The present invention relates generally to fields of pipeline monitoring systems, and more specifically, to non-contact condition monitoring (NCM) autonomous ground vehicle (AGV) for inspecting condition of one or more metallic pipelines buried underground and the associated method thereof.
BACKGROUND
[0002] Many modern services rely upon a network of pipes to carry or distribute fluids. Examples include fresh water, waste water and sewage, and fuels such as oil or gas. It is common to monitor the operation of the network and the condition of pipes. In this manner, blockages, leaks or other issues can be identified and scheduled for repair.
[0003] Where pipes are provided above ground, monitoring may be achieved by visual inspection of the pipe exterior. In many cases, pipes are not accessible to visual inspection, being buried underground (for example, buried oil and gas pipelines). Accordingly, pressure/audio sensors or the like may be utilized to detect vibrations of the pipe and thereby provide information on conditions within a pipe.
[0004] Condition monitoring and defect inspection in the buried oil and gas pipelines, made of metals, has always been a challenge for all organizations operating in the Oil and Gas sector.
[0005] There can be two ways to inspect pipelines internally or externally. Normally industry standard intelligent pigs are used internally for the inspection of corrosion and other such integrity issues in transportation pipes. This inspection mechanism is also called as Inline Inspection (ILI) and in its current form it interferes with regular production process of the plants. Since, these ILI Tools (intelligent pigs) are driven by pressure difference of the fluid flowing through the pipe therefore de-pressurized or fluctuating pressure pipelines cannot be inspected by this technology. Accurate localization of the ILI tool and detected pipe anomalies inside the metallic pipeline is still a serious challenge. Sometime these ILI tools get stuck inside the pipelines then locating ILI tool and taking out of the pipeline is another difficult challenge associated with ILI inspection. Inspection by ILI tool is not only intrusive but also require special external field infrastructures like pig launchers and receivers along with special pre-preparations before inspection like internal cleaning of the pipelines and recording all the operational parameters for a period etc. The ILI tool sensors only detect the thickness profile of the pipeline, while completely neglecting the internal dynamics of the microstructure in the core of the metallic pipes, which is ultimately responsible for most of the anomalies. Also based on traditional ILI data and corrosion models alone, the future status of the currently detected cracks, corrosion, and other defects cannot be estimated accurately in essence Predictive Maintenance is not possible. Due to number of factors such as overall cost, infrastructure requirement, pre-preparation conditions, and interference with regular production process etc ILI inspections are performed with significant gaps 3 to 5 years. Once anomalies are detected in ILI, there is no knowledge how detected anomalies are evolving with time till next ILI is performed. Therefore, currently, there is a substantial blind gap between knowledge of health condition of a pipeline between its two successive ILI inspections.
[0006] Currently external inspection is manually performed by a group of workers driving a vehicle along the transportation pipelines performing visual inspection of the transportation pipes for detection of leakage or any other kind of damage. But such external manual inspection processes, currently employed in industry, are highly inefficient, expensive and hazardous as well. Another problem with current external manual inspection process is that such simple visual inspection just doesn’t give any significant information for the anomalies brewing in the buried pipes or cathodic protection layer and it can be helpful only in the worst cases of leakage and deformations. Currently there is no commercially available mechanism to detect corrosion and other such related defects inside the metallic volume of pipes and its dual surfaces from outside well above the ground.
[0007] In fact, there is no proper non-contact external condition monitoring system available to the industry for the buried pipelines to inform the operators about any kind of current defect or prediction of future defects such as corrosion, high stress and pitting of transportation pipes before any kind of serious damage takes place.
[0008] With ever increasing global demand and depleting resources for fossil fuels, oil and gas industry is now positively looking for advanced robotic solutions to increase their productivity and safety. With time easy resources of the fossil fuels are shrinking and newly searched reservoirs, to feed supply demands of global consumption, are mostly located in extreme environmental conditions such as hot deserts, deep water and arctic zone etc. Production of the fossil fuels, in such inhospitable environmental conditions, poses difficult challenges to health, safety and environment (HSE). Tragic incidents like Exxon Valdez and Deepwater Horizon oil spills are examples of such challenges. Therefore, oil and gas industry has lot to learn from successful implementation of robotics and automation for dull, dirty and dangerous (3D) tasks of manufacturing industry. Most of the robotics technologies, currently used in the oil and gas industry, are mainly focused on inspection, maintenance and repair (IMR) of plant facilities with higher frequency and accuracy. Fundamental idea, involved in the automatization of these processes, is based on the principle of teleoperation with skilled operator. Automation of 3D tasks not only improves HSE standards but also lead to much needed economic efficiency by reducing production cycle, floor space and number of staff members required for continuous inspection and manipulation of plant facilities. Considering the risks involved in this industry usage of completely autonomous robots till now, first without achieving very high reliability, is a difficulty choice.
