Lightning protection control device

Abstract

The invention provides a lightning protection control device. A power supply input is connected to a power grid providing commercial alternating current; the power output is connected to a load and provides power for the system; the grounding part is electrically connected with a circuit connected with the power input and the power output; a plurality of movable contacts including a first contact located at one end facing the power input end and a second contact located at the opposite end and electrically connected with the power output end; the guide rod piece is formed by combining a plurality of movable contact rod pieces between the guide rod piece and the movable contact rod pieces; the motor is driven, the guide piece moves in the first direction, at least one part of the guide piece is meshed with the guide piece, a first contact of the movable contact piece makes contact with the power input end, the first contact of the movable contact piece and the power input end move in the second direction opposite to the first direction, and the second contact of the movable contact piece makes contact with the power input end. The first contact of the movable contact piece is separated from the power supply input end; the lightning prediction sensor comprises an electric field sensor for detecting electric change in the atmosphere, a magnetic field sensor for detecting magnetic field change in the atmosphere, a weather sensor for detecting weather conditions in real time and a radar sensor for detecting charge distribution in cloud in the atmosphere; and the driving motor and the control unit are electrically connected to the thunder and lightning prediction sensor.

Description

TECHNICAL FIELD

[0001] The present invention relates to a lightning protection control device for preventing lightning by predicting the occurrence of lightning. BACKGROUND

[0002] The power system is one of the key infrastructures of modern society, and its stability directly affects the continuity and quality of power supply. However, lightning, as one of the natural phenomena, can cause great damage to the power system, and the damage caused by lightning to power equipment can cause serious economic losses. Lightning is a discharge in the atmosphere, in which a high-voltage current is generated instantaneously, and can cause overvoltage and overcurrent by powering devices and related equipment through the power grid. The damage caused by lightning can cause power outages, equipment damage, and fires, especially those using high voltage, such as transformers, transmission lines, power plants, and large electronic devices. In order to prevent such damage, it is necessary to apply lightning protection technology to the power system.

[0003] PRIOR ART DOCUMENTS

[0004] PATENT DOCUMENTS

[0005] Patent Document 1: Korean Patent Registration No. 10-2328862

[0006] Patent Document 2: Korean Patent Registration No. 10-2128952

[0007] Patent Document 3: Korean Patent Registration No. 10-1481625

[0008] Patent Document 4: Korean Patent Registration No. 10-0935642 SUMMARY

[0009] PROBLEMS TO BE SOLVED BY THE INVENTION

[0010] In order to prevent lightning damage to power equipment, lightning protection devices such as lightning arresters can be used. The lightning arrester can protect the equipment by rapidly discharging the lightning-induced multiple voltage to the ground. In addition to this, the grounding system has developed into an important defense measure to reduce the direct impact of lightning on the power system. These protection technologies can effectively reduce the physical damage caused by lightning, but their limitation is that they cannot be activated before the occurrence of lightning.

[0011] METHOD FOR SOLVING THE PROBLEM

[0012] According to one embodiment of the present disclosure, the lightning protection control device comprises: a power input connected to a system providing alternating current for commercial use, a power output connected to a load powered by the system, a grounding part electrically connected to the circuit connecting the power input and the power output, a first contact facing the power input and a second contact unit opposite and electrically connected to the power output, and a plurality of movable contact pieces including the second contact unit opposite and electrically connected to the power output; the plurality of movable contact pieces are combined with the guide piece, at least a part of which is engaged with the guide piece, the first contact of the movable contact piece is moved relative to the power input in a first direction, so that the first contact of the movable contact piece is moved relative to the movable contact piece in a first direction, and the first contact part of the movable contact piece is separated from the power input, so that the guide piece moves towards the movable contact piece in a second direction opposite to the first direction; it can include a lightning prediction sensor, including an electric field sensor for detecting changes in the electric field in the atmosphere, a magnetic field sensor for detecting changes in the magnetic field in the atmosphere, a weather sensor for detecting weather conditions in real time, a radar sensor for detecting the distribution of cloud charge in the atmosphere, and a control unit electrically connected to the driving motor and the lightning prediction sensor. The control unit receives data collected by the electric field sensor, the magnetic field sensor, the radar sensor and the weather sensor from the lightning prediction sensor, controls the driving motor to separate the first contact of the movable contact piece from the power input and the first contact of the movable contact piece, predicts the occurrence of lightning based on the data sent by the lightning prediction sensor when the data exceeds the preset threshold, the guide piece moves in the second direction and separates the power input, and by detecting the contact between the moving contact piece and the grounding part, the driving motor can be controlled to stop the movement of the guide piece.

