KR102648377B1 - Intelligent integrated wiring board with network cable quality monitoring - Google Patents

Abstract

The present invention relates to an AI integrated distribution board capable of monitoring network cable quality. More specifically, an AI integration capable of monitoring network cable quality that continuously monitors network quality and uses collected data to prevent potential network failures in advance. It's about the wiring board.

Description

AI integrated wiring board capable of monitoring network cable quality and method using the same {Intelligent integrated wiring board with network cable quality monitoring}

The present invention relates to an AI integrated distribution board capable of monitoring network cable quality. More specifically, network cable quality monitoring that continuously monitors network quality with AI technology and uses collected data to prevent potential network failures in advance. It is about a possible AI integrated wiring board.

In general, network devices such as hubs, switches, routers, patch panels, 110 blocks, terminal blocks, IP cameras, NVRs, etc. are electrically connected to communication equipment on the subscriber side through LAN cables to enable communication with each other. In other words, a local area network (LAN) connects large numbers of personal computers, workstations, printers, and file servers within a modern office.

A LAN system is constructed by physically connecting all of these devices with a copper conductor LAN cable, and an 8-wire cable consisting of four twisted wire pairs is generally most widely used. Some LAN cables contain a thin layer of flexible foil that acts as an electrostatic shield.

Each end of the cable is connected into a connector that complies with industry standards. Generally, LAN cables are installed through walls, floors, and ceilings of buildings. Additionally, LAN cable systems require continuous maintenance, upgrades, and troubleshooting because LAN cables and connectors are prone to breakage and require moving offices and equipment, and further adding new equipment.

In order to build a LAN system in a building, etc., the LAN cable and network equipment necessary for communication must be equipped. In the normal construction stage, LAN cable installation work is first done, and once this work is completed, Each network device is connected to the LAN cable and installed.

Once the LAN cable installation and connection work is completed, technicians first check whether the cable is properly installed and connected without any problems, and then install the network device. In other words, technicians connect LAN cable checkers to both ends of the cable to check whether the LAN cable is operating normally and check for abnormalities.

Next, once the LAN cable is confirmed, each network device is connected to the LAN cable as previously designed, the equipment parameters are set, and the operation of the network as a whole is checked. All of these construction steps must be completed before the overall LAN system can operate normally.

As multiple LAN cables are connected to network devices, it takes a lot of time to identify the lines, line numbers, disconnections, short circuits, etc. of the LAN cables during initial installation or maintenance of network devices and subscriber-side communication equipment. It is becoming.

In addition, even after installation is complete, there are cases where communication does not occur properly or abnormal operation occurs. In this case, in order to check for abnormalities in the LAN cable, two communication workers must connect a LAN cable checker to both ends of the LAN cable. However, if the distance between communication workers is large, there is the inconvenience of having to use separate communication equipment to communicate to check for cable abnormalities.

In addition, there was an inconvenience in that one communication worker was placed at the terminal and one person at the IDF had to manually check the line number, and one communication worker was placed at each IDF and MDF, so there was the inconvenience of having to manually check and record each line.

In particular, in government offices, cable quality is inspected once for the first time, but there was the inconvenience of having to connect both ends of the 100 to 800 lines in the building one by one to test the quality.

Therefore, there is an urgent need to resolve physical line failures that occur when lines are changed due to organizational changes in government offices, etc.

In addition, stable line management for public health centers and water supply system operation systems is necessary, and it is necessary to establish a sustainable control system for incomplete network management that is completely dependent on maintenance companies.

Korean Patent Registration No. 2527242 Korean Patent Registration No. 2428890 Korean Patent Registration No. 2430094

The present invention was devised to improve the above-mentioned problems, and the present invention enables network cable quality monitoring that can be monitored on a regular basis by quickly checking the quality of the network cable using AI technology when building or changing the system of the on-premise network and telephone infrastructure. The purpose is to provide an AI integrated wiring board.

In addition, the purpose of the present invention is to provide an intelligent integrated wiring board using AI technology and an AI integrated wiring board capable of monitoring the quality of network cables that can monitor the quality of each section through remote control and periodic monitoring.

In addition, according to an embodiment of the present invention, the purpose is to provide an AI integrated wiring board capable of monitoring network cable quality by continuously monitoring network quality with AI technology and using collected data to prevent potential network failures in advance. .

In order to achieve the above object, an embodiment of the present invention includes an intelligent integrated wiring board for establishing a stable communication infrastructure and preventing network failures; And a control system connected to the intelligent integrated wiring board and a network to enable quality monitoring for each section through remote control and periodic monitoring, wherein the intelligent integrated wiring board detects the normal connection rate of the LAN connector of the intelligent integrated wiring board. Micro-resistance measurement module to prevent computer failure; A frequency-based measurement module for measuring network cable quality of the intelligent integrated distribution board; A line abnormality diagnosis module that collects resistance data by the resistance data collection module of the intelligent integrated distribution board; and a failure prediction module of a terminal device by analyzing power consumption data of the POE line of the intelligent integrated distribution board.

The grounding strength measures the LAN connector connection rate and displays it for on-site confirmation or sends it by email, and the failure prediction module collects and analyzes resistance data for the POE line and provides information for each terminal device through the MAC address of the terminal device. Calculate the failure rate.

The control system includes a resistance data collection software application for collecting resistance data of the RJ45 LAN connector of the microresistance measurement module; Cable frequency measurement software (Software) application for checking disconnection to measure network cable quality of the frequency-based measurement module; A delay rate/loss rate detection software application for detecting an abnormal delay rate or loss rate of the line abnormality diagnosis module; It includes a failure prediction power measurement software application for predicting failure of the failure prediction module.

The intelligent integrated wiring board measures the connection rate for each LAN pair (8 strands) and transmits an alarm control command transmission signal to the control system to alert the failure factor in advance.

The intelligent integrated distribution board further includes a POE support module mounted on the integrated distribution board, to which PoE (Power over Ethernet) is applied and capable of detecting abnormal symptoms using the POE power consumption data set.

Each integrated distribution panel of the above intelligent integrated distribution panel acts as an RX/TX gateway, enabling measurement through the RX/TX sensor for line electromagnetic delay measurement for each section or the RX/TX sensor for line resistance measurement, and can also be used by complex path construction organizations.

In order to detect the quality of the network cables installed in the building and the connected equipment of the intelligent integrated distribution board, it is possible to detect line defects for each network cable.

In order to detect the location of disconnection in the intelligent integrated distribution board in advance, it is possible to automatically generate quantitative data and analysis result reports on line quality through a line loss dB measurement sensor.

The control system is used to identify and classify abnormal or abnormal events from sensors, and supports anomaly type analysis and push alarm edge designed to detect and respond to anomalies in real time from the quantitative data of each identified and classified sensor. connected to the module;

In order for the AI module to detect the quality of network cables installed in linked equipment and buildings, line defect detection data for each network cable is analyzed using deep learning recognition technology, and to detect line defects in network cables. Data collection process to collect a sufficient amount of training data, line fault data is network cable image or signal data in normal and faulty states, and must include various line fault types to ensure diversity for accurate analysis; A data preprocessing process that removes noise from the collected data and converts it into an appropriate format. If it is graph image data, resize it or apply rotation and movement, and convert the data into a form suitable for learning so that the model can understand it; We design a deep learning model to be used for line fault detection and use a convolutional neural network (CNN), which is specialized for image or time series data processing, and build a network by setting the model's layer configuration and hyperparameters. Deep learning model design process; The data learning process begins by using preprocessed data for the constructed deep learning model and optimizes the model to learn how to distinguish between normal and defective states. The training data is generally divided into training, validation, and test sets. ; Once training is complete, apply the model to test data to evaluate accuracy and performance. If the performance of the model is satisfactory, a model evaluation process that uses it to perform line fault detection on real data; It includes a real-time application process that detects network cable defects by applying the trained deep learning model to a real-time environment, thereby enabling real-time network cable quality monitoring in an intelligent integrated distribution board.

