Abnormality detection method and abnormality detection system

Abstract

To advantageously locate a telegraph pole subjected to a dangerous event, such as a collision of a vehicle and other detectable impacts.SOLUTION: A system and a method based on distributed optical fiber sensing (DFOS) and artificial intelligence (AI) for executing location of a position of a dangerous event for a telegraph pole locate an affected telegraph pole from among multiple telegraph poles by using a machine learning method. Data are collected by using a DFOS technique of a telecommunication optical fiber cable, and data collected for specifying an event are analyzed by using an AI engine. The AI engine recognizes various vibration patterns when the event occurs, and advantageously locates the event for the specific telegraph pole and the position on the telegraph pole precisely. The AI engine can analyze the event in real time with a precision of 90% or higher.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present disclosure generally relates to distributed fiber optic sensing (DFOS) systems, methods, and structures. More particularly, the present disclosure describes DFOS and artificial intelligence (AI) systems and methods for locating hazardous events occurring on poles that suspend / support antennas / cables.

Background Art

[0002] As will be readily understood by those skilled in the art, distributed fiber optic sensing systems and methods provide various useful services, such as sensing various physical parameters including temperature, vibration, strain, etc., thereby enabling a new era of infrastructure monitoring and being shown to be very useful.

[0003] Poles are well-known and recognized in modern society and are generally associated with stopping telecommunication facilities, power facilities, and other infrastructure facilities, thereby providing efficient distribution of telecommunication traffic and power. Considering their importance in modern society, it is extremely important to quickly identify an abnormal state and report it to an appropriate responder when an abnormality occurs in telecommunication facilities or electrical facilities.

Summary of the Invention

[0004] Advances in the art are made in accordance with aspects of the present disclosure directed to DFOS and AI systems and methods for performing location identification of hazardous events on poles.

[0005] In clear contrast to the prior art, the systems and methods of the present invention according to aspects of the present disclosure advantageously identify poles that have been subjected to hazardous events such as collisions by vehicles or other detectable impacts. Further, the systems and methods according to aspects of the present disclosure use machine learning methods to uniquely identify a pole affected from among a plurality of poles and the location on the affected pole where the hazardous event occurred. The approach of the present invention uses data collected using a distributed fiber optic sensing (DFOS) with a telecommunication optical fiber cable. An AI engine is used to analyze the collected data to identify events on the poles and their locations. The AI engine analyzes events in real time with high accuracy (above 90%).

Brief Description of the Drawings

[0006] A more complete understanding of the present disclosure can be realized by referring to the accompanying drawings.

[0007]

Figure 1

[0008]

Figure 2

[0009]

Figure 3

[0010]

Figure 4

Modes for Carrying Out the Invention

[0011] Exemplary embodiments are more fully described by the drawings and the detailed description. However, the embodiments according to the present disclosure can be implemented in various forms and are not limited to the specific or exemplary embodiments described in the drawings and the detailed description.

[0012] The following merely illustrates the principles of the present disclosure. Therefore, it will be understood that those skilled in the art can devise various configurations that embody the principles of the present disclosure and are included within its spirit and scope, although not explicitly described or illustrated herein.

[0013] Furthermore, all examples and conditional terms described in this specification are intended solely for the educational purpose of helping the reader understand the concepts contributed by the inventors to facilitate the principles and techniques of the present disclosure, and should not be construed as being limited to such specifically recited examples and conditions.

[0014] Furthermore, all descriptions of this specification that describe the principles, aspects, and embodiments of the present disclosure, as well as specific examples thereof, are intended to encompass both their structural and functional equivalents. Furthermore, such equivalents are intended to include both currently known equivalents and equivalents developed in the future, i.e., developed elements that perform the same function regardless of structure.

[0015] Thus, for example, it will be understood by those skilled in the art that any block diagram in this specification represents a conceptual diagram of an exemplary circuit implementing the principles of the present disclosure.

[0016] Unless otherwise specified herein, the figures constituting the drawings are not drawn to scale.