[0013] Inventive effect

[0014] According to one embodiment of the present disclosure, the occurrence of lightning can be predicted in advance, and based on this, a lightning protection control device can be proposed to disconnect the connection between the power input connected to the power grid and the power output connected to the load (for example, household electronic products). Since the occurrence of lightning is foreseen in advance, the lightning protection control device can disconnect the electrical connection between the power input and the power output, because the movable contact part electrically connecting the power input and the power output is disconnected from the power input. Therefore, the lightning protection control device and the load can be prevented from being damaged by lightning. BRIEF DESCRIPTION OF DRAWINGS

[0015] For the purpose of the drawings description, the same or similar reference can be used for the same or similar elements.

[0016] Figure 1A A schematic diagram of the power input and the movable contact pieces in the contacts according to one embodiment of the present disclosure.

[0017] Figure 1B and Figure 1C The figure is shown where the guide is separated from the power input so that the power input and the movable contact piece are separated to prevent lightning.

[0018] Figure 2 is a configuration diagram of a lightning prediction sensor according to one embodiment of the disclosure.

[0019] Figure 3 is a diagram for explaining learning of an artificial intelligence model according to one embodiment of the disclosure.

[0020] BRIEF DESCRIPTION OF DRAWINGS

[0021] 100: lightning protection control unit;

[0022] 101: power input;

[0023] 102: power output;

[0024] 103: ground point;

[0025] 110: no movable contact;

[0026] 111: first contact;

[0027] 112: second contact;

[0028] 120: member;

[0029] 130: no guide;

[0030] 140: driving motor;

[0031] 141: linear section;

[0032] 200: lightning prediction sensor;

[0033] 201: electric field sensor;

[0034] 202: magnetic field sensor;

[0035] 203: radar sensor;

[0036] 204: weather sensor;

[0037] 300: control unit. DETAILED DESCRIPTION

[0038] The embodiments will be described in detail below with reference to the accompanying drawings. However, various changes can be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. Any changes, equivalents, or substitutions of the embodiments should be understood to be included in the scope of the right.

[0039] The specific structural or functional description of the embodiments is initiated only for illustrative purposes, and can be changed and implemented in various forms. Therefore, the embodiments are not limited to the specific form of disclosure, and the scope of the specification includes changes, unifications or substitutions included in the descriptive concept.

[0040] Terms such as first or second can be used to describe various components, but the interpretation of these terms should be used only to distinguish one component from another. For example, the first component can be named the second component, and likewise, the second component can be named the first component.

[0041] When a component is referred to as "connected" to another component, it is understood that it can be directly connected to or connected to another component, but there can be another component between the two.

[0042] The terms used in the embodiments are only for illustrative purposes and should not be interpreted as limiting. The singular expression includes the plural expression unless the context clearly means otherwise. In this specification, the term "include" or "have" should be understood to mean the presence of the functions, numbers, steps, actions, components, parts or combinations thereof described herein, and should not exclude the presence or addition of one or more other functions or numbers, steps, actions, components, parts or combinations thereof.

[0043] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person having ordinary knowledge in the art to which the embodiments belong. Terms such as defined in a general dictionary should be interpreted to have the same meaning as the meaning in the relevant description context, and should not be interpreted in an idealistic or overly formal sense unless specifically defined in this application.

[0044] In addition, when describing the drawings, the same reference numerals are given to the same elements regardless of the drawing codes, and the same repeated description is omitted. In describing the embodiments, if it is judged that a specific description of the relevant notification technology can unnecessarily obscure the gist of the embodiments, detailed explanation should be omitted.