The AI module enables automatic generation of quantitative data and an analysis result report on line quality through a line loss dB measurement sensor to detect the location of a disconnection in the intelligent integrated distribution board in advance, and the analysis result report includes information on each network cable. Line fault detection result information that provides line fault information detected through deep learning technology; Line quality measurement result information, which quantifies the quality of network cables and provides dB measurements; Network cable location information that indicates the location of each network cable and the disconnection location on the wiring board, allowing confirmation at each point; Time-specific data information that records changes in network cable quality over time and allows you to check trends; Abnormal pattern detection information that uses deep learning recognition technology to detect abnormal patterns or abnormal signs and report unusual events; Recommended action information, which provides recommended actions to resolve the problem if a line fault or deterioration in quality is detected; Graphs and visualization information provided in the form of graphs and charts to visually represent data for easy understanding; Include additional statistics or relevant information as needed.

The control system, which is connected to the AI module and a network, is used to identify and classify abnormal or abnormal events from sensors, and is designed to detect and respond to abnormalities in real time from the quantitative data of each identified and classified sensor. Connected to the anomaly type analysis support and push alarm edge module; the anomaly type analysis support and push alarm edge module collects data from several sensors through the control system, and the sensors are used to monitor network cable quality and line loss A sensor data collection unit that includes not only dB measurement sensors but also sensors that detect various environmental factors, allowing the sensors to transmit data to the control system in real time; Collected sensor data identifies and classifies events through an AI module. The AI module uses deep learning recognition technology to analyze data, distinguish between normal and ideal states, and include an event identification and classification unit that detects network cable line defects and environmental changes; An anomaly detection unit that classifies events and detects anomalies in real time based on quantitative data from each sensor, and determines the event as an anomaly when it detects data outside the normal range or an abnormal pattern; When an abnormality is detected, the control system performs an abnormality type analysis, analyzes specifically what type of abnormality occurred, derives a response plan appropriate for each type, classifies the type of line defect, and determines the indoor temperature. An abnormality type analysis unit that classifies abnormalities in environmental factors such as humidity; And when an abnormality is detected, the control system immediately sends a push alarm to the manager or person in charge. The push alarm includes information about the abnormality and details about the event, enabling real-time response, and the manager can immediately respond. It includes a push alarm unit that can take action against abnormal phenomena, thereby minimizing potential failures or problems and maintaining system stability.

According to an embodiment of the present invention, AI technology can detect abnormal delay rates and loss rates of individual server equipment to prevent network cable quality and network failures in advance.

In addition, according to an embodiment of the present invention, AI technology makes it possible to detect the quality of network cables installed in buildings as well as connected equipment, detect line defects, and detect the location of disconnections.

In addition, the present invention systematizes quality measurement using AI technology, enabling regular measurement without additional costs, thereby reducing labor costs and shortening the time required.

In addition, the present invention uses AI technology to automatically generate and easily check quantitative data and analysis result reports on line quality.

In addition, the present invention uses AI technology to enable automatic quality upgrades and alarms regardless of line volume.

In addition, the present invention enables automatic generation and easy confirmation of quantitative data and analysis result reports on line quality through AI technology, and automatic quality improvement is possible regardless of line volume according to the analysis result report.

Figure 1 is a block diagram schematically showing the relationship with the goal of implementing an AI integrated wiring board capable of monitoring network cable quality according to an embodiment of the present invention.
Figure 2 is a configuration diagram of an AI integrated wiring board capable of monitoring network cable quality according to an embodiment of the present invention.
Figure 3 is a detailed configuration diagram of an AI integrated wiring board capable of monitoring network cable quality according to an embodiment of the present invention.
Figure 4 is a detailed configuration diagram of the control system of an AI integrated wiring board capable of monitoring network cable quality according to an embodiment of the present invention.
Figure 5 is a diagram comparing a deep learning method according to an embodiment of the present invention with other methods.
Figure 6 is a diagram showing converting waveform data into OCR and extracting it using the CNN method according to an embodiment of the present invention.
Figure 7 is a diagram showing the image analysis method of Figure 6.
Figure 8 is a diagram showing a learning step and a testing step according to an embodiment of the present invention.

The present invention as described above will be described in detail through the attached drawings and examples.

It should be noted that the technical terms used in the present invention are only used to describe specific embodiments and are not intended to limit the present invention. In addition, the technical terms used in the present invention, unless specifically defined in a different sense in the present invention, should be interpreted as meanings generally understood by those skilled in the art in the technical field to which the present invention pertains, and are not overly comprehensive. It should not be interpreted in a literal or excessively reduced sense. Additionally, if the technical term used in the present invention is an incorrect technical term that does not accurately express the idea of the present invention, it should be replaced with a technical term that can be correctly understood by a person skilled in the art. In addition, general terms used in the present invention should be interpreted according to the definition in the dictionary or according to the context, and should not be interpreted in an excessively reduced sense.

Additionally, as used in the present invention, singular expressions include plural expressions unless the context clearly dictates otherwise. In the present invention, terms such as “consists of” or “comprises” should not be construed as necessarily including all of the various components or steps described in the invention, and some of the components or steps are included. It may not be possible, or it should be interpreted as including additional components or steps.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numbers regardless of the reference numerals, and duplicate descriptions thereof will be omitted.

Additionally, when describing the present invention, if it is determined that a detailed description of related known technologies may obscure the gist of the present invention, the detailed description will be omitted. In addition, it should be noted that the attached drawings are only intended to facilitate easy understanding of the spirit of the present invention, and should not be construed as limiting the spirit of the present invention by the attached drawings.

The present invention is to quickly check the quality of network cables when building or changing the system of network and telephone infrastructure in administrative buildings, complex buildings, control centers, etc., and to monitor the quality of network cables that can be monitored regularly, so as to build a stable communication infrastructure and network. Intelligent integrated wiring board (20) to prevent failures in advance; and a control system 10 connected to a network to enable quality monitoring for each section through remote control and periodic monitoring with the intelligent integrated distribution panel.

The intelligent integrated wiring board 20 includes a microresistance measurement module for preventing computer failure by detecting a normal connection rate of the LAN connector of the intelligent integrated wiring board; A frequency-based measurement module for measuring network cable quality of the intelligent integrated distribution board; A line abnormality diagnosis module that collects resistance data of the intelligent integrated distribution board; and a failure prediction module of the terminal device by analyzing power consumption data of the POE (Power over Ethernet) line of the intelligent integrated distribution board. Here, each module plays a key role in detecting abnormal conditions, which will be described later.

The grounding strength measures the LAN connector connection rate and displays it for on-site confirmation or sends it by email, and the failure prediction module collects and analyzes resistance data for the POE line and provides information for each terminal device through the MAC address of the terminal device. Calculate the failure rate.

Specifically, as shown in FIG. 3, if described in detail with reference to the microresistance measurement module 11, the frequency-based measurement module 12, the line abnormality diagnosis module 13, and the failure prediction module 14, the microresistance The measurement module 11 is a module for detecting the connection rate of the LAN connector of the intelligent integrated distribution board.

If the LAN connector is not connected properly, the connection with communication equipment may become unstable, causing computer failure.

The micro-resistance measurement module 11 measures the micro-resistance connected to the LAN connector and determines the connection state of the LAN connector.

The frequency-based measurement module 12 is a module for measuring network cable quality of an intelligent integrated distribution board.

The frequency-based measurement module 12 measures the quality of the cable using frequency and determines the condition of the cable based on the measured data.

The line abnormality diagnosis module 13 is a module for diagnosing abnormal line conditions.

For example, the line abnormality diagnosis module 13 collects voltage and current values applied to the line and calculates the resistance value of the line.

In other words, the line abnormality diagnosis module 13 diagnoses the abnormal state of the line through this and enables quick response.

As another embodiment, the line abnormality diagnosis module 13 may operate as a module that predicts a malfunction of a terminal device by analyzing power consumption data of the POE line.

The line abnormality diagnosis module 13 analyzes power data transmitted through the POE line, predicts failure of the terminal device through the failure prediction module 14, and can prevent network failures due to failure of the terminal device in advance. there is.

Among the display contents of the administrator terminal, the disconnection location and the induced power noise location are redisplayed through 3D modeling and 3D mapping, and the disconnection point and fault location are displayed through the 3D mapping.

In the intelligent integrated distribution panel, monitoring and alarming can be performed by calculating the network abnormality delay rate or data abnormality packet transmission/reception rate of the central intelligent integrated distribution panel, the intelligent integrated distribution panel for each floor, and the outlet section within the premises.

