dynamic current baseline, the tangential force data of the end effector and the encoder displacement data are collected, and the time difference ratio of the tangential force-displacement curve and the current-displacement curve is calculated accordingly. It is then determined whether the time difference ratio is within a preset ratio threshold. If the timing difference ratio is determined to be within the preset ratio threshold, a sliding window standard deviation algorithm is used to identify several extreme points in the current-displacement curve, and the extreme points are fitted to generate the corresponding actual current-displacement curve.

[0013] Furthermore, the step of calculating the characteristic parameters of the hysteresis loop area, nonlinearity coefficient, and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve includes: Using the extreme point of the no-load tangential motion reference curve as the anchor point, the phase cross-correlation algorithm is used to perform sub-pixel-level phase alignment on the actual current-displacement curve to calculate the coordinates of the intersection point of the two aligned curves, and the entire motion cycle is divided into a forward loading segment, a peak holding segment, a reverse unloading segment, and a valley holding segment based on the coordinates of the intersection point. Closed polygons are constructed for the forward loading end and the reverse unloading segment respectively. The area of ​​the two closed polygons is calculated by the adaptive Gauss-Legend integral method and the average value is taken to obtain the initial hysteresis loop area. The overlapping area of ​​the two curves in the peak holding segment and the valley holding segment caused by mechanical clearance is removed to generate the hysteresis loop area. Linear regression fitting is performed on the forward loading end and the reverse unloading section to calculate the ratio of the standard deviation of the fitting residual to the total amplitude of the curve, and the nonlinearity coefficient of each segment is obtained accordingly. The maximum current difference between the two curves at the same displacement point is detected and set as the peak current deviation.

[0014] The second aspect of the present invention proposes: A system for monitoring the operational status of wheeled equipment used in power transmission towers, wherein the system comprises: The data acquisition module is used to control the end effector of the current wheel leg to perform a reciprocating motion with constant amplitude along a preset tangential direction when each wheel leg of the wheel-stepping equipment is in a completely suspended state, and to collect the current-displacement mapping data of the drive motor to construct the no-load tangential motion reference curve of each wheel leg. The control module is used to control the normal joint output torque of the current wheel leg to zero after any wheel leg performs a ground contact action and reaches the preset target position, and to send the corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve. The calculation module is used to collect the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command, so as to generate the actual current-displacement curve and calculate the characteristic parameters of hysteresis loop area, nonlinearity coefficient and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve. The comparison module is used to compare the feature parameters with preset parameter thresholds, and when each feature parameter is lower than its corresponding parameter threshold, the current wheel leg is determined to be in a virtual leg state.

[0015] Furthermore, the acquisition module is specifically used for: When the wheel leg is completely suspended in the air and the fuselage remains stationary, a composite excitation signal is sent to the tangential drive joint, and the corresponding current, displacement and acceleration data are collected to separate the effective current component with the same frequency as the excitation signal through cross-correlation analysis. The complete motion cycle of the wheel-walking equipment is divided into several motion stages, and the slope and intercept of the current change with displacement in each motion stage are calculated to detect the viscous friction coefficient and equivalent damping coefficient of the tangential joint under different motion states, and a phased no-load dynamic model is constructed accordingly. Based on the no-load dynamic model, a theoretical no-load current-displacement curve is generated. The actual collected effective current data is then compared with the theoretical no-load current-displacement curve using residual analysis. Target data is selected accordingly, and the least squares method is used to fit the target data to construct the no-load tangential motion reference curve.

[0016] Furthermore, the acquisition module is specifically used for: Granger causality tests were performed on the current-displacement time series data for each motion stage to verify whether the current change was driven unidirectionally by the displacement change, and data segments with invalid causal relationships were removed to generate intermediate data. Within the intermediate data, residual data of all wheel legs in the same motion stage and displacement range are detected to construct a multi-wheel leg residual matrix. The Pearson correlation coefficient between each wheel leg residual and all other wheel-rail residuals is calculated to mark several data points to be verified. Based on the rated current, stall current, maximum output torque of the joint, and maximum motion acceleration of the end effector of the tangential drive motor, the physical upper and lower limits of the current for each motion stage are constructed, and it is determined whether the current value of each data point to be verified is within the physical upper and lower limits of the current, so as to filter out the target data accordingly.

[0017] Furthermore, the control module is specifically used for: The no-load tangential motion reference curve is piecewise fitted and decoupled from physical parameters to detect the natural angular frequency, equivalent viscous damping coefficient and joint transmission clearance compensation value of the current wheel leg under no-load conditions, and a reference motion parameter set containing the individual mechanical characteristics of the wheel leg is generated accordingly. Based on the reference motion parameter set, with the natural angular frequency as the fundamental frequency, corresponding harmonics and harmonic components are superimposed to construct a composite sinusoidal excitation signal. The amplitude of each harmonic component is weighted and corrected according to the equivalent viscous damping coefficient. The phase of the composite sinusoidal excitation signal is pre-offset based on the joint transmission clearance compensation value to generate an initial motion command. The initial motion command is dynamically modified to generate the sinusoidal reciprocating motion command.

[0018] Furthermore, the control module is specifically used for: Structural vibration spectrum data of transmission towers under environmental wind loads are collected to detect the first-order and second-order natural vibration frequencies of the tower body. A corresponding environmental vibration interference feature library of the transmission tower is constructed, and the vibration acceleration amplitude and phase angle at the current moment are calculated based on the environmental vibration interference feature library. The time-domain signal of the initial motion command is converted into a frequency-domain signal to filter out harmonic components that overlap with the inherent vibration frequency of the transmission tower. The remaining harmonic components are then phase-inversely canceled according to the vibration acceleration amplitude and the phase angle to generate an anti-interference motion command. During the tangential motion confidence process, the tangential joint current fluctuation data of the other stable ground-contact wheel legs are collected simultaneously to calculate the cross-correlation coefficient of the current fluctuation of the multiple wheel legs. When the cross-correlation coefficient is lower than the preset coefficient threshold, the anti-interference motion command is set to the sinusoidal reciprocating motion command.