[0045] The embodiments can be implemented in various types of products, including personal computers, notebook computers, tablet computers, smartphones, televisions, smart home appliances, smart cars, kiosks, and wearable devices.

[0046] In an embodiment, an artificial intelligence (AI) system is a computer system that implements human-level intelligence, unlike existing rule-based intelligent systems, which are systems that make judgments and learn by themselves. The more the AI system is used, the more the recognition rate is improved, and the seller's taste is more accurately understood, so the existing rule-based intelligent system is gradually being replaced by an AI system based on deep learning.

[0047] Artificial intelligence technology includes machine learning and element technology using machine learning. Machine learning is an algorithm technology that classifies / learns features of input data by itself, and element technology is a technology that simulates functions such as human brain cognition and judgment using machine learning algorithms such as deep learning, and consists of technical fields such as language understanding, visual understanding, reasoning / prediction, knowledge representation, and gesture representation.

[0048] Various fields to which artificial intelligence technology is applied are as follows. Language understanding is a technology for recognizing, adapting, and processing human language / text, and includes natural language processing, machine translation, a dialogue system, question and answer, and speech recognition / synthesis. Visual understanding is a technology for recognizing and processing objects like human vision, and includes object recognition, object tracking, image search, human recognition, scene understanding, space understanding, and image improvement. Reasoning prediction is a technology for logically reasoning and predicting information based on judgment, and includes knowledge / probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge representation is a technology for automatically processing human experience information into knowledge data, and includes knowledge building (data generation / classification) and knowledge management (data utilization). Motion control is a technology for controlling the motion of autonomous vehicles and robots, and includes motion control (navigation, collision, driving) and operation control (behavior control).

[0049] In general, in order to apply a machine learning algorithm to real life, training is performed in a trial-and-error method due to the nature of the basic method of machine learning. In particular, deep learning requires hundreds of thousands of iterations. Since it is not possible to do this in an actual physical external environment, the actual physical external environment is virtually implemented on a computer and learns through simulation.

[0050] Figure 1A is a schematic view of a power input and a movable contact piece in a contact according to one embodiment of the present disclosure. FIGS. 1b and 1c show the view in which the guide piece is separated from the power input so that the power input and the movable contact piece are separated from each other to prevent a lightning strike. Figure 2 is a configuration schematic view of a lightning prediction sensor according to one embodiment of the present disclosure.

[0051] Figure 3 is a schematic view for explaining learning of an artificial intelligence model according to one embodiment of the present disclosure.

[0052] According to one embodiment of the present disclosure, as shown in FIGS. 1a to 1c and Figure 2

[0053] The lightning protection control device 100 can be activated to prevent lightning damage to the power unit and the load. The lightning protection control unit 100 can include a power input 101 connected to a system providing commercial alternating current, a power output 102 connected to a load powered by a carrying power grid, and a grounding point 103 (e.g., ground) electrically connected to the circuit connecting the power input 101 and the power output 102. In one embodiment, the lightning protection control device 100 is a power device that powers the load, which can be configured to predict lightning conditions, thereby cutting off the electrical connection between the power input 101 and the power output 102.

[0054] In one embodiment, referring to FIGS. 1a to 1c, the lightning protection control device 100 can include a plurality of movable contact pieces 110. The movable contact pieces 110 can be configured to electrically connect the power input 101 and the power output 102. The movable contact pieces 110 can include a first contact 111 and a second contact 112. The first contact 111 faces the power input 101 and can be in contact with the power input 101. The second contact 112 is electrically connected to the power output 102. In one embodiment, the second contact 112 and the power output 102 can be electrically or physically connected by a spring piece 120 made of a conductive material. As described below, the spring piece 120 can be elastically deformed according to the movement of the guide piece 130 and the movable contact piece 110 according to the driving of the driving motor 140.