As an embodiment, the present invention is an intelligent integration that includes a frequency-based measurement module for measuring network cable quality, a line abnormality diagnosis module through resistance data collection, and a terminal device failure prediction module by analyzing power consumption data of the POE line. The wiring board is designed to prevent network failures in advance, and includes a micro-resistance measurement module to prevent computer failures by detecting the normal connection rate of the LAN connector. Quality monitoring for each section is possible through remote control and periodic monitoring with the integrated wiring board. A network-connected control system is also built to do so.

Therefore, the present invention can prevent computer failures by detecting the connection rate of the LAN connector of the intelligent integrated wiring board through the microresistance measurement module as shown below.

(1) The quality of network cables can be measured using a frequency-based measurement module, contributing to the establishment of a stable communication infrastructure.

(2) By collecting resistance data through the resistance data collection module, the line abnormality diagnosis module can diagnose the abnormal state of the line, enabling quick response to failures.

(3) By analyzing the power consumption data of the POE line, failure of the terminal device can be predicted, allowing for proactive response.

Through these functions, network failures can be prevented in advance and a stable communication infrastructure can be established.

As shown in Figure 4, the present invention consists of resistance data collection software 15, cable frequency measurement software 16, delay rate/loss rate detection software 17, and failure prediction power measurement software 18.

In the control system 10 according to one embodiment, resistance data collection software 15, cable frequency measurement software 16, delay rate/loss rate detection software 17, and failure prediction power measurement software 18 each have the fine Resistance data collection software (Software) application for collecting resistance data of the RJ45 LAN connector of the resistance measurement module; Cable frequency measurement software (Software) application for checking disconnection to measure network cable quality of the frequency-based measurement module; A delay rate/loss rate detection software application for detecting an abnormal delay rate or loss rate of the line abnormality diagnosis module; It includes a failure prediction power measurement software application for predicting failure of the failure prediction module.

In addition, it includes a fastening detection patch panel (hardware) that is displayed and connected to detect the normal fastening rate of the LAN connector of the intelligent integrated distribution board 20.

The intelligent integrated distribution board according to one embodiment measures the connection rate for each LAN pair (8 strands) and transmits an alarm control command transmission signal to the control system to alarm the failure factor in advance.

Therefore, the intelligent integrated distribution board can detect network connection problems in advance.

The simple alarm of the conventional invention sounded a warning only when the network connection was interrupted, but in the present invention, by measuring the connection rate of each LAN cable, a warning can be sounded in advance before a problem occurs.

With this change, even if a network failure occurs, you can be warned in advance and take action, saving money and time related to the time the problem occurs. In addition, since problems such as business interruption or data loss due to problems can be prevented, the stability and reliability of the system can be increased.

Additionally, since this new system measures the connection rate of each LAN cable, it is easy to determine which LAN cable is having a problem. Therefore, a quick response is possible when a problem occurs, which can shorten the problem resolution time.

The intelligent integrated distribution board according to one embodiment further includes a POE support module mounted on the integrated distribution board to which PoE is applied and capable of detecting abnormal symptoms using the POE power consumption data set.

Therefore, the PoE support module is used to prevent computer failures by detecting the normal connection rate of the LAN connector to detect and prevent network connection problems. This module is electrically connected to the micro-resistance measurement module, frequency-based measurement module, and line abnormality diagnosis module, and is an effective tool for preventing network failures by interconnecting them, contributing to increasing the safety and reliability of the intelligent integrated distribution board.

According to one embodiment, each integrated distribution panel of the intelligent integrated distribution panel performs the role of an RX/TX gateway, enabling measurement through a line propagation delay measurement RX/TX sensor or line resistance measurement RX/TX sensor for each section, and a complex path construction agency. It can also be used in

Therefore, the intelligent integrated wiring board detects the normal connection rate of the LAN connector to prevent computer failures, and can measure line propagation delay for each section through the line propagation delay measurement RX/TX sensor or the line resistance measurement RX/TX sensor. , It can also be used by complex path construction organizations.

These functions play a significant role in preventing and responding to network failures, and by being built together with the control system 10, they contribute to increasing the efficiency of the control system for periodic monitoring.

In order to detect the quality of the network cables installed in the building and the linked equipment of the intelligent integrated distribution board according to one embodiment, it is possible to detect line defects for each network cable.

In one embodiment, a fine resistance measurement module, a frequency-based measurement module, and a circuit abnormality diagnosis module are further included in addition to a line fault detection module capable of detecting line faults for network cables, and are used as an effective tool for preventing network failures. By doing this, time and costs due to failures can be saved.

In one embodiment, the connection rate of each LAN cable can be measured, so it is possible to easily determine which LAN cable is causing the problem. This allows a quick response when a problem occurs, thereby reducing troubleshooting time.

In addition, the line fault detection module includes a circuit abnormality diagnosis module, so it can detect abnormalities in real time and respond quickly.

The function of this line fault detection module is an important element that increases the efficiency of the safety monitoring system, and is connected to a networked control system to enable periodic monitoring and remote control by zone, thereby enabling quality monitoring.

This contributes to minimizing the possibility of safety accidents and increasing the efficiency of the safety monitoring system.

Therefore, it detects the normal connection rate of network cables to prevent computer failures, and uses RX/TX sensors to measure propagation delay or resistance in zoned circuits, allowing organizations with complex path configurations to prevent and respond to network failures. , These functions can be combined with the control system to enable periodic monitoring and remote control, contributing to increasing the efficiency of the safety monitoring system.

According to one embodiment, it is possible to automatically generate quantitative data on line quality and an analysis result report through a line loss dB measurement sensor to detect the location of a disconnection in the intelligent integrated distribution board in advance.

The control system according to one embodiment is used to identify and classify abnormalities or abnormal events from sensors, and analyzes abnormality types designed to detect and respond to abnormalities in real time from quantitative data of each identified and classified sensor. Supports and connects to push alarm edge module;

Therefore, it is possible to monitor quantitative data from each sensor in real time and quickly respond to abnormal phenomena before they cause serious damage or disruption.

As an embodiment, the present invention includes an intelligent integrated wiring board (20) connected to a control system for cloud-based periodic monitoring and an optical network, and is connected to an abnormality information collection device through the intelligent integrated wiring board (20).

Through this, monitoring of the real-time intelligent integrated wiring board (20), deep learning-based anomaly detection, information security transmission, real-time anomaly monitoring, anomaly monitoring decision support, safety level indexing monitoring, scenario-based emergency response, big data-based risk prediction, real-time It consists of a control system for cloud-based periodic monitoring that provides the functions of the intelligent integrated distribution board (20) to integrated surveillance monitoring, deep learning prediction system, and information comprehensive system.

The present invention includes a risk assessment module for applying environmental conditions that collects and evaluates abnormal information of an intelligent integrated wiring board, a module for deriving failure and damage accident scenarios by receiving real-time safety-related information, a module for presenting safety measure standards for decision-making support, and a safety measure standard draft. The presentation module includes a system improvement module that provides system improvement scenarios through data including safety measure standards. In addition to this, additional sensors connected to the intelligent integrated distribution panel 20 may be included.

Therefore, the present invention enables automatic generation and easy confirmation of quantitative data and analysis result reports on line quality, and enables automatic quality improvement by system improvement scenarios regardless of line volume according to the analysis result report.

The present invention consists of a cloud-based section-specific quality monitoring and integrated safety monitoring system that includes a control system 10 for cloud-based periodic monitoring.

Through this, it is possible to simulate risk factors that may occur in an intelligent integrated distribution board by analyzing risk factors based on scenarios to predict and prepare for them in advance, and to cope with them in advance through scenarios.

The present invention monitors abnormal information of the intelligent integrated wiring board in real time through a cloud-based control system for periodic monitoring and an intelligent integrated wiring board (20) to predict and prepare for risks that may occur in the intelligent integrated wiring board in advance. It's purpose.

To this end, the present invention collects data using various sensors and surveillance equipment, and analyzes and visualizes it through a cloud-based integrated surveillance and monitoring system.

Through this, risk situations that may occur in the intelligent integrated distribution panel 20 can be predicted in advance, a response plan can be established, and risks can be minimized.