[0019] Furthermore, the calculation module is specifically used for: Two equally spaced zero-velocity dwell segments are inserted within each sinusoidal motion cycle, and current data within each zero-velocity dwell segment are collected to calculate the average baseline current. The corresponding dynamic current baseline for the entire cycle is then generated using piecewise linear interpolation. Based on the dynamic current baseline, the tangential force data of the end effector and the encoder displacement data are collected, and the time difference ratio of the tangential force-displacement curve and the current-displacement curve is calculated accordingly. It is then determined whether the time difference ratio is within a preset ratio threshold. If the timing difference ratio is determined to be within the preset ratio threshold, a sliding window standard deviation algorithm is used to identify several extreme points in the current-displacement curve, and the extreme points are fitted to generate the corresponding actual current-displacement curve.

[0020] Furthermore, the calculation module is specifically used for: Using the extreme point of the no-load tangential motion reference curve as the anchor point, the phase cross-correlation algorithm is used to perform sub-pixel-level phase alignment on the actual current-displacement curve to calculate the coordinates of the intersection point of the two aligned curves, and the entire motion cycle is divided into a forward loading segment, a peak holding segment, a reverse unloading segment, and a valley holding segment based on the coordinates of the intersection point. Closed polygons are constructed for the forward loading end and the reverse unloading segment respectively. The area of ​​the two closed polygons is calculated by the adaptive Gauss-Legend integral method and the average value is taken to obtain the initial hysteresis loop area. The overlapping area of ​​the two curves in the peak holding segment and the valley holding segment caused by mechanical clearance is removed to generate the hysteresis loop area. Linear regression fitting is performed on the forward loading end and the reverse unloading section to calculate the ratio of the standard deviation of the fitting residual to the total amplitude of the curve, and the nonlinearity coefficient of each segment is obtained accordingly. The maximum current difference between the two curves at the same displacement point is detected and set as the peak current deviation.

[0021] The third aspect of the present invention proposes: A computer includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the above-described method for monitoring the working status of wheeled equipment for power transmission towers.

[0022] The fourth aspect of the present invention proposes: A readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the above-described method for monitoring the working status of wheeled equipment for power transmission towers.

[0023] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0024] Figure 1 A flowchart of a method for monitoring the working status of wheeled equipment for power transmission towers provided in the first embodiment of the present invention; Figure 2 This is a structural block diagram of a wheeled equipment working status monitoring system for power transmission towers provided in the third embodiment of the present invention.

[0025] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0026] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0027] It should be noted that when a component is said to be fixed to another component, it can be directly on the other component or there may be an intervening component. When a component is said to be connected to another component, it can be directly connected to the other component or there may be an intervening component. The terms vertical, horizontal, left, right, and similar expressions used in this document are for illustrative purposes only.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The terminology used herein includes, and / or encompasses, any and all combinations of one or more of the associated listed items.

[0029] Please see Figure 1The image shows a method for monitoring the working status of wheeled walking equipment for power transmission towers according to the first embodiment of the present invention. This method eliminates interference from normal loads by ensuring the normal torque returns to zero after ground contact. It uses only three characteristic parameters—hysteresis loop area, nonlinearity coefficient, and peak current deviation—extracted from existing drive motor current and encoder data for fusion judgment, eliminating the need for additional force sensors. This results in low hardware cost and strong real-time performance. Furthermore, each wheel leg independently constructs an unloaded reference curve, effectively eliminating identification deviations caused by individual differences in wheel legs, installation errors, and long-term wear. This significantly improves the accuracy and robustness of identifying "virtual legs," thereby ensuring the gait stability and walking safety of the wheeled walking equipment in complex terrain.

[0030] Specifically, this embodiment provides: A method for monitoring the working status of wheeled equipment for power transmission towers, wherein the method includes: Step S10: When each wheel leg of the wheel-stepping equipment is in a completely suspended state, control the current wheel leg end effector to perform a reciprocating motion with constant amplitude along a preset tangential direction, and collect the current-displacement mapping data of the drive motor accordingly to construct the no-load tangential motion reference curve of each wheel leg. It should be noted that when the wheel leg is completely suspended in the air, its motion resistance comes only from the inherent mechanical characteristics of the joint, such as viscous friction, transmission clearance, and motor internal resistance, and is not affected by external loads. Therefore, the current-displacement mapping data collected under these conditions exhibits extremely high stability and repeatability. This step simulates the actual motion direction after ground contact by controlling the wheel leg to reciprocate along the tangential direction of the tower's angle steel surface, ensuring the kinematic consistency between the reference curve and the actual operating conditions. Simultaneously, each wheel leg is individually calibrated to eliminate machining errors and assembly differences between different wheel legs, providing a precise individual benchmark for subsequent determination of the "virtual leg" (or "virtual leg").

[0031] Step S20: After any wheel leg performs a ground contact action and reaches the preset target position, control the normal joint output torque of the current wheel leg to be zero, and send the corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve. It should be noted that after the wheel leg reaches the preset contact position, the normal joint output torque is first controlled to zero to eliminate the interference of the normal preload on the tangential motion, ensuring that the resistance to the tangential motion comes only from the interaction between the ground and the end effector. Then, based on the no-load reference curve, a sinusoidal reciprocating motion command adapted to the individual characteristics of the wheel leg is generated, driving the wheel leg to perform small-amplitude, low-frequency tangential sliding along the angle steel surface. If the wheel leg is a solid leg, friction and normal pressure will be generated between the end effector and the angle steel surface, causing the drive motor to output a larger current to overcome the ground resistance. If the wheel leg is a dummy leg, the end effector is not in effective contact with the angle steel, the resistance to the tangential motion is consistent with the no-load state, and the motor current will also coincide with the no-load reference curve. Through this active excitation method, the contact state, which is difficult to detect directly, is transformed into the easily measurable difference in motor current, realizing the indirect characterization of the dummy leg state.

