Conductor galloping monitoring method and related devices

Abstract

The application provides a conductor galloping monitoring method and related equipment. The method comprises the following steps: acquiring the angular velocity and acceleration of the conductor at each sampling time; calculating the first sampling angle of the conductor at the nth sampling time based on the angular velocity and acceleration of the conductor at the nth sampling time; calculating the predicted angle at the (n+1)th sampling time based on the first sampling angle at the nth sampling time; calculating the second sampling angle of the conductor at the (n+1)th sampling time based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time; calculating the galloping angle at the (n+1)th sampling time based on the second sampling angle at the (n+1)th sampling time and the predicted angle at the (n+1)th sampling time; determining that the galloping state of the conductor is abnormal, and outputting warning information. The application calculates the galloping angle of the conductor through the angular velocity and acceleration data of the conductor, realizes the fusion and complementation of multiple data sources, and can improve the accuracy and comprehensiveness of the galloping angle monitoring.

Description

Technical Field

[0001] This application relates to the field of line monitoring, and in particular to a method and related equipment for monitoring conductor galloping. Background Technology

[0002] Transmission lines are numerous and widely distributed, and subject to complex weather conditions, constantly facing the threat of natural disasters such as icing. The conductor galloping caused by the combined effects of wind and snow increases mechanical stress, potentially leading to fatigue damage, wire breakage, and other mechanical failures. During galloping, conductors colliding with each other or coming into contact with other equipment can cause electrical faults such as short circuits and power outages. Furthermore, conductor galloping can cause conductors to detach, endangering the safety of personnel and equipment on the ground, and increasing maintenance difficulty and costs. Using high-precision sensors, data acquisition systems, and wireless transmission technology to monitor conductor galloping in real time and promptly detect and warn of potential safety hazards has become a mainstream preventative measure in the market. This effectively controls and reduces the impact of conductor galloping, ensuring the safe and reliable operation of the transmission system. In the construction of modern smart grids, the application of conductor galloping monitoring devices will play a crucial role in improving the safety and stability of grid operation.

[0003] Current conductor galloping monitoring solutions typically involve installing angular velocity and accelerometer sensors at multiple monitoring points along the conductor. By collecting angular velocity and acceleration data from each monitoring point, the galloping information of the conductor is obtained through calculation. However, because angular velocity and accelerometer sensors are greatly affected by external environmental factors, the monitoring accuracy and sensitivity are insufficient. Summary of the Invention

[0004] To address the problems in the prior art, this application provides a method and related equipment for monitoring conductor galloping, thereby improving the accuracy and comprehensiveness of the galloping angle.

[0005] This application provides a method for monitoring conductor galloping, including:

[0006] Obtain the angular velocity and acceleration of the conductor at each sampling time;

[0007] Based on the angular velocity and acceleration of the conductor at the nth sampling time, calculate the first sampling angle of the conductor at the nth sampling time; where n is a positive integer;

[0008] Based on the first sampling angle of the conductor at the nth sampling time, calculate the predicted angle of the conductor at the (n+1)th sampling time;

[0009] Based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time, calculate the second sampling angle of the conductor at the (n+1)th sampling time;

[0010] Based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time, the galloping angle of the conductor at the (n+1)th sampling time is calculated.

[0011] If the dancing angle of the conductor at the (n+1)th sampling time is used to determine that the dancing state of the conductor is abnormal, a warning message is output.

[0012] In one embodiment, calculating the first sampling angle of the conductor at the nth sampling time based on the angular velocity and acceleration of the conductor at the nth sampling time includes:

[0013] Based on the angular velocity of the conductor at the nth sampling time, the angle of the conductor is calculated to obtain the first angle value;

[0014] Based on the acceleration of the conductor at the nth sampling time, the angle of the conductor is calculated to obtain the second angle value;

[0015] The first angle value and the second angle value are subjected to complementary filtering processing based on a preset complementary filtering algorithm to obtain the first sampling angle of the conductor at the nth sampling time.

[0016] In one embodiment, obtaining the angular velocity and acceleration of the conductor at each sampling time includes:

[0017] A three-axis coordinate system is constructed with the conductor as the center; the three-axis coordinate system includes an x-axis, a y-axis, and a z-axis.

[0018] The angular velocities of the conductor along the x-axis, y-axis, and z-axis at each sampling time are obtained.

[0019] Obtain the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at each sampling time;

[0020] The first angle value includes the x-axis angle value, the y-axis angle value, and the z-axis angle value; the second angle value includes the pitch angle value and the roll angle value; the first sampling angle includes the x-axis first sampling angle, the y-axis first sampling angle, and the z-axis first sampling angle.

[0021] In one embodiment, calculating the angle of the conductor based on the angular velocity of the conductor at the nth sampling time to obtain a first angle value includes:

[0022] Calculate the integrals of the x-axis angular velocity, y-axis angular velocity, and z-axis angular velocity of the conductor at the nth sampling time, respectively, to obtain the x-axis angle, y-axis angle, and z-axis angle of the conductor at the nth sampling time;

[0023] The x-axis angle, y-axis angle, and z-axis angle are filtered first to obtain the x-axis angle value, y-axis angle value, and z-axis angle value of the conductor at the nth sampling time.

[0024] In one embodiment, calculating the angle of the conductor based on the acceleration of the conductor at the nth sampling time to obtain a second angle value includes:

[0025] Based on the acceleration of the conductor at the nth sampling time, determine the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at the nth sampling time;

[0026] Based on the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at the nth sampling time, determine the pitch angle and roll angle of the conductor at the nth sampling time;

[0027] The pitch angle and roll angle are filtered a second time to obtain the pitch angle value and roll angle value respectively.

[0028] In one embodiment, calculating the predicted angle of the conductor at the (n+1)th sampling time based on the first sampling angle of the conductor at the nth sampling time includes:

[0029] Based on the first sampling angle of the conductor at the nth sampling time, the estimated angle of the conductor at the (n+1)th sampling time is determined;

[0030] Based on the estimated angle of the conductor at the (n+1)th sampling time, determine the angle estimation error of the conductor at the (n+1)th sampling time;

[0031] Based on the estimated angle and angle estimation error of the conductor at the (n+1)th sampling time, the predicted angle of the conductor at the (n+1)th sampling time is determined.