In addition, the present invention provides risk prediction and abnormality detection functions by utilizing big data technology and deep learning technology based on data collected through a control system for periodic monitoring, thereby strengthening safety measures and improving the safety level. Supports safe operation through exponential monitoring.

The present invention also collects abnormal information using local monitoring equipment connected to a network, such as an intelligent integrated wiring board, and analyzes and visualizes this in connection with a control system for periodic monitoring through the intelligent integrated wiring board 20.

In addition, the present invention supports access to the control system for periodic monitoring through various terminals such as a terminal connected to the control system 10 through a network, transmits data through the network, and links with various linked organizations. Support.

Lastly, the present invention can collect and evaluate safety-related real-time information that collects and evaluates abnormal information of the intelligent integrated distribution board and transmit it to the control system 10 to derive failure and damage accident scenarios.

The cloud-based section-specific quality monitoring and integrated safety monitoring system 30 also provides functions for detecting and quickly responding to various abnormal conditions that occur while operating the intelligent integrated distribution board 20.

To this end, sensors connected to the intelligent integrated distribution panel 20 collect information such as the normal connection rate of the LAN connector in real time, and through this, detect any abnormalities using a deep learning-based abnormality detection algorithm. When an abnormality is detected, a warning message is immediately sent to the control system 10 for periodic monitoring so that appropriate response measures can be taken.

In addition, the present invention also provides a function to collect abnormal state data occurring in the intelligent integrated distribution panel 20, analyze it, and predict it.

To this end, the control system 10 for the cloud-based periodic monitoring can utilize a big data-based risk prediction model to predict risk factors that may occur in the intelligent integrated distribution board 20 in advance and establish response plans. Let it happen.

In addition, the data collected from the intelligent integrated wiring board (20) is visualized based on GIS to collect all the safety situations of the intelligent integrated wiring board (20) in each building and identify them in real time, thereby indexing the safety level and establishing safety measures. Supports.

Therefore, the present invention provides a cloud-based section-specific quality monitoring and integrated safety monitoring system (30) that can predict and respond to various abnormalities that may occur while operating the intelligent integrated wiring board (20), thereby providing a safe intelligent integrated wiring board (20). (20) Supports operation and provides functions to minimize abnormal elements of the intelligent integrated distribution panel (20).

Example 1

A POE support module mounted on an integrated distribution board capable of detecting abnormal symptoms using a POE power consumption data set including POE power consumption data according to an embodiment of the present invention is capable of detecting abnormal phenomena occurring in a POE device, It ensures stable operation and can extend the life of the equipment.

The POE support module uses the POE power consumption data set to detect abnormalities. In POE devices, power supply and network connection occur simultaneously, so if power supply is not smooth, the network connection is also affected. This module analyzes POE power consumption data and detects any abnormality in power supply. By taking appropriate action against abnormalities detected in this way, stable operation of the equipment can be ensured and its lifespan can be extended.

Additionally, the POE support module is mounted on an integrated distribution board, making it easy to use. Since it is mounted on an integrated distribution board, management and maintenance of the POE device becomes easy. Additionally, this module expands the range of support for POE devices, allowing more devices to be used. This module, which increases the stability and reliability of POE devices, will become an important driving force in the development of POE technology.

The POE power consumption dataset is data that measures the amount of power consumed by a POE device and is used to increase the stability of the POE device. In POE devices, power supply and network connection occur simultaneously, so smooth power supply is important.

At this time, the POE power consumption data set can be used to detect if there is an abnormality in power supply and take appropriate measures to ensure stable operation of the equipment and extend the life of the equipment.

Here, the POE power consumption dataset accurately measures the amount of power consumed by POE equipment and can detect abnormalities. Detected abnormalities can be taken to ensure stable operation of equipment and extend its lifespan by taking appropriate measures.

If lightning or impulses are introduced through induction or grounding, malfunction or damage to the PoE device can be minimized, and additional impulse surge protection and immunity circuits can be added to ensure safe operation and reliable communication quality of the PoE system.

An additionally installed AI-based control board can be further improved by adding overheating prevention and remote monitoring functions.

These features improve the safety and reliability of POE-enabled modules and help maintain high-quality communications.

In another embodiment, additional shock wave protection and recovery circuitry may be added to the PoE support module to minimize damage from shock waves that may occur through lightning, induction, or grounding.

The temperature of the AI-based control board can be monitored in real time to prevent overheating, and remote monitoring and management functions can be added to increase convenience.

Therefore, a more accurate and reliable monitoring system can be built using an AI-based control board.

These improvements can improve the performance and durability of POE-enabled modules, ensuring safe and reliable operation.

Additionally, the safety and reliability of the system can be further improved by implementing a fail-safe module (fail-safe mechanism) that automatically shuts down the POE support module when overheating.

This can be implemented through the additional use of temperature sensors and intelligent control systems that monitor temperature and shut down the system in case of overheating.

Additionally, the convenience and accessibility of the system can be greatly improved by implementing a remote management system that allows users to monitor and control the system from a smartphone or computer.

This can be achieved by using IoT technology and user-friendly interfaces to allow users to access and control the system.

Example 2

First, when external power is applied, the intelligent integrated distribution panel 20 connected to the control system 10 operates.

And, the control system 10 checks the operating status of the intelligent integrated distribution panel 20 through frequency or voltage monitoring.

For example, the control system 10 compares the frequency or voltage measurement value of the intelligent integrated distribution panel 20 with a reference value to check the operating state.

Finally, when a failure of the intelligent integrated wiring board 20 is detected, the control system 10 generates an alarm in the monitoring and control device 244, transmits an alarm signal to each intelligent integrated wiring board 20, and sends an alarm signal to the user terminal 144. ) Controls the notification to sound.

At this time, it can be shown that the wiring board operates, monitors the quality status, receives a control signal through a communication protocol, the internal wiring board operates to check the quality status, and an alarm sounds when an abnormality in the wiring board is detected.

Additionally, the intelligent integrated distribution board 20 receives control signals through a communication protocol while monitoring the operating status through the fault prediction module 14.

The failure prediction module 14 checks the operating status of the intelligent integrated distribution board 20, compares frequency or voltage measurements to see if the intelligent integrated distribution board 20 is operating normally, and sounds an alarm when a failure is detected.

In one embodiment, the present invention includes the steps of determining, by the failure prediction module 14, the status information of the intelligent integrated distribution board 20 and the priority and type of the corresponding failure alarm alert; and differentially generating, by the failure prediction module 14, an alarm alert that stimulates at least one of the senses of sight, hearing, and touch based on the priority and type of the alarm alert.

As an example, the failure prediction module 14 determines the priority of the failure alarm and the type of the alarm based on the current state information received from the intelligent integrated distribution board 20 and the corresponding response scenario; and differentially generating, by the failure prediction module 14, an alarm alert that stimulates at least one of the senses of sight, hearing, and touch based on the priority, type, and response scenario of the alarm alert. .

When the failure prediction module 14 utilizes this scenario, there are the following advantages.

Failure of the intelligent integrated distribution board (20) can be quickly and accurately detected according to a certain scenario and appropriate measures can be taken.

Additionally, the user can be notified in various ways depending on the priority and type of the alarm.

In addition, the user's attention can be further focused through an alarm alert that optimally stimulates other senses, such as sight, hearing, and touch, depending on the scenario.

As an example, the failure prediction module 14 can detect and analyze the power supply status, temperature, vibration, etc. of the intelligent integrated distribution board 20 in real time.

Additionally, the failure prediction module 14 predicts the possibility of failure of the intelligent integrated distribution board 20 and sets a response scenario accordingly. For example, if the power supply condition is unstable, measures such as replacing the power supply or using a backup power source can be taken.

In addition, the failure prediction module 14 determines the priority of the alarm according to the risk of failure of the intelligent integrated distribution board 20. For example, if the temperature rises above a certain level, an urgent alarm can be triggered first.

Additionally, the failure prediction module 14 determines the type of alarm alert. For example, visual alarms include blinking or color changes in an LED lamp, auditory alarms include buzzers or speakers, and tactile alarms include vibration or shock.

Additionally, the failure prediction module 14 differentially generates an alarm that stimulates at least one of the senses of sight, hearing, and touch based on the priority, type, and response scenario of the alarm.