[0032] Step S30: Collect the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command to generate the actual current-displacement curve, and calculate the characteristic parameters of hysteresis loop area, nonlinearity coefficient and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve. It's important to note that when the wheel leg performs sinusoidal reciprocating motion, the change in current with displacement forms a closed hysteresis loop. The area of ​​the hysteresis loop reflects the energy dissipation during the motion: in the solid leg state, ground friction and plastic deformation consume a large amount of energy, resulting in a larger hysteresis loop area; in the virtual leg state, energy dissipation only comes from internal joint friction, resulting in a smaller hysteresis loop area. The nonlinearity coefficient reflects the linearity of the current-displacement relationship: in the solid leg state, ground friction exhibits nonlinear characteristics with changes in normal force and sliding speed, resulting in a higher nonlinearity coefficient; in the virtual leg state, joint friction characteristics are relatively stable, resulting in a lower nonlinearity coefficient. The peak current deviation reflects the difference in the maximum current during motion: in the solid leg state, ground resistance needs to be overcome, resulting in a significantly higher peak current than in the no-load state; in the virtual leg state, the peak current is basically consistent with the no-load state. By extracting these three complementary characteristic parameters, the contact state of the wheel leg can be comprehensively and accurately characterized.

[0033] Step S40: Compare the feature parameters with preset parameter thresholds, and when each feature parameter is lower than its corresponding parameter threshold, determine that the current wheel leg is in a virtual leg state.

[0034] It should be noted that the preset parameter thresholds are derived from statistical analysis of a large number of solid and dummy leg samples, and are dynamically updated as the mechanical characteristics of the wheel leg drift. A wheel is considered dummy only when all three characteristic parameters—hysteresis loop area, nonlinearity coefficient, and peak current deviation—are below their respective thresholds; if any one parameter is above its threshold, it is considered solid. This multi-parameter joint determination method effectively avoids misjudgments caused by interference from a single parameter, significantly improving the reliability of the detection.

[0035] Second Embodiment Furthermore, the step of acquiring the current-displacement mapping data of the corresponding drive motor to construct the no-load tangential motion reference curve of each wheel leg includes: When the wheel leg is completely suspended in the air and the fuselage remains stationary, a composite excitation signal is sent to the tangential drive joint, and the corresponding current, displacement and acceleration data are collected to separate the effective current component with the same frequency as the excitation signal through cross-correlation analysis. The complete motion cycle of the wheel-walking equipment is divided into several motion stages, and the slope and intercept of the current change with displacement in each motion stage are calculated to detect the viscous friction coefficient and equivalent damping coefficient of the tangential joint under different motion states, and a phased no-load dynamic model is constructed accordingly. Based on the no-load dynamic model, a theoretical no-load current-displacement curve is generated. The actual collected effective current data is then compared with the theoretical no-load current-displacement curve using residual analysis. Target data is selected accordingly, and the least squares method is used to fit the target data to construct the no-load tangential motion reference curve.

[0036] It should be noted that the motor current signal of wheeled walking equipment contains a large amount of noise components, such as electromagnetic interference, power fluctuations, and body vibration. This noise can seriously affect the accuracy of the baseline curve. This step uses a composite excitation signal containing the fundamental frequency, second harmonic, and third harmonic to fully excite the dynamic characteristics of the wheel legs at different frequencies. Then, through cross-correlation analysis, the collected current signal and the excitation signal are cross-correlated to extract the effective current component with the same frequency as the excitation signal. This effectively suppresses the influence of random noise and interference signals, and improves the signal-to-noise ratio.

[0037] The tangential motion of the wheel leg can be divided into three stages: acceleration, constant speed, and deceleration / braking. The frictional characteristics differ significantly between these stages: the acceleration stage is primarily affected by static friction, the constant speed stage by viscous friction, and the braking stage by both friction and damping. Traditional single dynamic models cannot accurately describe these stage characteristics, leading to significant errors in the baseline curve. This step divides the motion cycle into three stages, calculates the slope and intercept of the current-displacement curve for each stage, and then solves for the corresponding viscous friction coefficient and equivalent damping coefficient. Constructing a staged no-load dynamic model more accurately reflects the actual motion characteristics of the wheel leg.

[0038] This step first generates a theoretical no-load current-displacement curve based on a phased dynamic model. Then, it calculates the residual between the actual effective current data and the theoretical curve. Data points with excessively large residuals are usually caused by sudden disturbances or abnormal movements and are discarded. Finally, the least squares method is used to perform piecewise fitting on the filtered target data to obtain a continuous and smooth no-load tangential motion reference curve. This reference curve not only reflects the average mechanical characteristics of the wheel leg but also preserves the characteristic differences in different motion stages, providing a reliable reference standard for subsequent virtual leg detection.

[0039] Furthermore, the step of generating a theoretical no-load current-displacement curve based on the no-load dynamic model, and performing residual analysis between the actual collected effective current data and the theoretical no-load current-displacement curve to filter out the target data includes: Granger causality tests were performed on the current-displacement time series data for each motion stage to verify whether the current change was driven unidirectionally by the displacement change, and data segments with invalid causal relationships were removed to generate intermediate data. Within the intermediate data, residual data of all wheel legs in the same motion stage and displacement range are detected to construct a multi-wheel leg residual matrix. The Pearson correlation coefficient between each wheel leg residual and all other wheel-rail residuals is calculated to mark several data points to be verified. Based on the rated current, stall current, maximum output torque of the joint, and maximum motion acceleration of the end effector of the tangential drive motor, the physical upper and lower limits of the current for each motion stage are constructed, and it is determined whether the current value of each data point to be verified is within the physical upper and lower limits of the current, so as to filter out the target data accordingly.