[0055] In one embodiment, referring to FIGS. 1a to 1c, the movable contact pieces 110 can be used to electrically transfer or disconnect the power from the power grid to the load. For example, the movable contact pieces 110 can be moved relative to the power input 101 by the driving motor 140, as described below. In one embodiment, if the movable contact pieces 110 are in contact with the power input 101, power can be supplied to the load through the system-power input 101-movable contact pieces 110-power output 102. In one embodiment, the control unit 300 of the lightning protection control unit 100 controls the driving motor 140 to move the movable contact pieces 110 away from the power input 101 based on the prediction of lightning occurrence by the lightning prediction sensor 200, so that the movable contact pieces 110 are separated from the power input 101. Therefore, the electrical connection between the power input 101 and the power output 102 through the movable contact pieces 110 can be disconnected.

[0056] ​In one embodiment, referring to FIGS. 1a-1c, a guide 130 can be disposed between the plurality of movable contacts 110. The plurality of movable contacts 110 can be combined and fixed with a guide 130. In one embodiment, the guide 130 can be coupled with a portion of the drive motor 140. For example, the guide 130 can be combined with a linear portion 141 of the drive motor 140. In one embodiment, the drive motor 140 can be an actuator that moves the linear portion 141 in a linear direction. In one embodiment, the drive motor 140 can be a linear motor. In one embodiment, the linear portion 141 can be linearly moved in a first or second direction based on the driving of the drive motor 140. In one embodiment, the first direction can be from the power output 102 to the power input 101. The second direction can be a direction opposite to the first direction, from the power input 101 to the power output 102.

[0057] In one embodiment, the drive motor 140 can move the guide 130 to establish contact between the first contact of the movable contact and the power input, or remove the contact, if necessary.

[0058] According to one embodiment of the present disclosure, as shown in FIGS. 1a-1c, Figure 1C and Figure 2As shown, the lightning protection control device 100 can include a lightning prediction sensor 200. In one embodiment, the lightning prediction sensor 200 can include an electric field sensor 201, a magnetic field sensor 202, a weather sensor 204, and a radar sensor 203. Each sensor can detect changes in the electric, magnetic, and weather conditions in the atmosphere in real time. In one embodiment, the electric field sensor 201 is capable of detecting electrical changes in the atmosphere. The electric field sensor 201 can detect an increase in voltage difference before a lightning strike occurs. Generally, a pattern of rapid rise in electric field can occur before a lightning strike. The electric field sensor 201 can include a sensitive electric field measurement sensor and a signal processing device. The electric field sensor 201 can monitor data in real time and transmit the data to the control unit 300. In one embodiment, the control unit 300 can be a personal computer (PC) in communication with the lightning protection control unit 100. In one embodiment, the control unit 300 can be part of the lightning protection control unit 100. In one embodiment, the magnetic field sensor 202 is capable of detecting changes in the magnetic field in the atmosphere. Before a lightning strike, the current in the atmosphere increases, which can also enhance the magnetic field. The magnetic field sensor 202 can detect this change and predict the likelihood of a lightning strike. In one embodiment, the radar sensor 203 is capable of detecting the charge distribution of atmospheric clouds. The control unit 300 can track the distribution of internal charges in the cloud through the radar sensor 203 to predict the likelihood of a lightning strike. In one embodiment, the control unit 203 can calculate the location and time of a possible lightning strike through the radar sensor 203. In one embodiment, the weather sensor 204 can detect weather conditions such as barometric pressure, temperature, and humidity in real time. The sensors detect lightning precursors in the atmosphere and forward the detected data to the control unit 300. The control unit 300 can analyze the data transmitted from the lightning prediction sensor 200 to analyze the data. The control unit 300 can determine whether the data such as changes in the electric field, changes in the magnetic field, weather conditions, and charge distribution within the cloud exceed a pre-set threshold. The control unit 300 determines that a lightning strike is predicted when the collected data exceeds the pre-set threshold, and controls the drive motor 140 to move the guide 130 in the second direction to separate the first contact 111 of the movable contact piece 110 from the power input 101. In one embodiment, the control unit 300 can move the guide 130 in the second direction relative to the power input 101 to bring the movable contact piece 110 into contact and / or electrical connection with the ground portion 103. The control unit 300 can sense that the movable contact piece 110 is in contact and / or electrical connection with the ground portion 103. The control unit 300 can control the drive motor 140 to stop the guide 130 when the movable contact piece 110 is detected to be in contact and / or electrical connection with the ground portion 103. This allows the power supply to be cut off to prevent overvoltage or overcurrent from being transferred to the load (e.g., power supply unit) due to a possible lightning strike in the system or the control unit 100 for lightning protection.In one embodiment, the control unit 300 can connect the first contact 111 of the movable contact member 110 to the power input 101 by controlling the driving motor 140 to move the guide member 130 in the first direction according to the probability of lightning falling below a preset value. Accordingly, power can be supplied to the load through the system - power input 101 - movable contact member 110 - power output 102.