[0009] In Abu Dhabi, many oil and gas facilities are still using at least four-decade-old infrastructure of metallic pipes, carrying expensive and sensitive fluids, but unfortunately there is no availability of inspection solution, which can do non-contact external detection of various kind of defects creeping into such highly vulnerable aging pipes. These pipes are subjected to extreme weather conditions and has not been inspected for a long time leading to unexpected failure many times causing huge loss of revenue and environmental pollution. Whereas, earlier used manual inspection procedures for pipes are not only hazardous and expensive but also inefficient for the risks at stake. Recently, Abu Dhabi National Oil Company (ADNOC) conducted a workshop, “condition monitoring of deeply buried oil and gas pipes” has been mentioned as one of the most critical challenges.
[0010] At present, the Oil and Gas industry does not have any mechanism, for non-contact external inspection of the buried pipeline from the above ground without excavating the trench and exposing the pipe surface. For internal inspection also, tools installed on smart PIGs (Pipe Inspection Gauge), like ultrasonic testing (UT), magnetic flux leakage (MFL) and eddy current, require the sensors to be in close vicinity of the exposed pipe surface.
[0011] NCM inspection measures the distortions of self-magnetic flux leakage (SMFL) from the core metal of the pipeline with the variation of the pipeline’s metal magnetic permeability in stress concentration zone due to the combined influence of various factors such as residual stress, vibration, bending and loading of pipelines, installation stress, and temperature fluctuations, etc. The generation of strain in a metallic object under an applied magnetic field is called Magnetostriction while the reverse of it is called as Villari effect (observed in 1865) where there is a change in magnetization of a metal sample when stress is applied on it. Now this phenomenon is explained by the variation of a ferromagnetic material’s domain structure and also called a magnetoelastic effect. In reality, high-density accumulation of dislocations at microstructure level is finally responsible for every macro level metal loss, crack, local fracture or other such anomalies in the metal. This complex interaction of dislocations with microstructure level magnetic domain is modelled as magnetoelastic, magnetic-plastic and magnetomechanical effect which can be detected in a non-contact way by measurement of SMFL. Detected SMFL is a vector field with three major components, the longitudinal component along the pipeline, the two transverse components, perpendicular to the pipeline and earth and tangential to it. The different patterns of SMFL and correlations between different components of SMFL are used for detection of existing anomalies. Finally, batches of SMFL data, collected over the period, and are used for future fault predictions. At commissioning of the pipeline when hydro-test and the first ILI is performed to test the quality of installation, NCM test should also be performed to check the manufacturing and installation qualities of the newly laid pipeline for the complete quality control.
[0012] Further, the solutions currently available in these industries are the infrastructure inspection and management, like inspection of transportation pipelines, are performed manually which are not only expensive, labor intensive and inefficient but also pose serious threat to assets, environment and human operators in the cases accidents due to human error and negligence.
SUMMARY
[0013] The present invention overcomes the above-described and other problems and disadvantages in the prior art.
[0014] In response to the above-described and other problems and disadvantages in the prior art, the present invention provides a NCM, it is a smart multi-agent inspection mechanism which will enable new era in facility integrity management by deploying AI and Robotics based solutions in the field of condition monitoring and inspection of buried pipes. NCM significantly contributes to the ADNOC’s call for digital oil fields by practically using AI, robotics, machine learning and data-mining for building safer and efficient inspection mechanisms for the infrastructures of oil and gas industry.
[0015] The present invention provides a new alternate robotic inspection mechanism based on robotic agents i.e., autonomous ground vehicle (AGV), non-contact sensors and AI platforms are introduced to complement and overcome the shortcomings of the conventional inspection of the buried pipelines by ILI (In Line Inspection) tool and manual external inspection.
[0016] The present invention provides a capability of predicting the yet to happen future anomalies in the pipeline is not possible with the conventional inspection methods. So, the predictive maintenance mechanism based on Artificial Intelligence (AI) and Machine Learning (ML) for inspection and management of the buried metallic pipes are also integrated in the present invention.