For example, an urgent alarm alert can be generated as an alarm alert that stimulates visual, auditory, and tactile senses simultaneously, and a general alarm alert can be generated as an alarm alert that stimulates only either visual or auditory senses.

Hereinafter, the present invention describes a method for monitoring network cable quality using an intelligent integrated distribution board.

For example, when external power is applied, the wiring board operates, monitors the quality status and receives a control signal through a communication protocol, the internal wiring board operates to check the quality status, and when an abnormality in the wiring board is detected, an alarm sounds. do.

In addition, the failure prediction module can predict failures in the intelligent integrated distribution board and notify users in various ways depending on the priority and type of alarm.

Additionally, an alarm message notifying a dangerous situation is transmitted to the wearable device of a nearby co-worker and the control system 10, and is controlled to be displayed on the wearable device of the co-worker and the control system 10.

Example 3

Among the display contents, the disconnection location and the induced power noise location are displayed by 3D mapping on a screen with a 3D model of the intelligent integrated wiring board, and the expected disconnection point and fault location are displayed through the 3D mapping.

Augmented reality (AR) used in the present invention is a technology that displays virtual images overlaid on the real world. Using augmented reality, the location of disconnections and inductive power noise are mapped and displayed on a screen with a 3D model of an intelligent integrated wiring board. Additionally, 3D mapping can display expected disconnection points and fault locations.

For example, if a 3D model of an electrical facility is displayed on an augmented reality device and then the 3D mapping of the electrical facility is input into the augmented reality device, the augmented reality device can display the expected disconnection point and fault location in the 3D mapping.

Meanwhile, as shown in FIGS. 5 to 8, using AI techniques, failures in electrical equipment can be diagnosed more quickly and accurately.

In the present invention, we will use a deep learning method that can analyze and predict data information of complex cables as efficiently as possible (see Figure 5).

Specifically, deep learning and CNN (Convolutional Neural Networks) are very effective technologies for image recognition and classification. They learn the image of the 3D model of an intelligent integrated wiring board, and use the learned model to recognize and display disconnection points and fault locations. (see Figure 6).

In addition, reinforcement learning is a machine learning technology that learns how to maximize rewards by interacting with the environment, and can be used to learn the optimal policy to find expected disconnection points and fault locations (see Figure 8).

In particular, 3D models that have been point cloud processed are generally expressed as point clouds. Point cloud processing techniques can be used to analyze these data and estimate probable break points and fault locations.

Example 4

In measuring the LAN connector connection rate according to an embodiment of the present invention and displaying it for on-site confirmation or sending it by email, an alarm is set based on the cable data predicted value, or when an alarm is received, the predicted value and the actual value are divided. When checking the cable quality utilization rate by comparing, repeating cable quality utilization prediction and alarm settings, or sending an alarm to the administrator, the alarm or forecast/notification can be activated by operating the system even when the alarm or forecast/notification is over or in progress. The stability of the system can be increased by allowing the prediction process to be extended in real time without interruption.

Hereinafter, the process of searching for patterns related to the usage period of the cable connected to the intelligent integrated distribution board 20 will be described in detail in chronological order.

(1) The cable data collection module collects cable data related to the cable’s usage period to date. The cable data includes information such as cable operating time, stopping time, and operating status.

(2) The cable data preprocessing module preprocesses the collected cable data. In this process, cable data is sorted, missing values are processed, and outliers are removed.

Here, the missing value calculated through the intelligent integrated distribution panel 20 means that data is missing, and the outlier value means that the data is outside the normal range.

(3) The cable data analysis module is a process of analyzing preprocessed cable data, and during this process, statistical analysis, time series analysis, and pattern analysis of the cable's usage period are performed.

Here, statistical analysis through the intelligent integrated distribution panel 20 refers to the process of identifying the distribution and characteristics of data, time series analysis refers to the process of identifying change trends in data, and pattern analysis refers to regular patterns in data. It means the process of finding .

(4) The result summary module is a module that organizes the analysis results. In this process, the analysis results are summarized and visualized for easy understanding.

The above-described applications are AI (or machine learning modules, etc.) functions included in the intelligent integrated distribution board 20, and are modules that find correlations through analysis results and proceed with learning. Here, you can find patterns related to the failure rate of each terminal device according to cable use through machine learning algorithms (ex. SVM, KNN, Neural Network).

Once learning is completed through the above-mentioned modules, the expected usage period can be quickly predicted through the learning model by inputting new cable data.

Therefore, through the present invention, users can achieve better maintenance management and high work efficiency through real-time monitoring.

In particular, the AI algorithm for cables of the present invention can be applied to a wide range of applications because it can learn other types of cables in the same way.

Additionally, the algorithm of the present invention can easily predict repeated cable quality or utilization through just one learning.

Therefore, the present invention helps to timely recognize, analyze, and predict changes in failure rates for each cable terminal device through AI technology.

Meanwhile, when collecting and storing time series cable data of up to the plurality of cables, the cable data of each sensor may be analyzed using a simple average value or an exponentially weighted moving average.

In order for the intelligent integrated distribution panel 20 to collect and store time series cable data for multiple cables using a simple average value,

(1) Collect time series cable data of multiple cables with the intelligent integrated distribution board (20). This cable data is recorded by date or time zone.

(2) Create a database table to store the collected cable data. Here, the table has time zone, cable ID, sensor ID, sensor value, and average value columns, and is designed so that it can be recorded in the DB.

(3) Calculate the average value using the values of each sensor.

(4) Save the calculated average value, time zone, and values for multiple cables in the table.

(5) Perform visualization or analysis using the calculated average value.

By using simple average values like this, time series cable data can be accurately viewed/analyzed/visualized.

The AI function of the intelligent integrated wiring board (20) is based on the cables of the microresistance measurement module (11), frequency-based measurement module (12), line abnormality diagnosis module (13), and fault prediction module (14) attached to the integrated wiring board. You can explore changes in failure rates by terminal device by measuring/analyzing/diagnosing/predicting time series data.

In this case, using a simple average (or exponentially weighted moving average) gives more weight to recent cable data points, allowing for more accurate measurement, analysis, diagnosis, and prediction. This will better represent current trends in time series cable data.

In addition, the intelligent integrated distribution panel 20 can predict cable quality or blockage by checking linked information such as cable microresistance, frequency, and line abnormality that measure cable quality.

For example, by combining the cable data of the above sensors, the cable data after initial installation and the cable data at 10-day intervals are continuously synthesized and the average value is calculated to check the cable quality/usage rate during the period of use, and based on this, the remaining cable data can be checked in the future. Infer quality utilization rate.

In general, using an exponentially weighted moving average provides better performance than a simple moving average.

The exponentially weighted moving average algorithm can reduce the probability of relative differences appearing when amplifying time series information collection through sensors by smoothly connecting recent cable data points.

Therefore, the exponentially weighted moving average can play an important role in analyzing/evaluating/visualizing the final results of various AI modules connected to the failure prediction module of the AI cable system capable of self-diagnosis according to the present invention.

On the visualization side, using exponentially weighted moving averages can also generate additional information during recent measurements and produce more accurate visualization results.

In other words, exponentially weighted moving averages can be used to more accurately represent current trends in time series cable data and achieve better visualization results.

Better visualization results mean that, for example, when collecting continuous time series information, self-diagnosis is possible when detecting a measurement end state, and a more detailed understanding of the cause of the behavior of the AI system is possible when extracting recent information, such as recent cable data points. It means that it is possible.

In particular, the AI system capable of self-diagnosis based on the automated intelligent integrated distribution panel (20) can display externally during measurement so that continuous time series data patterns can be easily analyzed using accurate and consistent visualization solutions (external display devices, etc.). there is.

For example, exponentially weighted moving averages allow you to more accurately represent trend patterns in cable data, making it easier to identify and analyze changes in cable data over time.

This can therefore provide managers with more meaningful insights and a better understanding of the underlying patterns of cable data.

Intelligent distribution board 20's use of exponentially weighted moving averages also helps identify anomalies or outliers in cable data.

The present invention can infer the remaining usage period by calculating the daily fluctuation value up to the set value through the daily measured value (E).

In addition, the present invention calculates cable microresistance, frequency, line abnormality, etc. for each day and displays the remaining cable usable time on the screen.