[0040] It should be noted that, ideally, changes in motor current should be driven unidirectionally by changes in displacement. If current changes precede displacement changes, or if there is no causal relationship between the two, it indicates that the data has been affected by external interference, such as electromagnetic pulses or motor controller malfunctions. This step employs the Granger causality test to verify the causal relationship between the current and displacement time-series data for each motion stage. Only when the test results show that the displacement change is a Granger cause of the current change is the data segment retained; otherwise, it is discarded. Through causality testing, interference data unrelated to motion can be effectively filtered out, ensuring the causal rationality of the data.

[0041] Multiple wheel legs of the same equipment share similar mechanical designs and manufacturing processes. Their residuals within the same motion phase and displacement range should exhibit strong correlation. If the correlation between the residual of a particular wheel leg and the residuals of other wheel legs is significantly low, it indicates a potential anomaly in that data point. This step constructs a multi-wheel leg residual matrix, calculates the Pearson correlation coefficient between the residual of each wheel leg and the residuals of other wheel legs, and marks data points with correlation coefficients below a preset threshold as data points to be verified, providing a basis for subsequent physical boundary verification.

[0042] The motor current cannot exceed its stall current, nor can it fall below the minimum no-load current; simultaneously, according to Newton's second law, there is a definite physical relationship between current and acceleration. This step constructs the physical upper and lower limits of the current for each motion stage based on the motor's rated parameters and kinematic parameters. For data points to be verified, only those whose current values ​​are within the physical boundaries are retained; otherwise, they are considered abnormal data and discarded. Through three levels of data filtering, the final target data has extremely high reliability and can accurately reflect the true no-load characteristics of the wheel leg.

[0043] Furthermore, the step of sending a corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve includes: The no-load tangential motion reference curve is piecewise fitted and decoupled from physical parameters to detect the natural angular frequency, equivalent viscous damping coefficient and joint transmission clearance compensation value of the current wheel leg under no-load conditions, and a reference motion parameter set containing the individual mechanical characteristics of the wheel leg is generated accordingly. Based on the reference motion parameter set, with the natural angular frequency as the fundamental frequency, corresponding harmonics and harmonic components are superimposed to construct a composite sinusoidal excitation signal. The amplitude of each harmonic component is weighted and corrected according to the equivalent viscous damping coefficient. The phase of the composite sinusoidal excitation signal is pre-offset based on the joint transmission clearance compensation value to generate an initial motion command. The initial motion command is dynamically modified to generate the sinusoidal reciprocating motion command.

[0044] It should be noted that the unloaded tangential motion reference curve contains all the dynamic characteristic information of the wheel leg. By performing piecewise fitting and system identification, three key parameters can be decoupled: natural angular frequency, equivalent viscous damping coefficient, and joint transmission clearance compensation value. The natural angular frequency reflects the resonance characteristics of the wheel leg, the equivalent viscous damping coefficient reflects the energy dissipation characteristics of the joint, and the joint transmission clearance compensation value reflects the clearance size of the transmission system. These parameters have significant individual differences and are the basis for generating personalized motion commands.

[0045] This step uses the natural angular frequency of the wheel leg as the fundamental frequency, superimposing second and third harmonic components to construct a composite sinusoidal excitation signal, which can comprehensively excite the interaction between the wheel leg and the ground. Simultaneously, the amplitude of each harmonic component is weighted and corrected according to the equivalent viscous damping coefficient; the greater the damping, the larger the amplitude correction coefficient, ensuring that wheel legs with different damping characteristics can generate sufficient motion amplitude. Furthermore, the phase of the excitation signal is pre-shifted according to the joint transmission clearance compensation value to compensate for motion lag caused by transmission clearance, improving the accuracy of motion control. This personalized command generation method can effectively excite the current difference between the real and virtual leg states, improving detection sensitivity.

[0046] The initial motion command only considered the individual mechanical characteristics of the wheel legs and did not take into account the dynamic interference of the working environment. Further dynamic correction is needed to adapt to the complex high-altitude working environment.

[0047] Furthermore, the step of dynamically correcting the initial motion command to generate the sinusoidal reciprocating motion command includes: Structural vibration spectrum data of transmission towers under environmental wind loads are collected to detect the first-order and second-order natural vibration frequencies of the tower body. A corresponding environmental vibration interference feature library of the transmission tower is constructed, and the vibration acceleration amplitude and phase angle at the current moment are calculated based on the environmental vibration interference feature library. The time-domain signal of the initial motion command is converted into a frequency-domain signal to filter out harmonic components that overlap with the inherent vibration frequency of the transmission tower. The remaining harmonic components are then phase-inversely canceled according to the vibration acceleration amplitude and the phase angle to generate an anti-interference motion command. During the tangential motion confidence process, the tangential joint current fluctuation data of the other stable ground-contact wheel legs are collected simultaneously to calculate the cross-correlation coefficient of the current fluctuation of the multiple wheel legs. When the cross-correlation coefficient is lower than the preset coefficient threshold, the anti-interference motion command is set to the sinusoidal reciprocating motion command.

[0048] It should be noted that transmission towers experience low-frequency, high-amplitude vibrations under wind loads. These vibrations primarily occur near their first and second natural frequencies, typically between 0.5 and 3 Hz, overlapping with the frequency range of the wheel leg tangential motion, which can severely interfere with current signals. This step involves using accelerometers on the equipment to collect the tower's vibration spectrum data in real time, detecting the current first and second natural vibration frequencies, vibration acceleration amplitude, and phase angle. This data is then used to construct a dynamically updated environmental vibration interference feature library, providing a basis for subsequent interference cancellation.