[0032] In one embodiment, calculating the galloping angle of the conductor at the (n+1)th sampling time based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time includes:

[0033] The Kalman gain is calculated based on the angle prediction error of the conductor at the (n+1)th sampling time.

[0034] Based on the second sampling angle of the conductor at the (n+1)th sampling time and the Kalman gain, update the second sampling angle and angle prediction error of the conductor at the (n+1)th sampling time;

[0035] Based on the updated second sampling angle and angle prediction error of the conductor at the (n+1)th sampling time, the galloping angle of the conductor at the (n+1)th sampling time is calculated.

[0036] In one embodiment, the conductor is provided with multiple monitoring points; the conductor galloping monitoring method further includes:

[0037] Obtain the angular velocity and acceleration of m monitoring points at the nth sampling time; where m is an integer greater than or equal to 2;

[0038] Based on the angular velocity and acceleration of the m monitoring points at the nth sampling time, a forward state sequence and a reverse state sequence are generated;

[0039] Based on the forward state sequence and the reverse state sequence, the predicted dancing posture of the conductor at the nth sampling time is determined;

[0040] If the predicted galloping posture of the conductor at the nth sampling time is greater than a preset value, the galloping of the conductor at the nth sampling time is determined to be abnormal.

[0041] In one embodiment, the conductor galloping monitoring method further includes:

[0042] Based on the acceleration of the conductor at the (n+1)th sampling time, the amplitude and frequency of the conductor's galloping at the (n+1)th sampling time are determined.

[0043] Based on the dancing angle, dancing amplitude, and dancing frequency of the conductor at the (n+1)th sampling time, the dancing trajectory of the conductor is determined; the dancing trajectory is used to determine the dancing state of the conductor.

[0044] This application also proposes a computer storage medium, including a processor and a memory; the memory stores computer instructions that, when executed on the processor, cause the processor to perform the above-described wire galloping monitoring method.

[0045] This application also proposes a conductor galloping monitoring system, comprising:

[0046] The parameter acquisition module is used to detect the angular velocity and acceleration of the conductor;

[0047] The main control module is used to calculate the first sampling angle of the conductor at the nth sampling time based on the angular velocity and acceleration of the conductor at the nth sampling time; and to calculate the predicted angle of the conductor at the (n+1)th sampling time based on the first sampling angle of the conductor at the nth sampling time; the main control module is also used to calculate the second sampling angle of the conductor at the (n+1)th sampling time based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time; and to calculate the galloping angle of the conductor at the (n+1)th sampling time based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time; the main control module is also used to output a warning message when it determines that the galloping state of the conductor is abnormal based on the galloping angle of the conductor at the (n+1)th sampling time; where n is a positive integer.

[0048] In one embodiment, the parameter acquisition module includes:

[0049] The Beidou positioning module is used to obtain the position information of the conductor; the position information and the wagging angle are used to determine the wagging state of the conductor.

[0050] In one embodiment, the conductor galloping monitoring system further includes a photovoltaic module for converting solar energy into electrical energy to power the parameter acquisition module and the main control module.

[0051] This application calculates the galloping angle of the conductor by using angular velocity and acceleration data, achieving the fusion and complementarity of multi-source data, thereby improving the accuracy and comprehensiveness of the galloping angle. The comprehensive analysis of multi-source data can more accurately reflect the actual motion state of the conductor and external environmental conditions, providing more reliable data support for galloping monitoring. Updating the predicted galloping angle data with angle data obtained from actual sampling further reduces the error in the galloping angle. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the module structure of an embodiment of the conductor galloping monitoring system of this application.

[0053] Figure 2 This is a schematic diagram of the structure of an embodiment of the photovoltaic module of this application.

[0054] Figure 3 This is an application flow logic diagram of an embodiment of the conductor galloping monitoring system of this application.

[0055] Figure 4 This is a flowchart of an embodiment of the conductor galloping monitoring method of this application.

[0056] Figure 5 This is a flowchart illustrating an embodiment of obtaining angular velocity and acceleration for this application.

[0057] Figure 6 This is a schematic diagram of a three-axis coordinate system according to an embodiment of this application.

[0058] Figure 7 This is a flowchart illustrating an embodiment of calculating the first sampling angle for this application.

[0059] Figure 8 A flowchart illustrating an embodiment of calculating the first angle value for this application.

[0060] Figure 9 This is a flowchart of an embodiment of complementary filtering as defined in this application.

[0061] Figure 10 A flowchart illustrating an embodiment of calculating the second angle value for this application.

[0062] Figure 11 This is a flowchart illustrating one embodiment of the calculation of the predicted angle in this application.

[0063] Figure 12 This is a flowchart illustrating an embodiment of calculating the dancing angle for this application.

[0064] Figure 13 This is a flowchart of another embodiment of the conductor galloping monitoring method of this application.

[0065] Figure 14 This is a logic flowchart of another embodiment of the conductor galloping monitoring method of this application.

[0066] Figure 15 This is a flowchart of yet another embodiment of the conductor galloping monitoring method of this application.

[0067] Explanation of key component symbols: Conductor galloping monitoring system-100; Parameter acquisition module-110; Main control module-120; Beidou positioning module-111; Photovoltaic module-130; First control unit-121; Second control unit-122.

[0068] The following detailed description, in conjunction with the accompanying drawings, will further illustrate this application. Detailed Implementation

[0069] The following description will refer to the accompanying drawings to provide a more complete picture of the present application. The drawings illustrate exemplary embodiments of the present application. However, the present application may be implemented in many different forms and should not be construed as limited to the exemplary embodiments set forth herein. These exemplary embodiments are provided to make the present application thorough and complete, and to fully convey the scope of the present application to those skilled in the art. Similar reference numerals denote the same or similar components.