[0017] As some additional background, note that a distributed optical fiber sensing system interconnects an optoelectronic integrator to an optical fiber (or cable) and converts the fiber into an array of sensors distributed along the length of the fiber. In practice, the fiber becomes the sensor, and the interrogator generates / injects laser light energy into the fiber and senses / detects events along the fiber length.

[0018] As will be understood and appreciated by those skilled in the art, DFOS technology can be deployed to continuously monitor vehicle movement, human traffic, excavation activities, seismic activity, temperature, structural integrity, leakage of liquids and gases, and many other conditions and activities. This is used worldwide to monitor power plants, communication networks, railways, roads, bridges, borders, critical infrastructure, onshore and offshore power and pipelines, and downhole applications in oil, gas and enhanced geothermal power generation. Advantageously, distributed fiber optic sensing is not restricted by line of sight or remote power access and, depending on the system configuration, can be deployed in continuous lengths exceeding 30 miles with sensing / detection possible at every point along that length. Thus, the cost per sensing point over long distances is usually not comparable to competing technologies.

[0019] Fiber optic sensing measures changes in the "backscattering" of light that occurs in an optical sensing fiber when the sensing fiber encounters an event of vibration, strain, or temperature change. As described above, the sensing fiber functions as a sensor over its entire length, providing real-time information regarding the physical / environmental surrounding conditions, and the integrity / security of the fiber. Further, distributed fiber optic sensing data identifies the exact location of events and conditions occurring in or near the sensing fiber.

[0020] FIG. 1 shows a schematic diagram illustrating the generalized configuration and operation of a distributed fiber optic sensing system that includes artificial intelligence analysis and cloud storage / services. Referring to FIG. 1, an optical sensing fiber connected to an interrogator can be observed. As is known, modern interrogators are systems that generate an input signal to the fiber and detect / analyze the signal received after being reflected / scattered. The signal is analyzed and an output is generated that indicates the environmental conditions encountered along the fiber. The signal received in this way can result from reflections within the fiber such as Raman backscattering, Rayleigh backscattering, and Brillouin backscattering. It can also be a forward signal that utilizes the speed difference of multiple modes. Without loss of generality, the following description assumes a reflected signal, but the same approach can also be applied to transmitted signals.

[0021] As is understood, modern DFOS systems include an interrogator that periodically generates optical pulses (or any encoded signal) and injects them into the optical fiber. The injected optical pulse signal is transmitted along the optical fiber.

[0022] At positions along the fiber, a small portion of the signal is scattered / reflected and returned to the interrogator. The scattered / reflected signal transmits information used by the interrogator to detect, for example, a change in power level indicating mechanical vibration.

[0023] The reflected signal is converted to the electrical domain and processed inside the interrogator. Based on the pulse injection time and the time the signal is detected, the interrogator can determine from which position along the fiber the signal is coming and thus sense the activity at each position along the fiber.

[0024] Distributed Acoustic Sensing (DAS) / Distributed Vibration Sensing (DVS) systems detect vibrations and capture acoustic energy along an optical sensing fiber. Advantageously, existing optical fiber networks carrying traffic can be utilized to convert into distributed acoustic sensors and capture real-time data. Further, classification algorithms can be used to detect and locate events such as leaks, cable faults, intrusion activities, or other abnormal events including both acoustic and / or vibration.

[0025] Currently, various DAS / DVS technologies are in use, and the most common one is based on coherent optical time domain reflectometry (C-OTDR). C-OTDR utilizes Rayleigh backscattering and can detect acoustic frequency signals over long distances. The interrogator transmits coherent laser pulses along the optical sensor fiber (cable). Due to scattering sites within the fiber, the fiber functions as a distributed interferometer having a gauge length (e.g., 10 meters) equal to the pulse length. An acoustic disturbance acting on the sensor fiber generates microscopic elongation or compression (microstrain) of the fiber, resulting in changes in the phase relationship and / or amplitude of the optical pulses passing through the fiber.