[0049] This step first converts the initial motion command from the time domain to the frequency domain using a Fast Fourier Transform (FFT), filtering out harmonic components that overlap with the inherent vibration frequency of the tower. Then, based on the current vibration acceleration amplitude and phase angle, a compensation signal with the same amplitude but opposite phase as the vibration signal is generated and superimposed on the remaining harmonic components, thus actively canceling the tower vibration. This combination of frequency domain filtering and time domain compensation reduces the impact of tower vibration on the wheel leg movement by more than 90%, significantly improving the detection's anti-interference capability.

[0050] If the tower vibration interference is not completely canceled, the current of all ground-contacting wheel legs will fluctuate synchronously, causing a significant increase in the cross-correlation coefficient of the current fluctuations across multiple wheel legs. This step simultaneously collects the current data of the remaining stable ground-contacting wheel legs and calculates their cross-correlation coefficients. When the cross-correlation coefficient is below a preset threshold, it indicates that the vibration interference has been effectively canceled, and the anti-interference motion command is used as the final sinusoidal reciprocating motion command. Otherwise, the interference cancellation process is repeated until the requirements are met. Through multi-wheel leg cross-verification, the anti-interference effect of the command can be ensured, avoiding misjudgments caused by residual interference.

[0051] Furthermore, the step of acquiring the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command to generate an actual current-displacement curve includes: Two equally spaced zero-velocity dwell segments are inserted within each sinusoidal motion cycle, and current data within each zero-velocity dwell segment are collected to calculate the average baseline current. The corresponding dynamic current baseline for the entire cycle is then generated using piecewise linear interpolation. Based on the dynamic current baseline, the tangential force data of the end effector and the encoder displacement data are collected, and the time difference ratio of the tangential force-displacement curve and the current-displacement curve is calculated accordingly. It is then determined whether the time difference ratio is within a preset ratio threshold. If the timing difference ratio is determined to be within the preset ratio threshold, a sliding window standard deviation algorithm is used to identify several extreme points in the current-displacement curve, and the extreme points are fitted to generate the corresponding actual current-displacement curve.

[0052] It should be noted that the motor's no-load current increases with temperature, causing the current baseline to drift. Using a fixed baseline would introduce significant systematic errors. This step inserts two zero-speed pauses at the peak and trough positions of each sinusoidal motion cycle. During these pauses, the motor only needs to overcome static friction to remain stationary, and the current value becomes the current baseline. The average baseline current is calculated by collecting current data from these two pauses, and then a piecewise linear interpolation method is used to generate a dynamic current baseline for the entire cycle. This dynamic baseline correction method can track baseline drift caused by temperature changes in real time, controlling the baseline error to within 1%.

[0053] According to the motor's torque balance equation, the motor current is proportional to the output torque, and the output torque is proportional to the end tangential force. Therefore, the rate of change of current should maintain a fixed ratio with the rate of change of tangential force. This step simultaneously collects data from the tangential force sensor (for calibration and verification) and the encoder, calculating the time-series difference ratio between the tangential force-displacement curve and the current-displacement curve. If the ratio is within a preset threshold, the data is valid; otherwise, the data has been interfered with and needs to be re-collected. Time-series difference verification further eliminates abnormal data, ensuring the accuracy of the actual curves.

[0054] The original current data still contains high-frequency noise, resulting in a very coarse curve that affects subsequent characteristic parameter calculations. This step uses a sliding window standard deviation algorithm to identify extreme points in the curve, which contain the main morphological information of the curve. Then, cubic spline interpolation is used to fit these extreme points, obtaining a smooth and continuous actual current-displacement curve. This curve retains the key features of the original data while eliminating the influence of high-frequency noise, providing a high-quality data foundation for subsequent characteristic parameter calculations.

[0055] Furthermore, the step of calculating the characteristic parameters of the hysteresis loop area, nonlinearity coefficient, and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve includes: Using the extreme point of the no-load tangential motion reference curve as the anchor point, the phase cross-correlation algorithm is used to perform sub-pixel-level phase alignment on the actual current-displacement curve to calculate the coordinates of the intersection point of the two aligned curves, and the entire motion cycle is divided into a forward loading segment, a peak holding segment, a reverse unloading segment, and a valley holding segment based on the coordinates of the intersection point. Closed polygons are constructed for the forward loading end and the reverse unloading segment respectively. The area of ​​the two closed polygons is calculated by the adaptive Gauss-Legend integral method and the average value is taken to obtain the initial hysteresis loop area. The overlapping area of ​​the two curves in the peak holding segment and the valley holding segment caused by mechanical clearance is removed to generate the hysteresis loop area. Linear regression fitting is performed on the forward loading end and the reverse unloading section to calculate the ratio of the standard deviation of the fitting residual to the total amplitude of the curve, and the nonlinearity coefficient of each segment is obtained accordingly. The maximum current difference between the two curves at the same displacement point is detected and set as the peak current deviation.

[0056] It should be noted that due to factors such as transmission backlash and control delay, there may be a phase shift between the actual current-displacement curve and the no-load reference curve. Direct feature calculation would lead to significant errors in parameters such as hysteresis loop area and peak current deviation. This step uses the extreme points of the no-load reference curve as anchor points and employs a phase cross-correlation algorithm to perform sub-pixel-level phase alignment of the actual curves, controlling the phase difference between the two curves within 0.1 rad. Then, based on the coordinates of the intersection points of the aligned two curves, the entire motion cycle is divided into four stages: forward loading segment, peak holding segment, reverse unloading segment, and valley holding segment, laying the foundation for subsequent staged feature calculations.

