High-voltage transmission line arc detection method and system

Abstract

The invention relates to the technical field of high-voltage power transmission, in particular to a high-voltage power transmission line arc detection method and system, and the system mainly comprises a signal receiving module, an ultraviolet detection module, a sound receiving module, a transmission module and a neural network calculation module. And the signal processing module is used for receiving a flash signal generated by discharging in the air, receiving noise signals of thunder and electric sparks, receiving the processed signal through the neural network calculation module, and identifying that the input signal is a thunder signal or high-voltage discharging of the high-voltage power transmission line. Through the content, whether the result shows that high-voltage discharge exists when lightning occurs or not is determined, so that the probability of false alarm is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage power transmission, and in particular to a method and system for detecting arcs in high-voltage power transmission lines. Background Art

[0002] As the world's climate changes, there are more severe weather with dense rain and lightning. In addition, arc detection is becoming more and more important due to the development of high-voltage transmission lines. Since high-voltage transmission lines emit strong electromagnetic waves, it is easy to detect high-voltage electromagnetic waves with simple electronic circuits. Since natural lightning is also a source of strong electromagnetic waves, simple circuits of electromagnetic waves can cause false alarms.

[0003] The traditional method for detecting high-voltage discharge in transmission lines is to use ultraviolet sensors for quantitative measurement. The problem with this measurement is that it only measures ultraviolet signals, while the ultraviolet rays generated by line discharges will be overwhelmed by the ultraviolet rays generated by lightning discharges. As a result, lightning discharges will be misjudged as line discharges. Summary of the invention

[0004] The object of the present invention is to provide a method and system for detecting arc in a high voltage transmission line to solve the above-mentioned problems in the prior art.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] In a first aspect, the present invention provides a high-voltage transmission line arc detection system, comprising:

[0007] A signal receiving module is used to receive electromagnetic wave signals generated by lightning and high-voltage transmission lines;

[0008] Ultraviolet detection module, used to receive flash signals generated by discharge in the air;

[0009] A sound receiving module, used for receiving noise signals of thunder and electric sparks;

[0010] A plurality of transmission modules, wherein the plurality of transmission modules are respectively connected to the signal receiving module, the ultraviolet detection module and the sound receiving module, and are used to process the received signals and transmit them to the neural network calculation module;

[0011] The neural network calculation module is used to receive the processed signal and identify the input signal as a lightning signal or a high-voltage discharge of a high-voltage transmission line.

[0012] Preferably, the neural network calculation module includes:

[0013] Establishing a neural network model, wherein the neural network prediction model includes an input layer, a hidden layer and an output layer;

[0014] The results predicted by the neural network model are used to determine whether high voltage discharge exists when lightning occurs.

[0015] Preferably, the establishing of the neural network model comprises:

[0016] Obtain test data, extract and normalize test data features, and divide the test data into training sets and validation sets;

[0017] The neural network prediction model is trained through the training set to iteratively update the weights and biases, the optimal weights and biases are selected through the validation set, and the optimal neural network model is output.

[0018] Preferably, the neural network prediction model includes:

[0019] Set the objective function, which includes:

[0020]

[0021] In the formula, S a is the output of the objective function, ρ is the number of training samples, is the number of hidden layer nodes, Y j is the bias term of the jth node in the hidden layer, β is the input layer input, and λ ik is the kth feature of the i-th sample, X j is the connection weight from the jth hidden layer node to the output node, ξ is the regularization coefficient, μ is the weight coefficient, α ij is the output of the i-th sample at the j-th hidden layer node, γ a is the activation function, κ b is the dimension of the input feature, ε ij is the connection weight from the i-th node in the input layer to the j-th node in the hidden layer.

[0022] Preferably, a loss function is also included:

[0023]

[0024] Where BCELoss(x,y) is the output of the loss function, x i is the true label, y i is the probability value of the model probability, and n is the number of samples.

[0025] Preferably, determining whether high voltage discharge exists when lightning occurs according to the result predicted by the neural network model comprises:

[0026] Get the output of the objective function and determine whether the output value is greater than or equal to 0.5. If it is greater than or equal to 0.5, it is determined to be lightning.

[0027] If it is less than 0.5, it is judged as an arc.

[0028] Preferably, the transmission module includes a first circuit module, an analog-to-digital conversion module and a Fourier transform module, the first circuit module is connected to the analog-to-digital conversion module, the analog-to-digital conversion module is connected to the Fourier transform module, and the Fourier transform module is connected to the neural network calculation module;

[0029] The first circuit module is used to continue filtering and biasing the received signal and protect the existing circuit structure;

[0030] The analog-to-digital conversion module is used to continuously convert the input at a number of intervals when the input exceeds a set threshold;

[0031] The Fourier transform module is used to convert the input signal into a frequency domain diagram.