[0070] The terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to limit the application. As used herein, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” are intended to also include the plural forms. Furthermore, when used herein, “comprising” and / or “including” and / or “having,” integers, steps, operations, components, and / or components, but does not exclude the presence or addition of one or more other features, regions, integers, steps, operations, components, and / or groups thereof.

[0071] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. Furthermore, unless expressly defined herein, terms such as those defined in a general dictionary should be interpreted as having the same meaning as they have in the relevant art and in the content of this application, and will not be interpreted as having an idealized or overly formal meaning.

[0072] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments. It should be noted that components depicted in the drawings are not necessarily shown to scale; and identical or similar components will be designated with the same or similar reference numerals or similar technical terms.

[0073] Reference Figure 1 This application proposes a conductor galloping monitoring system 100, including a parameter acquisition module 110 and a main control module 120. The parameter acquisition module 110 is used to detect the angular velocity and acceleration of the conductor. The main control module 120 is used to calculate a first sampling angle of the conductor at the nth sampling time based on the angular velocity and acceleration of the conductor at the nth sampling time; and to calculate a predicted angle of the conductor at the (n+1)th sampling time based on the first sampling angle of the conductor at the nth sampling time; the main control module 120 is also used to calculate a second sampling angle of the conductor at the (n+1)th sampling time based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time; and to calculate the galloping angle of the conductor at the (n+1)th sampling time based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time; where n is a positive integer.

[0074] In this embodiment, the parameter acquisition module 110 may include a gyroscope and an accelerometer. By installing gyroscopes and accelerometers at various monitoring points on the conductor, the angular velocity and acceleration at different positions of the conductor can be detected. The main control module 120 can calculate the wagging angle of the conductor at each monitoring point based on the angular velocity and acceleration at each monitoring point.

[0075] A gyroscope can detect the angular velocity of a conductor. Integrating this angular velocity yields the conductor's angles (roll, pitch, and yaw), thus providing its attitude. However, the angular velocity detected by the gyroscope is subject to measurement errors, noise, and drift, which accumulate after integration. An accelerometer can acquire the conductor's acceleration. Calculating this acceleration provides the conductor's real-time attitude angles. Therefore, this embodiment uses the accelerometer's acceleration to obtain the conductor's real-time attitude angles, correcting the accumulated errors from the gyroscope-based angles and improving the accuracy of the conductor's galloping angles.

[0076] The main control module 120 can also calculate the predicted angle for the next sampling time based on the first sampling angle at the nth sampling time, and calculate the second sampling angle at the (n+1)th sampling time based on the angular velocity and acceleration detected at the (n+1)th sampling time. In this way, the predicted angle is updated based on the second sampling angle at the (n+1)th sampling time, making the difference between the final calculated dancing angle and the actual dancing angle smaller.

[0077] In one embodiment, the main control module 120 is further configured to output a warning message when it determines that the dancing state of the conductor is abnormal based on the dancing angle of the conductor at the (n+1)th sampling time.

[0078] For example, if the gyratory angle at the (n+1)th sampling time is greater than or equal to a preset angle, the main control module 120 can output a warning message to the tower to prompt monitoring personnel to perform timely maintenance. If the gyratory angle at the (n+1)th sampling time is less than the preset angle, monitoring continues.

[0079] In one embodiment, the main control module 120 can be mounted on the conductor. Alternatively, the main control module 120 can include a first control unit 121 mounted on the conductor and a second control unit 122 mounted on the tower. The first control unit 121 can collect angular velocity and acceleration data on the conductor, perform preliminary filtering on the angular velocity and acceleration data, and then send the data to the second control unit 122. The second control unit 122 analyzes the angular velocity and acceleration data to obtain data such as the waving angle, waving amplitude, and waving frequency, and displays and stores the data in the background. If the analyzed data is abnormal, for example, if the waving angle is greater than a preset angle, the second control unit 122 can also output a warning signal to prompt monitoring personnel to perform timely maintenance. Furthermore, the second control unit 122 can also transmit data such as the waving angle, waving amplitude, and waving frequency to the backend main station via a dedicated network.

[0080] In one embodiment, the parameter acquisition module 110 may further include a temperature sensor. One or more temperature sensors can be set at each monitoring point on the conductor to detect the conductor temperature and / or ambient temperature. The temperature sensor can be a platinum resistance thermometer. Platinum resistance thermometers can cover a wide temperature measurement range from -200°C to 850°C and have excellent anti-interference capabilities and adaptability, making them suitable for various environmental conditions. The linearity of platinum resistance thermometers is superior to other types of temperature sensors, and systematic errors can be reduced and measurement accuracy improved through lookup tables and software compensation methods. Furthermore, the three-wire or four-wire connection method used in platinum resistance thermometers effectively reduces the influence of conductor resistance on the measurement results, further enhancing their accuracy.

[0081] In one embodiment, the parameter acquisition module 110 may further include a humidity sensor for collecting the humidity of the wire and / or the ambient humidity.

[0082] In one embodiment, the conductor galloping monitoring system 100 further includes a BeiDou positioning module 111, which is used to acquire the position information of the conductor; the position information and the galloping angle are used to determine the galloping information of the conductor.

[0083] In one embodiment, the conductor galloping monitoring system 100 further includes a data transmission module, which transmits the position information and the galloping angle to the tower and / or the cloud. The data transmission module can be implemented using a LoRa antenna module, a 5G antenna module, or the like.

[0084] Leveraging the high positioning accuracy of the BeiDou positioning module 111, the main control module 120 can acquire the conductor's position information in real time and transmit it along with the galloping angle to the tower and / or cloud for analysis, ensuring the accuracy and real-time nature of data transmission. The tower-based computing center or cloud can calculate the conductor's galloping information, such as galloping posture and trajectory, based on the position information and the galloping angle, providing maintenance personnel with more accurate and reliable decision support.