[0057] The area of ​​the closed region between the forward loading segment and the reverse unloading segment is the hysteresis loop area, reflecting the energy dissipation during motion. This step uses the adaptive Gauss-Legendary integral method to calculate the area of ​​the closed polygon, which has higher computational accuracy and efficiency compared to the traditional trapezoidal integral method. Meanwhile, due to the presence of mechanical backlash, the two curves of the peak holding segment and the valley holding segment will overlap. This overlapping area is not caused by ground resistance and needs to be removed. Through staged integration and overlapping area removal, the true hysteresis loop area can be accurately calculated, reflecting the energy dissipation caused by ground resistance.

[0058] The nonlinearity coefficient is characterized by the ratio of the standard deviation of the fitted residuals to the total amplitude of the curve. A larger ratio indicates a higher degree of nonlinearity in the current-displacement relationship. The peak current deviation is the maximum current difference between the two curves at the same displacement point, reflecting the current increment caused by ground resistance. The final nonlinearity coefficient is obtained by averaging the nonlinearity coefficients of the forward loading and reverse unloading segments. Through the comprehensive calculation of these three characteristic parameters, the contact state of the wheel legs can be comprehensively and accurately characterized, providing a reliable basis for determining whether a wheel leg is faulty.

[0059] Please see Figure 2 The third embodiment of the present invention provides: A system for monitoring the operational status of wheeled equipment used in power transmission towers, wherein the system comprises: The data acquisition module is used to control the end effector of the current wheel leg to perform a reciprocating motion with constant amplitude along a preset tangential direction when each wheel leg of the wheel-stepping equipment is in a completely suspended state, and to collect the current-displacement mapping data of the drive motor to construct the no-load tangential motion reference curve of each wheel leg. The control module is used to control the normal joint output torque of the current wheel leg to zero after any wheel leg performs a ground contact action and reaches the preset target position, and to send the corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve. The calculation module is used to collect the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command, so as to generate the actual current-displacement curve and calculate the characteristic parameters of hysteresis loop area, nonlinearity coefficient and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve. The comparison module is used to compare the feature parameters with preset parameter thresholds, and when each feature parameter is lower than its corresponding parameter threshold, the current wheel leg is determined to be in a virtual leg state.

[0060] Furthermore, the acquisition module is specifically used for: When the wheel leg is completely suspended in the air and the fuselage remains stationary, a composite excitation signal is sent to the tangential drive joint, and the corresponding current, displacement and acceleration data are collected to separate the effective current component with the same frequency as the excitation signal through cross-correlation analysis. The complete motion cycle of the wheel-walking equipment is divided into several motion stages, and the slope and intercept of the current change with displacement in each motion stage are calculated to detect the viscous friction coefficient and equivalent damping coefficient of the tangential joint under different motion states, and a phased no-load dynamic model is constructed accordingly. Based on the no-load dynamic model, a theoretical no-load current-displacement curve is generated. The actual collected effective current data is then compared with the theoretical no-load current-displacement curve using residual analysis. Target data is selected accordingly, and the least squares method is used to fit the target data to construct the no-load tangential motion reference curve.

[0061] Furthermore, the acquisition module is specifically used for: Granger causality tests were performed on the current-displacement time series data for each motion stage to verify whether the current change was driven unidirectionally by the displacement change, and data segments with invalid causal relationships were removed to generate intermediate data. Within the intermediate data, residual data of all wheel legs in the same motion stage and displacement range are detected to construct a multi-wheel leg residual matrix. The Pearson correlation coefficient between each wheel leg residual and all other wheel-rail residuals is calculated to mark several data points to be verified. Based on the rated current, stall current, maximum output torque of the joint, and maximum motion acceleration of the end effector of the tangential drive motor, the physical upper and lower limits of the current for each motion stage are constructed, and it is determined whether the current value of each data point to be verified is within the physical upper and lower limits of the current, so as to filter out the target data accordingly.

[0062] Furthermore, the control module is specifically used for: The no-load tangential motion reference curve is piecewise fitted and decoupled from physical parameters to detect the natural angular frequency, equivalent viscous damping coefficient and joint transmission clearance compensation value of the current wheel leg under no-load conditions, and a reference motion parameter set containing the individual mechanical characteristics of the wheel leg is generated accordingly. Based on the reference motion parameter set, with the natural angular frequency as the fundamental frequency, corresponding harmonics and harmonic components are superimposed to construct a composite sinusoidal excitation signal. The amplitude of each harmonic component is weighted and corrected according to the equivalent viscous damping coefficient. The phase of the composite sinusoidal excitation signal is pre-offset based on the joint transmission clearance compensation value to generate an initial motion command. The initial motion command is dynamically modified to generate the sinusoidal reciprocating motion command.

[0063] Furthermore, the control module is specifically used for: Structural vibration spectrum data of transmission towers under environmental wind loads are collected to detect the first-order and second-order natural vibration frequencies of the tower body. A corresponding environmental vibration interference feature library of the transmission tower is constructed, and the vibration acceleration amplitude and phase angle at the current moment are calculated based on the environmental vibration interference feature library. The time-domain signal of the initial motion command is converted into a frequency-domain signal to filter out harmonic components that overlap with the inherent vibration frequency of the transmission tower. The remaining harmonic components are then phase-inversely canceled according to the vibration acceleration amplitude and the phase angle to generate an anti-interference motion command. During the tangential motion confidence process, the tangential joint current fluctuation data of the other stable ground-contact wheel legs are collected simultaneously to calculate the cross-correlation coefficient of the current fluctuation of the multiple wheel legs. When the cross-correlation coefficient is lower than the preset coefficient threshold, the anti-interference motion command is set to the sinusoidal reciprocating motion command.