[0032] In a second aspect, the present invention further provides a method for detecting arcs in a high-voltage transmission line, wherein arc detection is performed by using the above-mentioned arc detection system for a high-voltage transmission line.

[0033] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0034] The system provided by the present invention mainly includes a signal receiving module, an ultraviolet detection module, a sound receiving module, a transmission module and a neural network calculation module, which receive electromagnetic wave signals generated by lightning and high-voltage transmission lines, receive flash signals generated by discharge in the air, receive noise signals of thunder and electric sparks, and receive processed signals through the neural network calculation module, and identify the input signal as a lightning signal or a high-voltage discharge of a high-voltage transmission line. Through the above content, it is determined whether the result shows that there is a high-voltage discharge when lightning occurs, so as to reduce the probability of false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 It is a schematic diagram of the module structure of the present invention;

[0037] Figure 2 Schematic diagram of the neural network model of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0039] The division of modules in this application is a logical division. There may be other division methods when implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0040] In addition, the connection, coupling or communication in the present application may be a direct connection, coupling or communication between related objects, or an indirect connection, coupling or communication through other devices. Furthermore, the connection, coupling or communication between objects may be electrical or other similar forms, which are not limited in the present application.

[0041] The modules or submodules described independently may be physically separated or not: they may be implemented by software or hardware, and some modules or submodules may be implemented by software, and the processor may call the software to implement the functions of these modules or submodules, and other modules or submodules may be implemented by hardware, such as by hardware circuits. In addition, some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.

[0042] How to distinguish between arcs in high-voltage transmission lines and natural lightning. - Arcs in high-voltage transmission lines and natural lightning will both emit three types of signals: ultraviolet rays, electromagnetic waves, and sound waves.

[0043] Ultraviolet radiation: Ultraviolet light (i.e. UV light) is produced by the ionization of gases. We can use ultraviolet detectors to detect flashes produced by discharges in the air. This light travels from the source to the detector at the speed of light (C = 299,792,458 m / s).

[0044] Electromagnetic waves: Electromagnetic waves can be detected with antennas. This wave is generated by the impulse of electrons traveling in space. Lightning waves are generated by discharges of several million volts. According to experiments, lightning electromagnetic waves have a wave front shape of 1.2 to 50 microseconds and a wave tail of 200 to 500 microseconds. Electromagnetic waves propagate at the speed of light.

[0045] The voltage of a transmission line ranges from a few kilovolts to about 850 kilovolts (the limit of current technology), and this waveform lasts for hundreds of microseconds. Therefore, the shape of the electromagnetic wave's wavefront can determine whether the electromagnetic wave comes from lightning or a transmission line.

[0046] Thunder and lightning: Thunder is an acoustic shock wave caused by the extreme heat produced by lightning. Like all gases, air expands when heated. The faster it is heated, the faster it expands. When air is heated to 30,000°C in a fraction of a second, a phenomenon called "explosive expansion" occurs, forming a shock wave similar to a sonic boom. Sound waves travel at the speed of sound = 343 m / s.

[0047] Please refer to Figure 1-2 As shown, a high voltage transmission line arc detection system comprises:

[0048] The signal receiving module is used to receive electromagnetic wave signals generated by lightning and high-voltage transmission lines. Specifically, it is an electronic circuit with an antenna that can pick up electromagnetic waves generated by lightning and high-voltage transmission lines.

[0049] The ultraviolet detection module is used to receive the flash signal generated by the discharge in the air. Specifically, the central spectrum is 200 nanometers (200 nanometers ± 50 nanometers). In order to avoid interference from sunlight, the spectrum range of 150 nanometers to 250 nanometers is selected. This interval is blocked by the atmospheric ozone layer;

[0050] A sound receiving module, used for receiving noise signals of thunder and electric sparks;

[0051] A plurality of transmission modules, wherein the plurality of transmission modules are respectively connected to the signal receiving module, the ultraviolet detection module and the sound receiving module, and are used to process the received signals and transmit them to the neural network calculation module;

[0052] The neural network calculation module is used to receive the processed signal and identify the input signal as a lightning signal or a high-voltage discharge of a high-voltage transmission line.

[0053] The traditional method for detecting high-voltage discharge in transmission lines is to use ultraviolet sensors for quantitative measurement. The problem with this measurement is that it only measures ultraviolet signals, while the ultraviolet rays generated by line discharges will be overwhelmed by the ultraviolet rays generated by lightning discharges. As a result, lightning discharges will be misjudged as line discharges.