[0026] Before the next laser pulse is transmitted, the previous pulse must travel the full length of the sensing fiber and the time for its scattering / reflection to return. Thus, the maximum pulse rate is determined by the length of the fiber. Therefore, acoustic signals that vary in frequency up to the Nyquist frequency, which is usually half the pulse rate, can be measured. Higher frequencies decay very rapidly, so most of the frequencies relevant to event detection and classification are in the lower range of the 2 kHz range.

[0027] Figure 2 is a schematic diagram showing wires / cables suspended / supported in the air by utility poles according to aspects of the present disclosure. As can be seen from this figure, utility poles typically include aerial communication cables including telephone and cable TV wires / cables.

[0028] Of course, those skilled in the art will easily understand and recognize that utility poles, as shown in the figures, have been widely used to support electric wires that supply power from power companies to residents and enable the growth of telephone, television, and Internet networks. Such utility poles are generally made of wood, and the wooden material provides great flexibility for the replacement of hardware and cable devices. However, such wooden materials are vulnerable to external hazards such as damage from winter snowplows (under the communication cable), collisions with automobiles (under the communication cable), collisions of drones with utility poles (above the communication cable), and damage to trees (anywhere on the utility pole). These hazardous events occurring on utility poles can impart various vibration levels to the utility poles, and thus, various levels of attention are required. As is known, a distributed fiber optic sensing system can use an existing telecommunication optical fiber cable as a distributed sensor to capture the response of the optical cable resulting from mechanical impacts on a utility pole. However, current distributed fiber optic sensing systems are not designed to classify these events (under or above the communication cable). In the present invention, a machine learning algorithm for identifying the locations of these events is designed based on vibration data collected from distributed fiber optic sensing.

[0029] As can be easily understood and recognized, the above-described electric wires / cables installed on (suspended from) utility poles are subject to dynamic hazards such as fallen trees, animal activities, drones / kite flying, automobile accidents, and weather conditions that can affect the suspended electric wires / cables. Failure to appropriately address such hazards can result in serious service interruptions and threats to people / property.

[0030] As further understood, when a danger affects an electric wire / cable suspended from a utility pole, it is extremely important for service maintenance / restoration to quickly identify the affected wire / cable and report the exact location of the affected utility pole. This is even more important in the case of rural utility poles, as it would take an inordinate amount of time for human technicians to locate the affected / severed utility pole / wire / cable section by section without first accurately identifying which pole it is.

[0031] Figure 3 is a schematic flow diagram showing the overall determination of dangerous events for utility poles according to an aspect of the present disclosure. As exemplarily shown in this figure, a mechanical impact on a utility pole is detected / collected as a DFOS signal from the operation of the DFOS system. The data thus collected is processed and used to create a model. This model is then used to analyze / evaluate / predict the nature of the impact applied to the utility pole.

[0032] It should be noted that the approach of the present invention uses distributed fiber optic sensing in a telecommunication optical fiber cable to collect data. An AI engine is used to analyze the data collected for event identification and location determination. Advantageously, the AI engine of the present invention can recognize various vibration patterns with high accuracy even when events occur at various utility pole locations. The AI engine of the present invention can analyze events occurring in real time with high accuracy (above 90%).

[0033] As described above, the system and method of the present invention advantageously employ the collection of high-quality data and machine learning models to identify patterns.

[0034] For the data collection process, a distributed fiber optic sensing interrogator is connected to an aerial optical sensor cable to collect strain signals along a target path that may advantageously include multiple utility poles. To confirm the validity of the raw data, data quality checks, filtering, and windowing are applied. The data is collected using a "hammer test" where mechanical shocks are applied to the utility poles along the path (i.e., using a hammer), and during this test, DFOS data is collected.

[0035] In the case of the machine learning model, the process is divided into a training phase and a testing phase. The data is collected and preprocessed. In the training phase, half of the collected data is used, and in the testing phase, half of the collected data is used. In this model, 50,000 estimators are used for the ensemble method.