[0064] Furthermore, the calculation module is specifically used for: Two equally spaced zero-velocity dwell segments are inserted within each sinusoidal motion cycle, and current data within each zero-velocity dwell segment are collected to calculate the average baseline current. The corresponding dynamic current baseline for the entire cycle is then generated using piecewise linear interpolation. Based on the dynamic current baseline, the tangential force data of the end effector and the encoder displacement data are collected, and the time difference ratio of the tangential force-displacement curve and the current-displacement curve is calculated accordingly. It is then determined whether the time difference ratio is within a preset ratio threshold. If the timing difference ratio is determined to be within the preset ratio threshold, a sliding window standard deviation algorithm is used to identify several extreme points in the current-displacement curve, and the extreme points are fitted to generate the corresponding actual current-displacement curve.

[0065] Furthermore, the calculation module is specifically used for: Using the extreme point of the no-load tangential motion reference curve as the anchor point, the phase cross-correlation algorithm is used to perform sub-pixel-level phase alignment on the actual current-displacement curve to calculate the coordinates of the intersection point of the two aligned curves, and the entire motion cycle is divided into a forward loading segment, a peak holding segment, a reverse unloading segment, and a valley holding segment based on the coordinates of the intersection point. Closed polygons are constructed for the forward loading end and the reverse unloading segment respectively. The area of ​​the two closed polygons is calculated by the adaptive Gauss-Legend integral method and the average value is taken to obtain the initial hysteresis loop area. The overlapping area of ​​the two curves in the peak holding segment and the valley holding segment caused by mechanical clearance is removed to generate the hysteresis loop area. Linear regression fitting is performed on the forward loading end and the reverse unloading section to calculate the ratio of the standard deviation of the fitting residual to the total amplitude of the curve, and the nonlinearity coefficient of each segment is obtained accordingly. The maximum current difference between the two curves at the same displacement point is detected and set as the peak current deviation.

[0066] The fourth embodiment of the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for monitoring the working status of wheeled equipment for power transmission towers.

[0067] The fifth embodiment of the present invention provides a readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, it implements the above-described method for monitoring the working status of wheeled equipment for power transmission towers.

[0068] In summary, the method and system for monitoring the working status of wheeled equipment for power transmission towers provided by the above embodiments of the present invention can eliminate normal load interference by eliminating the normal torque after ground contact. It performs fusion judgment based solely on three characteristic parameters extracted from existing drive motor current and encoder data: hysteresis loop area, nonlinearity coefficient, and peak current deviation. No additional force sensor is required, resulting in low hardware cost and strong real-time performance. Furthermore, each wheel leg independently constructs an unloaded reference curve, effectively eliminating identification deviations caused by individual differences in wheel legs, installation errors, and long-term wear. This significantly improves the accuracy and robustness of identifying "virtual legs," thereby ensuring the gait stability and walking safety of wheeled equipment in complex terrain.

[0069] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0070] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, a computer-readable medium can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0071] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0072] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] In the description of this specification, references to the terms "an embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0074] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for monitoring the working status of wheeled equipment used in power transmission towers, characterized in that, The method includes: When each wheel leg of the wheel-stepping equipment is in a completely suspended state, the end effector of the current wheel leg is controlled to perform a reciprocating motion with a constant amplitude along a preset tangential direction, and the current-displacement mapping data of the drive motor is collected accordingly to construct the no-load tangential motion reference curve of each wheel leg. After any wheel leg performs a ground contact action and reaches the preset target position, the normal joint output torque of the current wheel leg is controlled to be zero, and a corresponding sinusoidal reciprocating motion command is sent to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve. According to the sinusoidal reciprocating motion command, the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg are collected to generate the actual current-displacement curve, and the characteristic parameters of hysteresis loop area, nonlinearity coefficient and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve are calculated. The feature parameters are compared with preset parameter thresholds, and when each feature parameter is lower than its corresponding parameter threshold, the current wheel leg is determined to be in a virtual leg state.

2. The method for monitoring the working status of wheeled equipment for power transmission towers according to claim 1, characterized in that, The step of acquiring the current-displacement mapping data of the corresponding drive motor to construct the no-load tangential motion reference curve of each wheel leg includes: When the wheel leg is completely suspended in the air and the fuselage remains stationary, a composite excitation signal is sent to the tangential drive joint, and the corresponding current, displacement and acceleration data are collected to separate the effective current component with the same frequency as the excitation signal through cross-correlation analysis. The complete motion cycle of the wheel-walking equipment is divided into several motion stages, and the slope and intercept of the current change with displacement in each motion stage are calculated to detect the viscous friction coefficient and equivalent damping coefficient of the tangential joint under different motion states, and a phased no-load dynamic model is constructed accordingly. Based on the no-load dynamic model, a theoretical no-load current-displacement curve is generated. The actual collected effective current data is then compared with the theoretical no-load current-displacement curve using residual analysis. Target data is selected accordingly, and the least squares method is used to fit the target data to construct the no-load tangential motion reference curve.

3. The method for monitoring the working status of wheeled equipment for power transmission towers according to claim 2, characterized in that, The steps of generating a theoretical no-load current-displacement curve based on the no-load dynamic model, performing residual analysis between the actual collected effective current data and the theoretical no-load current-displacement curve, and correspondingly filtering out the target data include: Granger causality tests were performed on the current-displacement time series data for each motion stage to verify whether the current change was driven unidirectionally by the displacement change, and data segments with invalid causal relationships were removed to generate intermediate data. Within the intermediate data, residual data of all wheel legs in the same motion stage and displacement range are detected to construct a multi-wheel leg residual matrix. The Pearson correlation coefficient between each wheel leg residual and all other wheel-rail residuals is calculated to mark several data points to be verified. Based on the rated current, stall current, maximum output torque of the joint, and maximum motion acceleration of the end effector of the tangential drive motor, the physical upper and lower limits of the current for each motion stage are constructed, and it is determined whether the current value of each data point to be verified is within the physical upper and lower limits of the current, so as to filter out the target data accordingly.