[0054] Not only can it quantitatively measure ultraviolet rays, but it can also quantitatively measure electromagnetic waves and sound waves.

[0055] For the measured physical values, a neural network algorithm is used to qualitatively determine whether it is a lightning discharge or a line discharge. This judgment is based on the data learning and prediction capabilities of the neural network, that is, the data transmitted to the neural network includes not only the characteristic signals of ultraviolet rays, electromagnetic waves, and sound waves emitted by lightning discharge, but also the characteristic signals of ultraviolet rays, electromagnetic waves, and sound waves emitted by line discharge. There is a difference between these two signals, which is distinguished by the neural network, and then a judgment is made as to whether line discharge really occurred or lightning discharge just occurred in nature. The interference scene of lightning discharge is then identified and eliminated, thereby improving the accuracy of line discharge detection.

[0056] In an exemplary embodiment of the present invention, the neural network computing module includes:

[0057] A neural network model is established, wherein the neural network prediction model includes an input layer, a hidden layer, and an output layer; and whether high voltage discharge exists when lightning occurs is determined according to the prediction result of the neural network model.

[0058] Specifically, establishing a neural network model includes:

[0059] Acquire test data, extract and normalize test data features, and divide the test data into training set and validation set; train the neural network prediction model through the training set to iteratively update weights and biases, select the optimal weights and biases through the validation set, and output the optimal neural network model.

[0060] The neural network prediction model includes: setting the objective function, which includes:

[0061]

[0062] In the formula, S a is the output of the objective function, ρ is the number of training samples, is the number of hidden layer nodes, Y j is the bias term of the jth node in the hidden layer, β is the input layer input, and λ ik is the kth feature of the i-th sample, X j is the connection weight from the jth hidden layer node to the output node, ξ is the regularization coefficient, μ is the weight coefficient, α ij is the output of the i-th sample at the j-th hidden layer node, γ a is the activation function, κ b is the dimension of the input feature, ε ij is the connection weight from the i-th node in the input layer to the j-th node in the hidden layer.

[0063] An exemplary embodiment of the present invention further includes a loss function:

[0064]

[0065] Where BCELoss(x,y) is the output of the loss function, x i is the true label, y i is the probability value of the model probability, and n is the number of samples.

[0066] Specifically, determining whether high voltage discharge exists when lightning occurs according to the prediction result of the neural network model includes:

[0067] Get the output of the objective function and determine whether the output value is greater than or equal to 0.5. If it is greater than or equal to 0.5, it is determined to be lightning; if it is less than 0.5, it is determined to be an arc.

[0068] In an exemplary embodiment of the present invention, the transmission module includes a first circuit module, an analog-to-digital conversion module and a Fourier transform module, wherein the first circuit module is connected to the analog-to-digital conversion module, the analog-to-digital conversion module is connected to the Fourier transform module, and the Fourier transform module is connected to the neural network calculation module;

[0069] The first circuit module is used to continue filtering and biasing the received signal and protect the existing circuit structure. Specifically, the output of this circuit is an analog signal. The circuit maintains a maximum value of at least 100 microseconds. If no new signal enters, this maximum value will decay to zero within 1000 microseconds.

[0070] The analog-to-digital conversion module is used to continuously convert the input at several intervals when the input exceeds a set threshold. Specifically, the analog-to-digital conversion module connected to the signal receiving module and the ultraviolet detection module continuously converts the input at intervals of 50 microseconds, and the analog-to-digital conversion module connected to the sound receiving module continuously converts the input at intervals of 1000 microseconds.

[0071] The Fourier transform module is used to convert the input signal into a frequency domain chart. In order to reduce the computing cost, the circuit only detects the moment when the derivative changes its sign.

[0072] The system provided by the present invention mainly includes a signal receiving module, an ultraviolet detection module, a sound receiving module, a transmission module and a neural network calculation module, which receive electromagnetic wave signals generated by lightning and high-voltage transmission lines, receive flash signals generated by discharge in the air, receive noise signals of thunder and electric sparks, and receive processed signals through the neural network calculation module, and identify the input signal as a lightning signal or a high-voltage discharge of a high-voltage transmission line. Through the above content, it is determined whether the result shows that there is a high-voltage discharge when lightning occurs, so as to reduce the probability of false alarms.

[0073] Not only can it quantitatively measure ultraviolet rays, but it can also quantitatively measure electromagnetic waves and sound waves.