[0085] In one embodiment, the conductor galloping monitoring system 100 further includes a photovoltaic module 130 for converting solar energy into electrical energy to power the parameter acquisition module 110 and the main control module 120. The photovoltaic module 130 can be implemented using one or more solar photovoltaic panels.

[0086] Reference Figure 2In one embodiment, the photovoltaic module 130 includes a solar photovoltaic panel, a solar management and charging management chip, a lithium-ion polymer battery, and a linear regulator. The solar photovoltaic panel converts solar energy into electrical energy to charge the lithium-ion polymer battery. The electrical energy output from the lithium-ion polymer battery is converted into a suitable operating voltage by the linear regulator and then output to the main control module 120 to power it.

[0087] Figure 3 This is a logic flowchart of an embodiment of the conductor galloping monitoring system 100 of this application. After the conductor galloping monitoring system 100 is started, the parameter acquisition module 110 collects data such as angular velocity, acceleration, temperature, and positioning. The first control unit 121 on the conductor filters, formats, and compresses the collected data, and then transmits it to the second control unit 122 on the tower via a LORA antenna for analysis, obtaining data such as galloping angle, galloping amplitude, and galloping frequency. The data is then transmitted to the backend main station via a dedicated network for display and storage in the background. If the analyzed data is normal, monitoring continues. If the analyzed data is abnormal, for example, if the galloping angle is greater than a preset angle, the second control unit 122 can also output a warning signal to prompt monitoring personnel to perform timely maintenance.

[0088] This application calculates the galloping angle of the conductor by using angular velocity and acceleration data, achieving the fusion and complementarity of multi-source data, thereby improving the accuracy and comprehensiveness of the galloping angle. The comprehensive analysis of multi-source data can more accurately reflect the actual motion state of the conductor and external environmental conditions, providing more reliable data support for galloping monitoring. Updating the predicted galloping angle data with angle data obtained from actual sampling further reduces the error in the galloping angle.

[0089] Reference Figure 4 This application also proposes a method for monitoring conductor galloping, comprising:

[0090] S1: Obtain the angular velocity and acceleration of the conductor at each sampling time.

[0091] In this embodiment, angular velocity can be detected by a gyroscope, and acceleration can be detected by an accelerometer. Multiple monitoring points can be set at intervals on the conductor according to actual needs. The angular velocity and acceleration of each monitoring point at each sampling time can be acquired and calculated to obtain the conductor's swaying angle at different monitoring points. This embodiment uses only one monitoring point as an example for illustration.

[0092] Reference Figure 5 In one embodiment, step S1 includes:

[0093] S11: Construct a three-axis coordinate system with the conductor as the center; the three-axis coordinate system includes the x-axis, y-axis and z-axis.

[0094] By constructing a three-axis coordinate system, the orientation of the conductor can be described intuitively. Figure 6 The diagram illustrates the coordinate system transformation during attitude calculation. The x, y, and z axes constitute the conductor's three-axis coordinate system. The a, b, and c axes constitute the Earth coordinate system. The angles of rotation from the Earth coordinate system to the conductor coordinate system are Euler angles, which can be expressed as (pitch angle θ, roll angle ψ, yaw angle...). () is used to visually describe the attitude of the conductor.

[0095] The rotation matrix based on Euler angles requires solving for nine elements, and gimbaling occurs when the pitch angle approaches ±90°. Therefore, Euler angles are severely limited in their use for attitude representation, hindering full-attitude measurement by the detection system. When calculating the angular attitude of a traverse line, the traverse coordinate system data is typically converted to the Earth coordinate system. The angular attitude of the traverse line can be represented using parameters such as quaternions.

[0096] Through this rotation matrix The transformation between the traverse coordinate system and the Earth coordinate system can be achieved during attitude calculation:

[0097]

[0098] Quaternions can be represented as:

[0099] q(q0,q1,q2,q3)=q0+q1i+q2j+q3k,

[0100] Where i, j, k are unit vectors, and the normalized quaternions satisfy:

[0101] q0 2 +q1 2 +q2 2 +q3 2 =1.

[0102] The conversion relationship between the initial quaternion and the initial Euler angles is as follows:

[0103]

[0104] After attitude calculation, the rotation axis and rotation direction information generated by the coordinate system transformation relationship can be used to obtain the three-axis attitude Euler angles in reverse using quaternions.

[0105] θ = -arcsin[2(q1q3-q0q2)],

[0106]

[0107] S12: Obtain the angular velocity of the conductor along the x-axis, y-axis, and z-axis at each sampling time.

[0108] The angular velocities of the x-axis, y-axis, and z-axis can be determined by gyroscope detection, so that the dancing angle can be calculated from the angular velocities of the three axes.

[0109] S13: Obtain the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at each sampling time.

[0110] An accelerometer can detect the magnitude and direction of gravitational acceleration. When an object is moving, gravitational acceleration is always downward. The three components of gravity projected onto the three-axis coordinate system of the conductor are the acceleration along the x-axis, y-axis, and z-axis, respectively.

[0111] S2: Based on the angular velocity and acceleration of the conductor at the nth sampling time, calculate the first sampling angle of the conductor at the nth sampling time; where n is a positive integer.

[0112] In this embodiment, the gyroscope can detect the angular velocity of the conductor. Calculating the angular velocity yields the conductor's wagging angle, which in turn determines its attitude. However, the angular velocity detected by the gyroscope may contain measurement errors, noise, and drift, resulting in accumulated errors after integration. The conductor's wagging angle can also be obtained by acquiring and calculating its acceleration. Therefore, by calculating the acceleration and angular velocity, a first sampling angle can be obtained. Employing multi-source data fusion and complementarity can yield more accurate angle data.

[0113] Reference Figure 7 In one embodiment, step S2 includes:

[0114] S21: Based on the angular velocity of the conductor at the nth sampling time, calculate the angle of the conductor to obtain the first angle value.

[0115] In this embodiment, the angular velocity is integrated and filtered to obtain the required first angle value, which is then used for subsequent fusion calculations.