The present invention can be used as an app through a mobile terminal (e.g., a smartphone), allowing site name input using the site name, access button, import button, and site name input button, and searching and finding the site name.

After setting the date of the cable data to be imported, the date and time are displayed as information on the import screen, and the cable microresistance, frequency, and line abnormality status are displayed by date.

You can enter or edit the number to send a text message through the SMS transmission number button or the edit/enter button.

Hereinafter, the process of calculating the above-mentioned average value and checking the cable quality usage rate during the period of use of the integrated wiring board will be described step by step.

The intelligent integrated wiring board (20),

(1) After cable installation, cable data for multiple cables is collected and stored.

(2) Collect and store cable data from sensors at intervals of 7 to 10 days.

(3) Data extracted from the cables of sensors (e.g., data through microresistance measurement module (11), frequency-based measurement module (12), line abnormality diagnosis module (13), failure prediction module (14), etc.) Sort in order.

(4) Calculate the average value of the collected cable data using simple average or exponentially weighted moving average.

(5) Based on the calculated average value, check the usage period of the cable and the usage rate while maintaining a certain quality.

(6) Through this, the remaining quality usage rate is inferred.

(7) Through this, it is possible to predict when to replace the cable.

If the simple average value described above is used and the exponentially weighted moving average is used, the present invention can have strengths in time series cable data analysis.

In other words, the simple average value calculates all cable data equally, which is when the weights of recent cable data and old cable data are the same.

On the other hand, the exponentially weighted moving average is calculated by placing greater weight on recent cable data, which reflects the influence of recent cable data to a greater extent when analyzing time series cable data, enabling more accurate predictions.

For example, an exponentially weighted moving average can be useful when giving more weight to more recent cable data than a simple average. This can also be applied to the cable quality utilization rate in cases where recent cable data has a greater impact on the quality utilization rate.

However, because the above-mentioned calculation method is complex and the algorithm configuration for cable data is also complex, the explanatory power may be low.

And in the present invention, the simple average value is simple, easy to understand, and generally usable.

As an example, the step of the intelligent integrated distribution board 20 predicting the cable quality utilization rate using a linear regression method in machine learning using an exponentially weighted moving average will be described in detail.

(1) Cable data collected from sensors is refined and an exponentially weighted moving average is calculated using the refined cable data.

(2) Train a linear regression model using an exponentially weighted moving average as input cable data.

(3) Using the learned model, input a new exponentially weighted moving average to predict the cable quality utilization rate.

(4) Analyze the predicted results to evaluate the cable quality utilization rate.

As another example, the step of predicting the cable quality utilization rate using a linear regression method in machine learning using a simple average will be described in detail.

The method by which the intelligent integrated distribution board 20 predicts cable quality or usage rate using a simple average can be explained in detail as follows.

(1) First, the cable data collected from sensors is refined and only the necessary cable data is extracted.

And (2) measure the cable usage period through the extracted cable data and calculate the average value for the cable usage period.

At this time, the machine learning method of predicting cable quality utilization using a simple average can be performed as follows.

(3) Cable data collected from multiple cable conditions (cable micro-resistance, frequency, line abnormality detection sensor, etc.) are integrated to generate cable quality usage rate cable data for the period from installation.

Then, (4) the generated cable data is used to separate training cable data and test cable data.

As an example, after the intelligent integrated distribution board 20 separates cable training data and cable test data related to the failure rate of each terminal device, the step of predicting the cable quality usage rate in machine learning using the simple average of the present invention is as follows. Same as

(1) Cable data from sensors related to the cable usage period is collected through the cable data collection module 410.

(2) Cable data collected through the cable data preprocessing module 420 is refined to remove outliers and extract only necessary information.

(3) Cable quality usage rate is predicted in machine learning using simple average of cable data preprocessed through the training cable data generation module 430.

As another embodiment, a method in which the intelligent integrated distribution panel 20 generates a simple average on the 1st day to a simple average on the 7th day and uses the exponentially weighted moving average to predict the cable quality usage rate by date will be described.

The exponentially weighted moving average module is,

(1) Cable data by period (e.g., data such as microresistance measurement module (11), frequency-based measurement module (12), line abnormality diagnosis module (13), failure prediction module (14), etc.) or failure rate by terminal device, etc. Collect and find the average.

(2) Predict the cable quality utilization rate by applying an exponentially weighted moving average to the average for each period.

(3) To obtain an exponentially weighted moving average, find the coefficient to apply the weight.

(4) Multiply the average and weight for each period to obtain the weighted average.

(5) The current weighted average is obtained by applying the weight of the weighted average to the previous weighted average.

(6) Modify the forecast value by comparing the average for each period with the current weighted average.

(7) Predicted values are stored by period and expressed in a graph.

(8) Evaluate prediction accuracy by comparing predicted and actual values.

(9) If necessary, modify and re-evaluate the prediction model.

(10) Predict the cable quality utilization rate using the final predicted value.

As another embodiment of the present invention, a method in which the intelligent integrated distribution panel 20 applies artificial intelligence or a transformer module to detect and provide an alarm in cases where it is difficult to predict in advance will be described.

The transformer module is,

(1) Select and learn an artificial intelligence model (ex. LSTM, GRU, Transformer).

(2) Collect cable quality utilization data and apply it as input to the artificial intelligence model.

(3) Predict cable quality utilization rate with the learned model.

(4) Evaluate prediction accuracy by comparing predicted and actual values.

(5) If the prediction accuracy is low, modify the model or apply another artificial intelligence model.

(6) If the prediction accuracy is high, set an alarm based on the predicted value.

(7) When an alarm is received, check the cable quality utilization rate by comparing the predicted value with the actual value.

(8) If the cable quality usage rate is below a certain level, request a cable replacement.

(9) When cable replacement is completed, retrain the artificial intelligence model.

(10) Repeat cable quality usage rate prediction and alarm settings.

As another embodiment, first, we will explain how the intelligent integrated distribution panel 20 applies the transformer method as a transformer module among the AI modules to identify and provide an alarm in cases where it is difficult to predict in advance.

The transformer module is,

(1) Train the artificial intelligence model. (ex: LSTM, CNN, Transformer)

(2) Apply the collected cable data to an artificial intelligence model to predict cable quality utilization rate.

(3) Evaluate prediction accuracy by comparing predicted results with actual values.

(4) If the prediction accuracy is low, modify the model and re-evaluate.

(5) Analyze prediction results to identify cases where prediction is difficult. (ex: When the prediction error is large)

(6) In cases where prediction is difficult, an alarm is provided.

(7) The person who receives the alarm checks the actual value and corrects the prediction result.

(8) Reapply the modified results to the artificial intelligence model to improve the model.

(9) Apply the improved model and recheck the prediction results.

(10) Use the final prediction result to predict the cable quality utilization rate and send it to the manager as an alarm.

Among the AI techniques used here, the Transformer method has the function of receiving cable data and predicting the cable data, so it can be used to predict cable quality. However, accuracy may decrease if you do not prepare enough training data for a specific problem. Therefore, in order to use cable quality prediction, sufficient cable data related to cable quality must be collected and preprocessed.

To this end, we will explain the AI module that controls the intelligent integrated distribution panel 20 to convert cable data related to cable quality (e.g., grounding strength) into big data and apply it.

This is because the better the quality of the cable, the higher the grounding strength, the stronger the wires inside the cable are, and the better the insulation is, which lowers the cable's electrical resistance and increases the grounding strength.

The AI module is,

(1) Collect cable data related to cable quality into big data (ex: cable microresistance, frequency, line abnormality, etc.)

(2) Preprocess the cable data and convert it into a form that can be used for learning and prediction.

(3) Build an AI model using the Transformer method.

(4) Train the model using preprocessed cable data.

(5) Evaluate the trained model of cable data with test data to check prediction accuracy.

(6) If the prediction accuracy is low, modify the model and train again.

(7) Analyze predicted results to identify cases where prediction is difficult. (ex: When the prediction error is large)

(8) In cases where prediction is difficult using only the above cable data, an alarm is notified to the administrator.

(9) The person who receives the alarm checks the actual value and corrects the prediction result.

(10) The revised results are reapplied to the artificial intelligence model and converted into big data (cable data) again.