[0059] In one embodiment, in the control unit 300, an algorithm for predicting lightning occurrence can be designed in such a way that data collected from the lightning prediction sensor 200 is processed in real time, and the possibility of lightning strike is warned when a certain threshold is exceeded. In one embodiment, the electric field sensor 201 can measure the magnitude of the atmospheric electric field in real time and calculate the rate of change (Δ) of the electric field over a certain period of time. The magnetic field sensor 202 measures the change in the magnetic field in the atmosphere, and the rate of change (Δ) of the magnetic field. The weather sensor 204 can collect data such as air pressure, temperature, humidity, wind, etc. in real time to monitor the amount of change in weather conditions (e.g., air pressure change (ΔP), temperature change (ΔT), humidity change (ΔH)).

[0060] In one embodiment, the threshold values can be input in advance into the control unit 300. For example, the threshold values can be stored in the memory of the control unit 300. The threshold values can include a threshold value for the rate of change of the electric field, a threshold value for the rate of change of the magnetic field, a threshold value for the rate of change of the charge amount, a threshold value for the rate of change of the air pressure, a threshold value for the rate of change of the temperature, and a threshold value for the rate of change of the humidity. The threshold values can be based on experimental data or historical lightning data. In one embodiment, the threshold value for the rate of change of the electric field can be 3 kV / differential, the threshold value for the rate of change of the magnetic field can be 2 nT / sec, the threshold value for the rate of change of the air pressure can be 1.5 hPa / 1 hour, the threshold value for the rate of change of the temperature can be 2℃ / 10 differential, the threshold value for the cloud charge can be 100 C (Coulomb), and the threshold value for the rate of change of the humidity can be 5% / 10 differential.

[0061] In one embodiment, the control unit 300 can calculate the probability of lightning strike based on the rate of change of the electric field collected from the electric field sensor 201, the rate of change of the magnetic field collected from the magnetic field sensor 202, and the rate of change of the charge amount collected from the radar sensor 203 all exceeding the preset threshold values, and at least two of the atmospheric pressure change, the temperature change, and the humidity change detected by the weather sensor 204 exceeding the set threshold values. The control unit 300 can predict the occurrence of lightning by determining that the probability based on the occurrence of lightning exceeds a critical probability, thereby controlling the driving motor 140 to separate the movable contact member 110 from the power input 101.

[0062] In one embodiment, the probability of lightning strike can be calculated based on the following mathematical equation 1 in the control unit 300.

[0063] Mathematical Equation 1:

[0064] In one embodiment, the maximum measurable rate of change of the electric field, the maximum measurable rate of change in the magnetic field, the maximum measurable amount of charge, the rate of change of the measurable air pressure, the maximum measurable rate of change of the temperature, and the maximum measurable rate of change of the humidity can be values predetermined from historical data collected in the control unit 300 through the lightning prediction sensor 200.

[0065] In one embodiment, W1 to W4 can be determined by considering the contribution of the electric field sensor 201, the magnetic field sensor 202, the radar sensor 203, and the air pressure sensor 204 to lightning prediction. For example, W1 = 0.4, W2 = 0.3, W3 = 0.2, W4 = 0.1.

[0066] In one embodiment, the probability of lightning can increase if the rate of change of the electric field Δ exceeds the threshold value Eth. In one embodiment, the probability of lightning can increase if the rate of change of the magnetic field ΔB exceeds the threshold value Bth. In one embodiment, the probability of lightning can increase if the rate of change of the electric field ΔC exceeds the threshold value Cth.