[0116] Reference Figure 8 and Figure 9 In one embodiment, step S21 includes:

[0117] S211: Calculate the integrals of the x-axis angular velocity, y-axis angular velocity, and z-axis angular velocity of the conductor at the nth sampling time, respectively, to obtain the x-axis angle, y-axis angle, and z-axis angle of the conductor at the nth sampling time.

[0118] S212: Perform a first filter on the x-axis angle value, the y-axis angle value, and the z-axis angle value respectively to obtain the x-axis angle value, y-axis angle value, and z-axis angle value of the conductor at the nth sampling time. The range of the first filter can be selected according to actual needs. For example, in the gyroscope's detection signal, there is a lot of low-frequency noise; the first filter can be set to a high-pass filter to remove low-frequency noise interference and retain the required high-frequency information.

[0119] S22: Based on the acceleration of the conductor at the nth sampling time, calculate the angle of the conductor to obtain the second angle value.

[0120] In this embodiment, the angle of the conductor is obtained by calculating the acceleration and then filtered to obtain the required second angle value, which is then used for subsequent fusion calculations.

[0121] Reference Figure 10 and Figure 9 In one embodiment, step S22 includes:

[0122] S221: Calculate the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at the nth sampling time to obtain the pitch angle and roll angle of the conductor at the nth sampling time.

[0123] In this embodiment, the pitch angle and roll angle values ​​can be calculated using the following formulas:

[0124]

[0125] Where pitch is the pitch angle value, roll is the roll angle value, and a x Let a be the acceleration along the x-axis. y Let a be the y-axis acceleration. z This is the acceleration along the z-axis.

[0126] S222: Perform a second filter on the pitch angle and roll angle respectively to obtain the pitch angle value and roll angle value. The range of the second filter can be selected according to actual needs. For example, accelerometers are easily affected by high-frequency noise; the second filter can be set to a low-pass filter to remove high-frequency noise interference and retain the required low-frequency information.

[0127] S23: Based on a preset complementary filtering algorithm, the first angle value and the second angle value are subjected to complementary filtering processing to obtain the first sampling angle of the conductor at the nth sampling time.

[0128] For example, in order to make the reconstructed angle signal after complementary filtering as close as possible to the actual sampled angle signal, the high-pass filter and low-pass filter can be set to meet the following conditions:

[0129] G1(s)+G2(s)=1,

[0130] Where G1(s) is the transfer function of the high-pass filter and G2(s) is the transfer function of the low-pass filter. Its general form is:

[0131]

[0132] Where C(s) is the transfer function of the PI controller, and s is a variable in the frequency domain. Error compensation using a PI controller can reduce signal noise and drift error.

[0133]

[0134] Where, k P k is the proportionality coefficient. I The integral coefficient is used to adjust the magnitude of the proportional coefficient and the integral coefficient, which can change the effect of PI control.

[0135] Therefore, the transfer functions of the high-pass filter and the low-pass filter are respectively:

[0136]

[0137] Thus, by performing high- and low-frequency complementary filtering on the first and second angle values ​​through the above calculations, high- and low-frequency noise during the data acquisition process can be effectively filtered out, thereby obtaining a first sampling angle that is close to the original attitude angle. In addition, the preset complementary filtering algorithm can be set to other conditions as needed, which are not limited here.

[0138] S3: Based on the first sampling angle of the conductor at the nth sampling time, calculate the predicted angle of the conductor at the (n+1)th sampling time.

[0139] In this embodiment, after calculating the first sampling angle of the conductor at the nth sampling time through the above steps, the angle of the conductor at the next sampling time can be predicted based on the first sampling angle, so as to obtain a more accurate sampling angle at the next sampling time. For example, the first sampling angle of the conductor at the nth sampling time can be input into the prediction model to predict the angle of the conductor at the next sampling time.

[0140] Reference Figure 11 and Figure 9 In one embodiment, step S3 includes:

[0141] S31: Based on the first sampling angle of the conductor at the nth sampling time, determine the estimated angle of the conductor at the (n+1)th sampling time.

[0142] In this embodiment, an extended Kalman filter can be used to predict the conductor angle at the next moment.

[0143] First, define the state vector x, (x∈R) n The state equation is:

[0144] x k =f(x) k-1 ,k-1)+m k-1 ,

[0145] Where k represents the current time. The nonlinear function f is the transfer function, describing the evolution of the system state over time. n ∈R n To obey N(0,Q) k The state equation for Gaussian white noise with a linear distribution is:

[0146] x k =f(x) k-1 ,u k-1 ,m k-1 ).

[0147] In this way, the angle of the conductor at the (n+1)th sampling time can be predicted.

[0148] S32: Based on the estimated angle of the conductor at the (n+1)th sampling time, determine the angle estimation error of the conductor at the (n+1)th sampling time.

[0149] In this embodiment, the angle error of the conductor at the (n+1)th sampling time can be predicted using the error covariance prediction equation. Error covariance matrix prediction predicts the error covariance at the next time step. This step considers process noise, which describes the uncertainty in system state prediction. The error covariance prediction equation is shown below:

[0150]

[0151] Among them, F k Let be the Jacobian matrix of the state transition function. The Jacobian matrix is ​​used to approximate nonlinear functions. State transition functions and observation functions are typically nonlinear, derived from the current state x. k The Jacobian matrix of these functions is calculated at each time step to facilitate the linearization of these nonlinear functions in subsequent calculations, thereby approximating the nonlinear problem at each time step.

[0152] This is the process noise covariance matrix, used in Kalman filters to describe the uncertainty of the system model, i.e., the random noise during the system state transition process. This matrix is ​​an important parameter for measuring the magnitude of internal system noise; it quantifies the dynamic behavior of the system and reflects the uncertainty of state prediction.

[0153] Taking the partial derivative of the function f, we obtain the Jacobian matrix of the state vector:

[0154]

[0155] S33: Based on the estimated angle and angle estimation error of the conductor at the (n+1)th sampling time, determine the predicted angle of the conductor at the (n+1)th sampling time.