Here, in step (8), despite training a model to predict cable quality by analyzing cable data, if it is difficult to predict cable quality using the quality prediction model, an alarm is sent to the administrator via SMS. , including those transmitted through app push, etc.

This is because, despite learning a quality prediction model, if quality prediction is difficult, there may be a problem with the model parameters.

In general, if there is insufficient training data or the quality of the training data is low, the model may not be properly trained, making quality prediction difficult.

Additionally, quality prediction may be difficult if the structure, algorithm used, parameter settings, etc. of the cable quality prediction model are not appropriate.

Therefore, before learning the quality prediction model, collect high-quality training data, design the structure of the model appropriately (structure of the cable quality prediction model, algorithm used, parameter settings, etc.), and adjust the algorithm and parameters appropriately over time. It is necessary to choose.

As another example, the AI module of the intelligent integrated distribution board 20 (1) collects self-diagnosis cable data and collects cable data up to a certain date.

(2) Average the cable data over a certain period of time to obtain the average value for each date.

(3) Calculate an exponentially weighted moving average to predict the available usage rate while maintaining a certain cable quality.

(4) Analyze cable quality utilization rate based on the calculated predicted value.

(5) Calculate how to optimize cable quality utilization based on the analysis results.

(6) Implement optimized methods to improve cable quality utilization rate.

(7) Evaluate the improved cable quality usage rate by comparing it with the conventional cable quality.

(8) Modify or repeat the number and method of cable replacement based on the evaluation results.

(9) Continuously improve cable quality utilization through an iterative learning process.

(10) Finally, the results of using the cable quality usage rate as efficiently as possible are transmitted to the administrator terminal.

At this time, the self-diagnosis cable data is used as basic data by the AI module to use the number and method of cable replacement as efficiently as possible.

Meanwhile, when an abnormality outside the normal value range is found in the information of the intelligent integrated distribution board 20 according to the present invention, an exponential moving average (EMA) or exponentially weighted moving average (EWMA) is used as an alarm technique. An alarm is generated only when the change in abnormal values for the cable is outside the normal value range.

Additionally, since the exponentially weighted moving average is more influenced by recent cable data, it places higher weight on the most recent data among sensor signals.

If you look closely,

Here, vt is the weighted exponentially weighted moving average of the t sensor signal data. And the β value is the optimal value as a hyperparameter and is preferably 0.9. Lastly, θt is the value of the tth data among the sensor signals.

First, if we change the value of β, if β is 0.9, vt is almost the same as the average of the previous 10 data.

In another way, if the intelligent integrated distribution panel 20 determines an abnormality outside the normal value range, the percentage area of signal change outside the normal value range of the intelligent integrated distribution panel 20 during the judged time is (EMAx - xi)/EMAx × It is obtained as dt and cumulatively added with the area recorded at the previous time, and the difference in the accumulated signal change rate area obtained by cumulatively summing the signal change rate area (EMAy - yi)/EMAy × dt calculated with the signal value of the intelligent integrated distribution panel 20. is compared with reference data (or big data) and if it differs by more than a certain amount, it is determined whether it is sensor data.

(Where EMA is an exponential moving average; difference in each measurement time)

Therefore, by using an exponential shift calculation formula that multiplies exponentially decreasing weights, the average value can be obtained much more simply without the need for conventional data storage and calculation processing. In terms of hardware, there is no need for a high-performance MCU with a minimum memory capacity, and power consumption is reduced with a small number of calculations. Consumption is low.

Another method, Exponential Moving Average, is that the weight for older data decreases exponentially but never goes to zero.

As shown in Equation 2 below, unlike the weighted moving average (a) for each sensor signal data Y, the exponentially weighted moving average (b) is calculated exponentially.

(Here, coefficient α is the weight reduction degree between 0 and 1, Y t is the value in period t, and S t is the exponential moving average value in any period t)

Example 5

The AI data conversion unit, which converts data such as cable microresistance and frequency, captures images of the graph representing each data and converts the operation form for each timeline into text (numeric) data.

In the step of converting the graph into text data, image processing technology is used to collect operational text data of each system.

For this purpose, an AI algorithm consisting of the steps of capturing the above images and converting the operational form by timeline into text data can be used to detect early signs of system failure and take corrective action before failure occurs.

Multimodal AI, which includes these algorithms, is a technology that considers various data modalities such as text, images, video, and voice to learn and express their relationships. Therefore, multimodal AI can perform a variety of tasks rather than utilizing just one modality.

Multimodal artificial intelligence is a technology that extracts information by processing various types of data. For example, images can be processed using image analysis technology that includes numerical information such as graphs, and text can be understood, analyzed, and processed using natural language processing technology (e.g., GPT, etc.).

Example 6

In order for the AI module to detect the quality of network cables installed in linked equipment and buildings, line defect detection data for each network cable is analyzed using deep learning recognition technology, and to detect line defects in network cables. Data collection process to collect a sufficient amount of training data, line fault data is network cable image or signal data in normal and faulty states, and must include various line fault types to ensure diversity for accurate analysis; A data preprocessing process that removes noise from the collected data and converts it into an appropriate format. If it is graph image data, resize it or apply rotation and movement, and convert the data into a form suitable for learning so that the model can understand it; We design a deep learning model to be used for line fault detection and use a convolutional neural network (CNN), which is specialized for image or time series data processing, and build a network by setting the model's layer configuration and hyperparameters. Deep learning model design process; The data learning process begins by using preprocessed data for the constructed deep learning model and optimizes the model to learn how to distinguish between normal and defective states. The training data is generally divided into training, validation, and test sets. ; Once training is complete, apply the model to test data to evaluate accuracy and performance. If the performance of the model is satisfactory, a model evaluation process that uses it to perform line fault detection on real data; It includes a real-time application process that detects defects in network cables by applying the trained deep learning model to a real-time environment, thereby enabling real-time network cable quality monitoring in an intelligent integrated distribution board.

Example 7

The AI module is capable of automatically generating quantitative data and analysis result reports on line quality through a line loss dB measurement sensor to detect the location of disconnections in the intelligent integrated distribution board in advance, and the analysis result report is included in each network cable. Line fault detection result information that provides line fault information detected through deep learning technology; Line quality measurement result information, which quantifies the quality of network cables and provides dB measurements; Network cable location information that indicates the location of each network cable and the disconnection location on the wiring board, allowing confirmation at each point; Time-specific data information that records changes in network cable quality over time and allows you to check trends; Abnormal pattern detection information that uses deep learning recognition technology to detect abnormal patterns or abnormal signs and report unusual events; Recommended action information, which provides recommended actions to resolve the problem if a line fault or deterioration in quality is detected; Graphs and visualization information provided in the form of graphs and charts to visually represent data for easy understanding; Other statistics and information, including additional statistics or related information, as necessary.

Example 8

The control system, which is connected to the AI module and a network, is used to identify and classify abnormal or abnormal events from sensors, and is designed to detect and respond to abnormalities in real time from the quantitative data of each identified and classified sensor. Connected to the anomaly type analysis support and push alarm edge module; the anomaly type analysis support and push alarm edge module collects data from several sensors through the control system, and the sensors are used to monitor network cable quality and line loss A sensor data collection unit that includes not only dB measurement sensors but also sensors that detect various environmental factors, allowing the sensors to transmit data to the control system in real time; Collected sensor data identifies and classifies events through an AI module. The AI module uses deep learning recognition technology to analyze data and distinguish between normal and ideal states. For example, an event identification and classification unit that detects line faults in network cables and environmental changes; An anomaly detection unit that classifies events and detects anomalies in real time based on quantitative data from each sensor, and determines the event as an anomaly when it detects data outside the normal range or an abnormal pattern; When an abnormality is detected, the control system performs an abnormality type analysis, analyzes specifically what type of abnormality occurred, derives a response plan appropriate for each type, classifies the type of line defect, and determines the indoor temperature. An abnormality type analysis unit that classifies abnormalities in environmental factors such as humidity; And when an abnormality is detected, the control system immediately sends a push alarm to the manager or person in charge. The push alarm includes information about the abnormality and details about the event, enabling real-time response, and the manager can immediately respond. It includes a push alarm unit that can take action against abnormal phenomena, thereby minimizing potential failures or problems and maintaining system stability.