[0036] The machined learning model is pre-trained and pre-tested. There are a total of 1400 data points, which are divided into half for training and half for testing. A tree-based ensemble method is used with 50,000 estimators. The training accuracy is 100% and the testing accuracy is 91%.

[0037] As shown in the figure, external events such as falling limbs, trees, and kites can affect or cause anomalies in the suspended wires / cables. When detected in this way, the location of the utility pole may be characterized.

[0038] Figure 4 is a schematic flow diagram showing the identification of a specific wire type related to anomalies in wires / cables suspended in the air from utility poles according to aspects of the present disclosure.

[0039] Referring to this figure, in step 1, it can be seen that a distributed optical fiber sensing interrogator is connected to an aerial optical sensor cable and collecting strain signals along a target path. In this step, data quality checks, filtering, and windowing are also applied to confirm the validity of the raw data.

[0040] In step 2, a "hammer test" is performed on the utility pole to simulate a mechanical impact event of the pole. In this step, vibration signals (DFOS return signals) generated by the mechanical impact are collected before, during, and after the mechanical impact event. Since the impacts of the hammer test are performed on the upper and lower halves of the utility pole, different effects are imposed on the vibration signals. These signals from the hammer tests provide information for the AI engine to identify subsequent events and their locations regarding individual utility poles and positions on the poles.

[0041] In step 3, the DFOS system and interrogator are connected to the target optical sensor fiber path and start continuous monitoring of the path.

[0042] In step 4, the position of each of the multiple utility poles is identified, and this position is then used for recording mechanical impact events resulting from abnormal impact events.

[0043] In step 5, when an anomaly occurs, the system automatically extracts raw data (DFOS vibration signals) at the position along the optical sensor fiber. At the same time, the vibration data set is transferred to the cloud storage system. After preprocessing of the data, downsampling and accumulation operations are performed, resulting in a stronger signal pattern.

[0044] In step 6, the extracted raw data (DFOS vibration signals) is processed and fed into a machine learning model for subsequent prediction.

[0045] Finally, in step 7, the machine learning model identifies which part of the utility pole is affected. The system triggers an alarm and sends the event to the cloud - based for recording and initiating subsequent service dispatch.

[0046] At this point, some specific examples have been used to present the present disclosure, but those skilled in the art will recognize that the present teachings are not so limited. Therefore, the present disclosure should be limited only by the claims appended hereto.

Claims

1. A receiving step of receiving backscattered light including a pattern according to a state of the utility pole from an optical fiber laid on the utility pole; and detecting at least one of damage caused by a snowplow, a car collision, and a drone collision as an anomaly to the utility pole based on a pattern contained in the backscattered light.

2. The anomaly detection method according to claim 1 , wherein the detection step detects the anomaly based on a vibration pattern contained in the backscattered light and a learning model.

3. The anomaly detection method according to claim 1 , wherein the detection step detects the anomaly based on an acoustic pattern contained in the backscattered light and a learning model.

4. The anomaly detection method according to claim 1 , further comprising a position identifying step of identifying a position of the utility pole on which the anomaly is detected by the detection step, based on a pattern contained in the backscattered light.

5. The anomaly detection method of claim 1 , further comprising: storing information about the anomaly in a cloud storage facility.

6. The anomaly detection method of claim 2 , wherein the learning model is generated using backscattered light that includes a pattern corresponding to a mechanical shock applied to the utility pole.

7. The anomaly detection method according to claim 6 , wherein the mechanical impact is applied to the utility pole above and below a position where the optical fiber is laid on the utility pole.

8. a receiving means for receiving backscattered light including a pattern corresponding to the state of the utility pole from an optical fiber laid on the utility pole; and an anomaly detection means for detecting at least one of damage caused by a snowplow, a car collision, and a drone collision as an anomaly to the utility pole based on a pattern contained in the backscattered light.

Images