[0156] By combining the above state equation and error covariance prediction equation, and substituting the angular velocity and acceleration values, the three-axis attitude angles at the current moment can be calculated, and used as the predicted angles for the next moment.

[0157] S4: Based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time, calculate the second sampling angle of the conductor at the (n+1)th sampling time.

[0158] In this embodiment, the principle of calculating the second sampling angle of the conductor at the (n+1)th sampling time is the same as that of calculating the first sampling angle of the conductor at the nth sampling time in step S2, and will not be repeated here.

[0159] S5: Based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time, calculate the dancing angle of the conductor at the (n+1)th sampling time.

[0160] In this embodiment, at the (n+1)th sampling time, the galloping angle of the conductor at the (n+1)th sampling time is obtained by calculating the predicted angle from the previous time and the second sampling angle obtained by actual measurement. The predicted angle is updated using the second sampling angle obtained by actual measurement, so that the final galloping angle is closer to the actual state of the conductor.

[0161] Reference Figure 12 In one embodiment, step S5 includes:

[0162] S51: Calculate the Kalman gain based on the angle prediction error of the conductor at the (n+1)th sampling time.

[0163]

[0164] in, H is the Jacobian matrix of the observation model with respect to the state. It is used to linearize nonlinear observation models and describes the local linear rate of change of the observation model with respect to the state variables. k and V k R is the measurement Jacobian matrix at time k. k Let k represent the measurement noise covariance matrix at time k. This represents the error covariance matrix at the previous time step.

[0165] In this embodiment, the state equation and error covariance equation can be updated using Kalman gain. Kalman gain is a coefficient that balances prediction error and measurement error, and it determines the proportion of the measured value in the state update.

[0166] S52: Based on the second sampling angle of the conductor at the (n+1)th sampling time and the Kalman gain, update the second sampling angle and angle prediction error of the conductor at the (n+1)th sampling time.

[0167] In this embodiment, Kalman gain and a second sampling angle are used to update the predicted angle. The state update equation combines the predicted state and the measurement residual (i.e., the difference between the measured value and the predicted state) to produce a more accurate state estimate:

[0168]

[0169] Define the observation vector z, (z∈R) m )

[0170] The observation equation is:

[0171] z k =h(x k )+v k ,

[0172] v n ∈R n To conform to N(0,R) k Gaussian white noise with a distribution of ) is linearized to obtain:

[0173] z k =h(x k ,v k ),

[0174] The nonlinear function h is the transfer function, which describes how to obtain observation data from the system's state variables.

[0175] S53: Based on the updated second sampling angle and angle prediction error of the conductor at the (n+1)th sampling time, calculate the dancing angle of the conductor at the (n+1)th sampling time.

[0176] Finally, the error covariance matrix is ​​updated. This step takes into account measurement noise, which describes the uncertainty of the measurement values:

[0177]

[0178] Thus, after fusion complementary filtering and extended Kalman filtering, and then converting the processed quaternions into Euler angle outputs, the galloping angles (pitch angle θ, roll angle ψ, yaw angle) of the conductor at the (n+1)th sampling time can be obtained. ).

[0179] S6: Based on the galloping angle of the conductor at the (n+1)th sampling time, determine that the galloping state of the conductor is abnormal and output a warning message. For example, if the galloping angle of the conductor at the (n+1)th sampling time is greater than a preset angle, output a warning message to remind monitoring personnel to perform timely maintenance.

[0180] Reference Figure 13 and Figure 14 In one embodiment, the conductor galloping monitoring method further includes:

[0181] S7: Obtain the angular velocity and acceleration of the m monitoring points at the nth sampling time; where m is an integer greater than or equal to 2.

[0182] In this embodiment, multiple monitoring points are typically installed evenly on the conductor. After acquiring the sampling data from multiple monitoring points, the BiLSTM (Bidirectional Long Short-Term Memory) algorithm can be used to process the sampled data. BiLSTM is a special type of recurrent neural network (RNN) widely used for processing sequential data. BiLSTM can be represented as a combination of two LSTMs. BiLSTM can consider both forward and reverse information. In overhead transmission lines, multiple monitoring points are generally installed evenly on the same conductor or ground wire on the windward side. Using BiLSTM can consider the influence between monitoring points bidirectionally, more accurately representing the cable status in the transmission line.

[0183] S8: Based on the angular velocity and acceleration of the m monitoring points at the nth sampling time, generate a forward state sequence and a reverse state sequence.

[0184] First, given an input sequence E = [e1, e2, ..., e T The internal mechanism of LSTM can be divided into forget gate, input gate, candidate memory units, update memory units, output gate, and hidden state. The function of the forget gate is mainly to determine how much information should be forgotten from the memory units of the previous sequence. Specifically, the formula for the forget gate is as follows:

[0185] f t =σ(W f ·[h t-1 ,e t ]+b f ),

[0186] Among them, W f This is the weight matrix of the forget gate, σ() is the sigmoid activation function, and bf It's a bias.

[0187] [h t-1 ,e t ] is the concatenated vector of the previous hidden state and the current input.

[0188] The function of the input gate is to determine how much information from the current sequence can be stored in the memory unit. Its formula is as follows:

[0189] i t =σ(W i ·[h t-1 ,e t ]+b i ),

[0190] Input gate i t Each element in the equation is processed by the sigmoid function to obtain an output between 0 and 1. The closer the value is to 1, the more the information at that position needs to be updated.

[0191] Candidate states are obtained by linearly combining the current sequence input with the hidden state from the previous time step and then activating it using the tanh function.

[0192]

[0193] Updating memory units is a core step, determining which information will be retained, as shown in the following formula:

[0194]

[0195] Where * represents element-wise multiplication, the output gate mainly controls how much information from memory units is transferred to the hidden state in each sequence, and its formula is as follows:

[0196] o t =σ(W o ·[h t-1 ,e t ]+b o ).