Example 9

It includes an AI module (19) with a remote monitoring function that monitors from a remote location, and detects line quality early through the AI module (19) before defects cause damage or cause safety hazards. ) gives a control command to repair the system, and the direction in which the AI module 19 moves the image sensor through learning is the tracking data converted by the trajectory input function by the tracking input module to rotate the image sensor. The signal is sent to the movement controller, and the movement controller adjusts the image sensor movement direction based on the tracking data, and has the function of measuring whether a specific part is properly connected (e.g., tension, etc.) at the adjusted position when the movement is completed. It monitors risk factors with a plurality of sensors and provides voice warnings, and the AI module 19 utilizes a trajectory input function that receives trajectory information from an image sensor to adjust the tension of the cable corresponding to the trajectory according to the input trajectory. Monitor with images.

10: Control system
11: Microresistance measurement module
12: Frequency-based measurement module
13: Line abnormality diagnosis module
14: Failure prediction module
15: Resistance data acquisition software
16: Cable frequency measurement software
17: Delay rate/loss rate detection software
18: Failure prediction power measurement software
20: Intelligent integrated wiring board

Claims (9)

Intelligent integrated distribution board (20) to establish a stable communication infrastructure and prevent network failures; And the intelligent integrated wiring board and a control system (10) connected to a network to enable quality monitoring for each section through remote control and periodic monitoring. In the AI integrated wiring board capable of monitoring network cable quality, including:
The intelligent integrated wiring board (20),
A micro-resistance measurement module for preventing computer failure by detecting the normal connection rate of the LAN connector of the intelligent integrated wiring board as micro-resistance, which is grounding strength;
A frequency-based measurement module for measuring network cable frequency quality of the intelligent integrated distribution board;
A line abnormality diagnosis module that diagnoses a line abnormality by collecting resistance data of the POE line of the intelligent integrated distribution board;
a failure prediction module that predicts a failure of a terminal device by analyzing power consumption data of the POE line of the intelligent integrated distribution board; and
AI captures the image of the graph displayed on the terminal device and converts the operation form for each timeline into numeric data, and in the step of converting the graph into numeric data, collects operational text data of each system using image processing technology. AI integrated distribution board capable of monitoring network cable quality, including modules.
In claim 1,
The control system 10,
Resistance data collection software application for collecting resistance data of the RJ45 LAN connector of the microresistance measurement module;
Cable frequency measurement software application for checking disconnection to measure network cable quality of the frequency-based measurement module;
A delay rate/loss rate detection software application for detecting an abnormal delay rate or loss rate of the line abnormality diagnosis module;
An AI integrated distribution board capable of monitoring network cable quality, including a failure prediction power measurement software application for predicting failure of the failure prediction module.
In claim 1,
The intelligent integrated wiring board,
An AI integrated wiring board capable of monitoring network cable quality, which measures the connection rate for each pair of LAN connectors and transmits an alarm control command signal to the control system to warn of failure factors in advance.
In claim 1,
The intelligent integrated wiring board,
An AI integrated wiring board capable of monitoring network cable quality, further including a POE support module mounted on an integrated wiring board that applies PoE and can detect abnormal symptoms using the POE power consumption data set.
In claim 1,
Each integrated distribution panel of the above intelligent integrated distribution panel acts as an RX/TX gateway, enabling measurement of line propagation delay for each section through an RX/TX sensor or line resistance measurement RX/TX sensor, and can also be used by complex path construction organizations. AI integrated wiring board capable of monitoring network cable quality.
In claim 1,
In order for the AI module to detect the quality of network cables installed in linked equipment and buildings, line fault detection data for each network cable is analyzed using deep learning recognition technology,
To detect line faults on network cables, more than a certain amount of training data is collected. Line fault data is image or signal data of network cables in normal and faulty states, and must include various types of line faults for accurate analysis. data collection processes to ensure diversity; A data preprocessing process that removes noise from the collected data and converts it into a certain format; if it is graph image data, it resizes or applies rotation and movement; and converts the data into a certain form for learning so that the model can understand it; We design a deep learning model to be used for line fault detection and use a convolutional neural network (CNN), which is specialized for image or time series data processing, and build a network by setting the model's layer configuration and hyperparameters. Deep learning model design process; The data learning process begins by using preprocessed data for the constructed deep learning model and optimizes the model to learn how to distinguish between normal and defective states. The training data is generally divided into training, validation, and test sets. ; Once training is complete, apply the model to test data to evaluate accuracy and performance. If the performance of the model is satisfactory, a model evaluation process that uses it to perform line fault detection on real data; A network cable characterized in that it includes a real-time application process that applies the trained deep learning model to a real-time environment to detect defects in the network cable, thereby enabling real-time network cable quality monitoring in an intelligent integrated distribution board. AI integrated wiring board capable of quality monitoring.
In claim 1,
The AI module enables automatic generation of quantitative data and analysis result reports on line quality through a line loss dB measurement sensor in order to detect the location of a disconnection in the intelligent integrated distribution board in advance,
The analysis result report includes line fault detection result information that provides line fault information detected through deep learning technology for each network cable; Line quality measurement result information, which quantifies the quality of network cables and provides dB measurements; Network cable location information that indicates the location of each network cable and the disconnection location on the wiring board, allowing confirmation at each point; Time-specific data information that records changes in network cable quality over time and allows you to check trends; Abnormal pattern detection information that uses deep learning recognition technology to detect abnormal patterns or abnormal signs and report unusual events; Recommended action information, which provides recommended actions to resolve the problem if a line fault or deterioration in quality is detected; Graphs and visualization information provided in the form of graphs and charts to visually represent data for easy understanding; An AI integrated distribution board capable of monitoring network cable quality, characterized in that it consists of additional statistics or other statistics and information including related information.
In claim 1,
The control system, which is connected to the AI module and a network, is used to identify and classify abnormal or abnormal events from sensors, and is designed to detect and respond to abnormalities in real time from the quantitative data of each identified and classified sensor. Connected to anomaly type analysis support and push alarm edge module;
Anomaly type analysis support and push alarm edge module,
Data is collected from several sensors through the control system, and the sensors include not only line loss dB measurement sensors used to monitor network cable quality, but also sensors that detect various environmental factors, so the sensors transmit data to the control system in real time. A sensor data collection unit to transmit to;
Collected sensor data identifies and classifies events through an AI module. The AI module uses deep learning recognition technology to analyze data, distinguish between normal and ideal states, and include an event identification and classification unit that detects network cable line defects and environmental changes;
An anomaly detection unit that classifies events and detects anomalies in real time based on quantitative data from each sensor, and determines the event as an anomaly when it detects data outside the normal range or an abnormal pattern;
When an abnormality is detected, the control system performs an abnormality type analysis, analyzes specifically what type of abnormality occurred, derives a response plan appropriate for each type, classifies the type of line defect, and determines the indoor temperature. An abnormality type analysis unit that classifies abnormalities in environmental factors such as humidity; and
When an abnormality is detected, the control system immediately sends a push alarm to the manager or person in charge. The push alarm includes information about the abnormality and details about the event, enabling real-time response, and the manager can immediately respond to the abnormality. An AI integrated distribution panel capable of monitoring network cable quality, which includes a push alarm unit that can take action on the phenomenon, thereby minimizing potential failures or problems and maintaining system stability.
In the method using the AI integrated wiring board capable of monitoring network cable quality of claim 1,
The grounding strength is measured by measuring the connection rate of the LAN connector and displayed or sent by email for confirmation on site;
The failure prediction module collects and analyzes resistance data for the POE line and determines the failure rate for each terminal device through the MAC address of the terminal device;
Redisplaying the disconnection location and the induced power noise location among the display contents through 3D modeling and 3D mapping, and displaying the disconnection point and fault location using the 3D mapping;
Calculating a network abnormal delay rate or a data abnormal packet transmission/reception rate of the central intelligent integrated distribution panel, the intelligent integrated distribution panel for each floor, and the outlet section in the premises in the intelligent integrated distribution panel, and monitoring and alarming;
The AI module processes the numerical data into numerical data using image analysis technology including an OCR function, and analyzes the numerical data using natural language processing technology. AI integration capable of monitoring network cable quality, including: Method using a wiring board.