[0067] In one embodiment, the control unit 300 determines to predict the occurrence of lightning according to the probability of lightning occurrence exceeding a critical probability, and controls the driving motor 140 so that the guide 130 and the movable contact 110 can be moved in the second direction with respect to the power input 101. In one embodiment, the critical probability can be set to 90%, and can be modified according to past lightning patterns.

[0068] According to one embodiment of the present disclosure, the lightning protection control device 100 can minimize damage to the power device (e.g., the lightning protection control device 100) and the load by turning off the power in advance before lightning occurs. In addition, the grounding point 103 can play a role in safely guiding the electric pulse generated by lightning into the ground. In one embodiment, in a state where lightning is expected to occur, the movable contact 110 is electrically connected to the grounding point 103, so that damage to the lightning protection control device 100 by lightning can be effectively prevented.

[0069] Specifically, the operation process of the present invention is as follows. The sensors that detect changes in atmospheric electric field and magnetic field, weather conditions, and the amount of charge in the cloud (e.g., the lightning prediction sensor 200) collect data in real time and transmit it to the control unit 300. The control unit 300 analyzes the probability of lightning based on the data, and, if the data exceeds the set threshold value, it can immediately activate the driving motor 140 to separate the power input 101 from the movable contact 110. This can prevent electric shock caused by lightning, which can be transmitted from the power grid to the load.

[0070] Further, the control unit 300 communicates with the Japan Meteorological Agency through a network to predict the possibility of lightning strike. If the probability of lightning strike at the installation location of the lightning protection control device 100 exceeds a critical probability, the driving motor 140 can be controlled to disconnect the movable contact member 110 from the power input 101.

[0071] In one embodiment, the controller 300 can include an artificial intelligence model trained to modify the threshold of the electric field change rate (E th), the threshold of the magnetic field change rate (B th), the threshold of the charge change rate (C th), the threshold of the air pressure change rate (P th), the threshold of the temperature change rate (T th), and the threshold of the humidity change rate (H th). The artificial intelligence model, for the electric field sensor 201, can learn historical data collected from the magnetic field sensor 202, the radar sensor 203, and the weather sensor 204, as well as actual lightning occurrence patterns. The artificial intelligence model can calculate the probability of lightning occurrence according to Mathematical Formula 1, based on data collected in real time from the electric field sensor, the magnetic field sensor, the radar sensor, and the weather sensor.

[0072] In one embodiment, artificial intelligence (AI) (e.g., an artificial intelligence model of the control unit) refers to a technology that simulates human learning, reasoning, and perception capabilities and implements them in a computer, and can include concepts such as machine learning, symbolic logic, etc. Machine learning (ML) is an algorithmic technique that classifies or learns the characteristics of input data on its own. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the analysis result, and can make a judgment or prediction based on the learning result. In addition, a technology that uses a machine learning algorithm to mimic human brain cognition, judgment, etc. can also be understood as a category of artificial intelligence. For example, it can include technical fields such as language understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control.

[0073] Machine learning can refer to a process of training a neural network model using experience of processing data. Machine learning can mean that computer software has improved its ability to process data on its own. A neural network model is constructed by modeling the correlation between data, which can be represented by a plurality of parameters. A new Kyung Network model extracts features from given data, analyzes them, and derives correlations between data, and it can be said that machine learning repeats this process to optimize the parameters of the neural network model. For example, a neural network model can learn a mapping (correlation) between input and output of data given in the form of input / output pairs. Alternatively, a neural network model can derive regularity between a given data set and learn a relationship even if only input data is given.

[0074] An artificial intelligence model or neural network model can be designed to replicate the structure of the human brain on a computer and can include a plurality of network nodes that mimic and weight neurons in a human neural network. The plurality of network nodes can mimic synaptic activity of neurons sending and receiving signals through synapses and are connected to each other. In an AI learning model, the plurality of network nodes can be located in layers of different depths and send and receive data according to their convolutional connections. For example, the AI learning model can be an artificial neural network, a convolutional neural network (CNN), etc. As an embodiment, the AI learning model can perform machine learning according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning can include decision trees, Bayesian networks, support vector machines, artificial neural networks, ada-boost, perceptrons, genetic programming, and clustering.