4. The method for monitoring the working status of wheeled equipment for power transmission towers according to claim 1, characterized in that, The step of sending the corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve includes: The no-load tangential motion reference curve is piecewise fitted and decoupled from physical parameters to detect the natural angular frequency, equivalent viscous damping coefficient and joint transmission clearance compensation value of the current wheel leg under no-load conditions, and a reference motion parameter set containing the individual mechanical characteristics of the wheel leg is generated accordingly. Based on the reference motion parameter set, with the natural angular frequency as the fundamental frequency, corresponding harmonics and harmonic components are superimposed to construct a composite sinusoidal excitation signal. The amplitude of each harmonic component is weighted and corrected according to the equivalent viscous damping coefficient. The phase of the composite sinusoidal excitation signal is pre-offset based on the joint transmission clearance compensation value to generate an initial motion command. The initial motion command is dynamically modified to generate the sinusoidal reciprocating motion command.

5. The method for monitoring the working status of wheeled equipment for power transmission towers according to claim 4, characterized in that, The step of dynamically correcting the initial motion command to generate the sinusoidal reciprocating motion command includes: Structural vibration spectrum data of transmission towers under environmental wind loads are collected to detect the first-order and second-order natural vibration frequencies of the tower body. A corresponding environmental vibration interference feature library of the transmission tower is constructed, and the vibration acceleration amplitude and phase angle at the current moment are calculated based on the environmental vibration interference feature library. The time-domain signal of the initial motion command is converted into a frequency-domain signal to filter out harmonic components that overlap with the inherent vibration frequency of the transmission tower. The remaining harmonic components are then phase-inversely canceled according to the vibration acceleration amplitude and the phase angle to generate an anti-interference motion command. During the tangential motion confidence process, the tangential joint current fluctuation data of the other stable ground-contact wheel legs are collected simultaneously to calculate the cross-correlation coefficient of the current fluctuation of the multiple wheel legs. When the cross-correlation coefficient is lower than the preset coefficient threshold, the anti-interference motion command is set to the sinusoidal reciprocating motion command.

6. The method for monitoring the working status of wheeled equipment for power transmission towers according to claim 1, characterized in that, The step of acquiring the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command to generate the actual current-displacement curve includes: Two equally spaced zero-velocity dwell segments are inserted within each sinusoidal motion cycle, and current data within each zero-velocity dwell segment are collected to calculate the average baseline current. The corresponding dynamic current baseline for the entire cycle is then generated using piecewise linear interpolation. Based on the dynamic current baseline, the tangential force data of the end effector and the encoder displacement data are collected, and the time difference ratio of the tangential force-displacement curve and the current-displacement curve is calculated accordingly. It is then determined whether the time difference ratio is within a preset ratio threshold. If the timing difference ratio is determined to be within the preset ratio threshold, a sliding window standard deviation algorithm is used to identify several extreme points in the current-displacement curve, and the extreme points are fitted to generate the corresponding actual current-displacement curve.

7. The method for monitoring the working status of wheeled equipment for power transmission towers according to claim 1, characterized in that, The steps for calculating the characteristic parameters of the hysteresis loop area, nonlinearity coefficient, and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve include: Using the extreme point of the no-load tangential motion reference curve as the anchor point, the phase cross-correlation algorithm is used to perform sub-pixel-level phase alignment on the actual current-displacement curve to calculate the coordinates of the intersection point of the two aligned curves, and the entire motion cycle is divided into a forward loading segment, a peak holding segment, a reverse unloading segment, and a valley holding segment based on the coordinates of the intersection point. Closed polygons are constructed for the forward loading end and the reverse unloading segment respectively. The area of ​​the two closed polygons is calculated by the adaptive Gauss-Legend integral method and the average value is taken to obtain the initial hysteresis loop area. The overlapping area of ​​the two curves in the peak holding segment and the valley holding segment caused by mechanical clearance is removed to generate the hysteresis loop area. Linear regression fitting is performed on the forward loading end and the reverse unloading section to calculate the ratio of the standard deviation of the fitting residual to the total amplitude of the curve, and the nonlinearity coefficient of each segment is obtained accordingly. The maximum current difference between the two curves at the same displacement point is detected and set as the peak current deviation.

8. A monitoring system for the working status of wheeled equipment used in power transmission towers, characterized in that, The system includes: The data acquisition module is used to control the end effector of the current wheel leg to perform a reciprocating motion with constant amplitude along a preset tangential direction when each wheel leg of the wheel-stepping equipment is in a completely suspended state, and to collect the current-displacement mapping data of the drive motor to construct the no-load tangential motion reference curve of each wheel leg. The control module is used to control the normal joint output torque of the current wheel leg to zero after any wheel leg performs a ground contact action and reaches the preset target position, and to send the corresponding sinusoidal reciprocating motion command to the tangential drive joint of the current wheel leg based on the no-load tangential motion reference curve. The calculation module is used to collect the actual current data and encoder displacement data of the tangential drive joint of the current wheel leg according to the sinusoidal reciprocating motion command, so as to generate the actual current-displacement curve and calculate the characteristic parameters of hysteresis loop area, nonlinearity coefficient and peak current deviation between the current-displacement curve and the no-load tangential motion reference curve. The comparison module is used to compare the feature parameters with preset parameter thresholds, and when each feature parameter is lower than its corresponding parameter threshold, the current wheel leg is determined to be in a virtual leg state.

9. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for monitoring the working status of wheeled equipment for power transmission towers as described in any one of claims 1 to 7.

10. A readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for monitoring the working status of wheeled equipment for power transmission towers as described in any one of claims 1 to 7.

Images