[0074] For the measured physical values, a neural network algorithm is used to qualitatively determine whether it is a lightning discharge or a line discharge. This judgment is based on the data learning and prediction capabilities of the neural network, that is, the data transmitted to the neural network includes not only the characteristic signals of ultraviolet rays, electromagnetic waves, and sound waves emitted by lightning discharge, but also the characteristic signals of ultraviolet rays, electromagnetic waves, and sound waves emitted by line discharge. There is a difference between these two signals, and this difference can be easily distinguished through the neural network, and then a judgment is made as to whether the line discharge really occurred or just a lightning discharge occurred in nature, thereby identifying and eliminating the interference scene of lightning discharge, thereby improving the accuracy of line discharge detection.

[0075] Of course, in this embodiment, the test data may also be processed into a data matrix, the test data matrix may be optimized into an operation matrix, and the operation matrix may be used to calculate the prediction.

[0076]

[0077] Where W1(m,n) is the nth row and mth column of the first data matrix, W2(n,k) is the kth row and nth column of the first data matrix, W1(m,n) is the 1st row and kth column of the third data matrix, X m is the mth optimization coefficient of the first data matrix, L1 (n) is the nth optimization result of the first data matrix, Q1,n is the nth optimization coefficient of the second data matrix, L2 (k) is the k-th optimization result of the second data matrix, and Q1, k is the k-th optimization coefficient of the third data matrix.

[0078] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A high voltage transmission line arc detection system, characterized in that: include: A signal receiving module is used to receive electromagnetic wave signals generated by lightning and high-voltage transmission lines; Ultraviolet detection module, used to receive flash signals generated by discharge in the air; A sound receiving module, used for receiving noise signals of thunder and electric sparks; A plurality of transmission modules, wherein the plurality of transmission modules are respectively connected to the signal receiving module, the ultraviolet detection module and the sound receiving module, and are used to process the received signals and transmit them to the neural network calculation module; The neural network calculation module is used to receive the processed signal and identify the input signal as a lightning signal or a high-voltage discharge of a high-voltage transmission line.

2. A high voltage transmission line arc detection system according to claim 1, characterized in that: The neural network calculation module includes: Establishing a neural network model, wherein the neural network prediction model includes an input layer, a hidden layer and an output layer; The results predicted by the neural network model are used to determine whether high voltage discharge exists when lightning occurs.

3. A high voltage transmission line arc detection system according to claim 2, characterized in that: The establishment of the neural network model comprises: Obtain test data, extract and normalize test data features, and divide the test data into training sets and validation sets; The neural network prediction model is trained through the training set to iteratively update the weights and biases, the optimal weights and biases are selected through the validation set, and the optimal neural network model is output.

4. A high voltage transmission line arc detection system according to claim 3, characterized in that: The neural network prediction model includes: Set the objective function, which includes: In the formula, S a is the output of the objective function, ρ is the number of training samples, is the number of hidden layer nodes, Y j is the bias term of the jth node in the hidden layer, β is the input layer input, and λ ik is the kth feature of the i-th sample, X j is the connection weight from the jth hidden layer node to the output node, ξ is the regularization coefficient, μ is the weight coefficient, α ij is the output of the i-th sample at the j-th hidden layer node, γ a is the activation function, κ b is the dimension of the input feature, ε ij is the connection weight from the i-th node in the input layer to the j-th node in the hidden layer.

5. A high voltage transmission line arc detection system according to claim 4, characterized in that: It also includes the loss function: Where BCELoss(x,y) is the output of the loss function, x i is the true label, y i is the probability value of the model probability, and n is the number of samples.

6. A high voltage transmission line arc detection system according to claim 5, characterized in that: Determining whether high voltage discharge exists when lightning occurs based on the prediction result of the neural network model includes: Get the output of the objective function and determine whether the output value is greater than or equal to 0.

5. If it is greater than or equal to 0.5, it is determined to be lightning. If it is less than 0.5, it is judged as an arc.

7. The arc detection system for high-voltage transmission lines according to claim 1, characterized in that: The transmission module includes a first circuit module, an analog-to-digital conversion module and a Fourier transform module, the first circuit module is connected to the analog-to-digital conversion module, the analog-to-digital conversion module is connected to the Fourier transform module, and the Fourier transform module is connected to the neural network calculation module; The first circuit module is used to continue filtering and biasing the received signal and protect the existing circuit structure; The analog-to-digital conversion module is used to continuously convert the input at a number of intervals when the input exceeds a set threshold; The Fourier transform module is used to convert the input signal into a frequency domain diagram.

8. A method for detecting arc in a high voltage transmission line, characterized in that: Arc detection is performed using an arc detection system for a high-voltage transmission line as described in any one of claims 1 to 7.

Images