[0075] A CNN is a kind of multi-layer perceptron designed to use minimal preprocessing. A CNN consists of one or more product layers and typical neural network layers located thereon, and has additional weight and pooling layers. Due to this structure, a CNN can fully utilize input data from a two-dimensional structure. Compared to other deep learning structures, a CNN performs well in both video and audio fields. A CNN can also be trained through standard backpropagation. A CNN tends to be easier to train than other feedforward neural network techniques and has an advantage of using fewer parameters.

[0076] A convolutional network is a neural network containing a set of nodes with tied parameters. The increase in the size of available training data and the availability of computing power, along with algorithmic advances such as discriminative linear units and dropout training, have greatly improved many computer vision tasks. In large data sets, such as those available today for many tasks, overfitting is not a concern, and increasing network size can improve test accuracy. Optimal use of computing resources is a limiting factor. For this, a decentralized, scalable implementation of a deep neural network can be used.

[0077] Figure 3 FIG. 1 is a schematic diagram for explaining learning of an artificial intelligence model according to one embodiment of the disclosure.

[0078] In one embodiment, the controller 300 can include an artificial intelligence model trained to modify the threshold of the electric field change rate (E th), the threshold of the magnetic field change rate (B th), the threshold of the charge change rate (C th), the threshold of the atmospheric pressure change rate (P th), the threshold of the temperature change rate (T th), and the threshold of the humidity change rate (H th). The artificial intelligence model, for the electric field sensor 201, can learn historical data collected from the magnetic field sensor 202, the radar sensor 203, and the weather sensor 204, and an actual lightning occurrence pattern. The artificial intelligence model can calculate the probability of lightning occurrence according to Mathematical Formula 1 according to data collected in real time from the electric field sensor, the magnetic field sensor, the radar sensor, and the weather sensor. The artificial intelligence model can increase the threshold of the electric field change rate, the threshold of the magnetic field change rate, the threshold of the charge amount, the threshold of the atmospheric pressure change rate, the threshold of the temperature change rate, and the threshold of the humidity change rate by a predetermined value in a state in which the probability of lightning occurrence exceeds a critical probability and lightning occurrence is not detected. Accordingly, the accuracy of lightning detection can be improved. In addition, if the probability of lightning does not match the actual occurrence rate of lightning, the AI model can dynamically adjust the value of the threshold to adjust the probability of lightning. Accordingly, the artificial intelligence model can dynamically adjust the threshold by training a machine learning algorithm on the collected data. The artificial intelligence model can improve prediction accuracy by using Logistic Regression, Random Forest, Neural Networks, etc., which can learn lightning patterns from historical data.

[0079] In one embodiment, the control unit 300 can train the artificial intelligence model using Gradient Descent (GD) or Stochastic Gradient Descent (SGD) techniques. In one embodiment, the control unit 300 can utilize a loss function designed by the output of the AI model and a label.

[0080] In one embodiment, the control unit 300 can calculate a training error using a predefined loss function. The loss function can define a label, an output, and a parameter as input variables, where the parameter can be set by a weight in the AI model. For example, the loss function can be designed in the form of an entropy form or the like, and various techniques or methods can be employed in embodiments in which the loss function is designed.

[0081] In one embodiment, the control unit 300 can use a backpropagation technique to identify a weight that affects the training error. Here, the weight can be a relationship between nodes in the AI model. The control unit 300 can use an SGD technique that uses a label and an output to optimize the weight found by the backpropagation technique. For example, the control unit 300 can update the weight of the loss function defined based on the label, the output, and the weight using the SGD technique.

[0082] The above-described embodiments can be implemented as hardware components, software components, and / or combinations of hardware and software components. For example, the devices, methods, and components described in the embodiments can be implemented using one or more general purpose or special purpose computers, such as processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable gate arrays (FPGAs), programmable logic units (PLUs), microprocessors, or any other devices capable of executing and responding to instructions. The processing units can execute an operating system (OS) and one or more software applications that execute on the operating system. The processing units can also access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing units can be described as a single processing unit, but one of ordinary skill in the art can know that the processing units can include multiple processing elements and / or multiple types of processing elements. For example, a processing unit can include multiple processors or a processor and a controller. Other processing configurations can also be used, such as parallel processors.