[0197] The formula for the hidden state is as follows:

[0198] h t =o t *tanh(c t ).

[0199] The first step in the feature vectorization process is to convert the features of each sampled data at the nth sampling time into a One-Hot vector form, where the corresponding position of the One-Hot vector is assigned the specific value represented by that feature. Subsequently, these One-Hot vectors are input into a hidden layer of length j for deep encoding, which follows the formula:

[0200] e i =δ(WX) i +b i ),

[0201] Where X i =[x1,x2,…,x j [W] represents the one-hot encoded vector corresponding to the i-th sampling time, j is the number of features, and W∈R m×n Let b be a trainable matrix, b be the bias, and δ() be a non-linear activation function. In this embodiment, the softmax activation function can be used.

[0202] A forward LSTM generates a forward hidden state sequence. The inverse LSTM produces a reversed sequence of hidden states. BiLSTM can combine forward and backward information to obtain two hidden states. Forward hidden state sequence and the reverse hidden state sequence The data is then stitched together and used as input for the next section. This mechanism enables BiLSTM to demonstrate higher efficiency and accuracy when handling tasks such as overhead transmission line galloping, which require comprehensive consideration of front and rear position information.

[0203] S9: Based on the forward state sequence and the reverse state sequence, output the predicted dancing posture of the conductor at the nth sampling time.

[0204] After obtaining the BiLSTM hidden state output containing sequence information, it is input into the prediction layer. A fully connected layer transforms the embedding containing sequence information into the predicted value, as shown in the following formula:

[0205]

[0206] in, To predict probabilities, the range is between 0 and 1, σ() is the sigmoid function, and W r ∈

[0207] R 1×2m h i This represents the stitched embedding calculated by BiLSTM at the i-th sampling time. The predicted probability value for the conductor can be obtained using the above formula, and the conductor's state (normal or abnormal) can be determined based on this predicted probability value.

[0208] The model uses the cross-entropy loss function for optimization, and the formula is as follows:

[0209]

[0210] r iLet r be the true label at the i-th sampling time, where r n =1, r j =0, which is 0 if the conductor is in normal condition at time j, and 1 if the conductor exhibits abnormal galloping at time i. The output value represents the model's prediction for that time step. The model uses the Adam algorithm for gradient descent with a learning rate of 0.001. Dropout is set to 0.3 to reduce the risk of overfitting. L2 normalization is performed on each embedding layer.

[0211] S10: If the predicted dancing posture of the conductor at the nth sampling time is greater than the preset value, it is determined that the dancing of the conductor at the nth sampling time is abnormal.

[0212] In this embodiment, a preset value can be set according to the actual application. If the predicted probability value is greater than or equal to the preset value, then the conductor galloping is determined to be abnormal. For example, the preset value is set to 0.6. If the value is greater than or equal to 0.6, then the conductor galloping is considered abnormal.

[0213] Reference Figure 15 In one embodiment, the conductor galloping monitoring method further includes:

[0214] S20: Based on the acceleration of the conductor at the (n+1)th sampling time, determine the amplitude and frequency of the conductor's galloping at the (n+1)th sampling time.

[0215] In this embodiment, the acceleration of the conductor at the (n+1)th sampling time can be double-integrated to obtain the conductor's displacement. The displacement of the conductor at multiple sampling times within a preset time is analyzed to extract the conductor's galloping amplitude and zero-crossing count. Based on the zero-crossing count and the preset time, the conductor's galloping frequency can be obtained.

[0216] S30: Determine the dancing trajectory of the conductor based on the dancing angle, dancing amplitude, and dancing frequency of the conductor at the (n+1)th sampling time.

[0217] In this embodiment, by collecting the angular velocity and acceleration of multiple monitoring points at the (n+1)th sampling time and performing the above calculations, the galloping angle, galloping amplitude, and galloping frequency of the multiple monitoring points at the (n+1)th sampling time can be obtained, thereby generating the galloping trajectory of the conductor at the (n+1)th sampling time. Thus, based on the galloping trajectories of the conductor at multiple sampling times, a continuous galloping trajectory can be generated, allowing monitoring personnel to more intuitively obtain the state of the conductor.

[0218] This application calculates the galloping angle of the conductor by using angular velocity and acceleration data, achieving the fusion and complementarity of multi-source data, thereby improving the accuracy and comprehensiveness of the galloping angle. The comprehensive analysis of multi-source data can more accurately reflect the actual motion state of the conductor and external environmental conditions, providing more reliable data support for galloping monitoring. Updating the predicted galloping angle data with angle data obtained from actual sampling further reduces the error in the galloping angle.

[0219] The specific embodiments of this application have been described above with reference to the accompanying drawings. However, those skilled in the art will understand that various changes and substitutions can be made to the specific embodiments of this application without departing from the spirit and scope of this application. All such changes and substitutions fall within the scope defined by this application.

Claims

1. A method for monitoring conductor galloping, characterized in that, include: Obtain the angular velocity and acceleration of the conductor at each sampling time; Based on the angular velocity and acceleration of the conductor at the nth sampling time, calculate the first sampling angle of the conductor at the nth sampling time; where n is a positive integer; Based on the first sampling angle of the conductor at the nth sampling time, calculate the predicted angle of the conductor at the (n+1)th sampling time; Based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time, calculate the second sampling angle of the conductor at the (n+1)th sampling time; Based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time, the galloping angle of the conductor at the (n+1)th sampling time is calculated. If the dancing angle of the conductor at the (n+1)th sampling time is used to determine that the dancing state of the conductor is abnormal, a warning message is output.

2. The conductor galloping monitoring method as described in claim 1, characterized in that, The calculation of the first sampling angle of the conductor at the nth sampling time based on the angular velocity and acceleration of the conductor at the nth sampling time includes: Based on the angular velocity of the conductor at the nth sampling time, the angle of the conductor is calculated to obtain the first angle value; Based on the acceleration of the conductor at the nth sampling time, the angle of the conductor is calculated to obtain the second angle value; The first angle value and the second angle value are subjected to complementary filtering processing based on a preset complementary filtering algorithm to obtain the first sampling angle of the conductor at the nth sampling time.