[0083] The method according to the present embodiment can be implemented in the form of program instructions that can be executed by various computer means and recorded on a computer-readable medium. The computer-readable medium can include program instructions, data files, data structures, etc. individually or in combination. The program commands recorded in the medium can be designed and configured specifically for the embodiments, or can be known and available to computer software artisans. Examples of computer-readable recording media include magnetic media (such as hard disks, floppy disks, and magnetic tapes), optical media (such as CD-ROM and DVD), magneto-optical disk media (such as Floptical disk), and hardware devices specially configured to store and execute program commands (such as ROM, RAM, flash memory, etc.). Examples of program instructions include machine code (e.g., code generated by a compiler) and high-level language code (e.g., executable by a computer using an interpreter). The hardware device can be configured to run as one or more software modules to perform the operations of the embodiments, or vice versa.

[0084] The software can include one or more combinations of computer programs, codes, instructions, or the like, and can be configured to cause the processing unit to operate as intended, or can individually or collectively instruct the processing unit. The software and / or data can be permanently or temporarily included in any type of machine, component, physical device, virtual device, computer storage medium, or apparatus, or in a signal wave transmitted to the processing unit for interpretation or to provide instructions or data to the processing unit. The software is distributed on a networked computer system and can be stored or executed in a distributed manner. The software and data can be stored on one or more computer-readable recording media.

[0085] While the above embodiments have been described with limited drawings, various technical modifications and changes can be made by those skilled in the art based on the above circumstances. For example, if the described technology is executed in a different order from the described method, and / or if the components of the described system, structure, device, circuit, etc. are combined or combined in a different way from the described method, or are replaced or replaced by other components or equivalents, appropriate results can be obtained.

[0086] Therefore, other embodiments, other implementations, and those equivalent to the patent claims also belong to the scope of the claims described below.

Claims

1. A lightning protection control device, in, include: The power input is connected to the power grid that provides commercial AC power; A power output connected to the load and providing power to the system; The grounding part is electrically connected to the circuit that connects the power input and the power output; Multiple movable contacts, including a first contact located at one end facing the power input terminal and a second contact located at the opposite end and electrically connected to the power output terminal; It is composed of a guide rod and multiple movable contact rods combined together; A drive motor is provided, the guide moves along a first direction such that at least a portion of it engages with the guide, the first contact of the movable contact contacts the power input terminal, and the first contact of the movable contact moves along a second direction opposite to the first direction such that the first contact of the movable contact separates from the power input terminal. Lightning prediction sensors include electric field sensors that detect changes in electrical activity in the atmosphere, magnetic field sensors that detect changes in magnetic fields in the atmosphere, weather sensors that monitor weather conditions in real time, and radar sensors that detect the distribution of electrical charges in clouds. The drive motor and control unit are electrically connected to the lightning prediction sensor. In the control unit, Data collected from electric field sensors, magnetic field sensors, radar sensors, and weather sensors is received from lightning prediction sensors. When the data sent by the lightning prediction sensor exceeds a preset threshold, based on the prediction of lightning occurrence, the guide moves along the second direction, controlling the drive motor to separate the first contact point of the movable contact from the power input. Based on the detection of contact between the movable contact and the grounding part, the drive motor is controlled to stop the movement of the guide component.

2. The lightning protection control device according to claim 1, wherein, It also includes a spring element, which is electrically connected to the second contact point of the movable contact element and the power output end, and elastically deforms based on the movement of the guide element.

3. The lightning protection control device according to claim 1, wherein, In the control unit, The rate of change of the electric field collected by the electric field sensor, the rate of change of the magnetic field collected by the magnetic field sensor, and the amount of charge collected by the radar sensor all exceeded the preset threshold. The probability of a lightning strike is calculated based on the fact that at least two of the changes in air pressure, temperature, and humidity detected by weather sensors exceed a set threshold. The occurrence of a lightning strike is predicted based on the probability of the strike exceeding a critical probability.

Images