3. The conductor galloping monitoring method as described in claim 2, characterized in that, The acquisition of the angular velocity and acceleration of the conductor at each sampling time includes: A three-axis coordinate system is constructed with the conductor as the center; the three-axis coordinate system includes an x-axis, a y-axis, and a z-axis. The angular velocities of the conductor along the x-axis, y-axis, and z-axis at each sampling time are obtained. Obtain the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at each sampling time; The first angle value includes the x-axis angle value, the y-axis angle value, and the z-axis angle value; the second angle value includes the pitch angle value and the roll angle value; the first sampling angle includes the x-axis first sampling angle, the y-axis first sampling angle, and the z-axis first sampling angle.

4. The conductor galloping monitoring method as described in claim 3, characterized in that, The step of calculating the angle of the conductor based on the angular velocity of the conductor at the nth sampling time to obtain a first angle value includes: Calculate the integrals of the x-axis angular velocity, y-axis angular velocity, and z-axis angular velocity of the conductor at the nth sampling time, respectively, to obtain the x-axis angle, y-axis angle, and z-axis angle of the conductor at the nth sampling time; The x-axis angle, y-axis angle, and z-axis angle are filtered first to obtain the x-axis angle value, y-axis angle value, and z-axis angle value of the conductor at the nth sampling time.

5. The conductor galloping monitoring method as described in claim 3, characterized in that, The calculation of the angle of the conductor based on the acceleration of the conductor at the nth sampling time to obtain the second angle value includes: Based on the acceleration of the conductor at the nth sampling time, determine the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at the nth sampling time; Based on the x-axis acceleration, y-axis acceleration, and z-axis acceleration of the conductor at the nth sampling time, determine the pitch angle and roll angle of the conductor at the nth sampling time; The pitch angle and roll angle are filtered a second time to obtain the pitch angle value and roll angle value respectively.

6. The conductor galloping monitoring method as described in claim 1, characterized in that, The step of calculating the predicted angle of the conductor at the (n+1)th sampling time based on the first sampling angle of the conductor at the nth sampling time includes: Based on the first sampling angle of the conductor at the nth sampling time, the estimated angle of the conductor at the (n+1)th sampling time is determined; Based on the estimated angle of the conductor at the (n+1)th sampling time, determine the angle estimation error of the conductor at the (n+1)th sampling time; Based on the estimated angle and angle estimation error of the conductor at the (n+1)th sampling time, the predicted angle of the conductor at the (n+1)th sampling time is determined.

7. The conductor galloping monitoring method as described in claim 6, characterized in that, The calculation of the conductor's galloping angle at the (n+1)th sampling time, based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time, includes: The Kalman gain is calculated based on the angle prediction error of the conductor at the (n+1)th sampling time. Based on the second sampling angle of the conductor at the (n+1)th sampling time and the Kalman gain, update the second sampling angle, predicted angle, and angle prediction error of the conductor at the (n+1)th sampling time; Based on the updated second sampling angle, predicted angle, and angle estimation error of the conductor at the (n+1)th sampling time, the galloping angle of the conductor at the (n+1)th sampling time is calculated.

8. The conductor galloping monitoring method as described in claim 1, characterized in that, The conductor is equipped with multiple monitoring points; the conductor galloping monitoring method further includes: Obtain the angular velocity and acceleration of m monitoring points at the nth sampling time; where m is an integer greater than or equal to 2; Based on the angular velocity and acceleration of the m monitoring points at the nth sampling time, a forward state sequence and a reverse state sequence are generated; Based on the forward state sequence and the reverse state sequence, the predicted dancing posture of the conductor at the nth sampling time is determined; If the predicted galloping posture of the conductor at the nth sampling time is greater than a preset value, the galloping of the conductor at the nth sampling time is determined to be abnormal.

9. The conductor galloping monitoring method as described in claim 1, characterized in that, The conductor galloping monitoring method also includes: Based on the acceleration of the conductor at the (n+1)th sampling time, the amplitude and frequency of the conductor's galloping at the (n+1)th sampling time are determined. Based on the dancing angle, dancing amplitude, and dancing frequency of the conductor at the (n+1)th sampling time, the dancing trajectory of the conductor is determined; the dancing trajectory is used to determine the dancing state of the conductor.

10. A computer storage medium, characterized in that, It includes a processor and a memory; the memory stores computer instructions that, when executed on the processor, cause the processor to perform the wire galloping monitoring method as described in any one of claims 1 to 8.

11. A conductor galloping monitoring system, characterized in that, include: The parameter acquisition module is used to detect the angular velocity and acceleration of the conductor; The main control module is used to calculate the first sampling angle of the conductor at the nth sampling time based on the angular velocity and acceleration of the conductor at the nth sampling time; Furthermore, based on the first sampling angle of the conductor at the nth sampling time, the predicted angle of the conductor at the (n+1)th sampling time is calculated; the main control module is also used to calculate the second sampling angle of the conductor at the (n+1)th sampling time based on the angular velocity and acceleration of the conductor at the (n+1)th sampling time; based on the second sampling angle of the conductor at the (n+1)th sampling time and the predicted angle of the conductor at the (n+1)th sampling time, the galloping angle of the conductor at the (n+1)th sampling time is calculated; the main control module is also used to output a warning message when it is determined that the galloping state of the conductor is abnormal based on the galloping angle of the conductor at the (n+1)th sampling time; where n is a positive integer.

12. The conductor galloping monitoring system as described in claim 11, characterized in that, The parameter acquisition module includes: The Beidou positioning module is used to obtain the position information of the conductor; the position information and the wagging angle are used to determine the wagging state of the conductor.

13. The conductor galloping monitoring system as described in claim 11, characterized in that, The conductor galloping monitoring system also includes: A photovoltaic module is used to convert solar energy into electrical energy to power the parameter acquisition module and the main control module.

Images