[0017] The present invention provides is aimed at bringing automation and digitalization to the facility integrity management of field-based industries like oil and gas, and power. The present invention replaces and complements the inefficient and risky manual inspection with a new alternative robotic inspection which will be not only efficient in terms of resource allocation and improving safety factor but also significantly reduce spending of companies on regular inspection, maintenance and repair (IMR) tasks. As described earlier the gap between two consecutive ILI inspection can be adequately filled by this newly proposed NCM inspection method so that critical decisions regarding the repair and maintenance of pipelines can be taken before any severe incident takes place. This predictive maintenance capability can enhance the overall life of the pipeline with timely interventions minimizing the overall maintenance and repair cost.
[0018] The present invention provides a robotic inspection mechanism i.e., an autonomous ground vehicle (AGV) NCM has four components, 1) autonomous robotic platforms (Autonomous Ground Vehicle -AGV), 2) non-contact location and inspection sensors (a first non-contact sensor (a general-purpose pipe locator or an underground pipe detector manufactured by VIVAX-METROTECH Corporation) and a second non-contact sensor (is a magnetic sensor, an Electromagnetic (EM) sensor, a Metal Magnetic Memory (MMM) sensor, or a combination thereof)), 3) Artificial Intelligence (AI) and Machine Learning (ML) based sensor data processing and 4) cloud deployment of the data for real-time streaming and analysis.
[0019] The present invention complements and overcome the shortcomings of the conventional inspection of buried pipelines by the ILI (In-Line Inspection) tool. The present invention enables to pre-predict yet to happen macro anomalies based on historical inspection data by using cutting edge Arti?cial Intelligence (AI) and Machine Learning (ML) techniques. The capability of prediction of yet to happen future anomalies in the pipeline is not possible with conventional inspection methods.
[0020] The present invention utilizes a noncontact magnetometric diagnosis (NCMD) technology which can predict anomalies even before any actual macro-level defect start appearing to be detected by ILI sensors or any other macro-defect inspection tools. The present invention can be utilized in the industries specially that has fields with high risk such as nuclear and oil and gas industry. By using cutting edge artificial intelligence and machine learning technologies new robotic platforms changes the scenario even in highly manual traditional oil and gas industry. In the onshore oil and gas industry robotic solutions are used both in upstream and downstream processes, such as site survey, drilling, production and transportation, mainly focused in the form of in-pipe inspection robots (IPIRs), tank inspection robots (TIRs), unmanned aerial vehicles (UAVs) and wireless sensor networks (WSNs) etc.
[0021] In contrast to the existing technologies of monitoring underground pipes, the present invention works on the principle of measuring distortions of residual magnetic fields conditions by the variation of the pipeline’s metal magnetic permeability in stress concentration zone due to the combined influence of various factors such as residual stress, vibration, bending and loading of pipelines, installation stress, and temperature fluctuations, etc. Therefore, the magnetic sensor of the present invention does not require an exposed surface for detection of anomalies.
[0022] Currently as state of art in this inspection technology, these anomalies are not quantified with size and exact nature rather qualitative analyses of anomalies are done. Therefore, these magnetic sensors are mainly used for detecting location of existing anomalies or vulnerable location for future upcoming anomalies without digging soil or removing coating and exposing pipelines. These measurements are also used to qualitatively categorize anomalies in different levels of severity based on relative comparison of inspection parameters like dimension and age of pipelines, pressure and temperature level and intensity of leaking magnetic field etc.
[0023] In the case of ILI inspection of the conventional technologies, a lot of pre-preparation of the pipeline is required with a collection of data and installation of infrastructures as pig-launcher and receivers are just adding to the additional cost. In the case of the present invention, the inspection sensors are always ready for inspection and do not require any pre-preparation of the pipeline or any supporting infrastructures such as launcher and receiver.
[0024] According to the present invention, the magnetic sensor is installed on an Autonomous Ground Vehicle (AGV), which is already equipped with various other navigation and inspection sensors such as GPS, camera, pipe-locator and acoustic sensor, etc. Final analysis of the collected magnetic data will not only identify the existing defects present in the pipelines like crack, stress-led-corrosion, wall thinning and pitting but also provide information for future planning for inspection and maintenance by identifying the most vulnerable zones for future anomalies. Identification of potential zones for future anomalies is based on the identification of stress concentration zones (SCZs) which can be due to plastic deformation or high density of dislocations where all future anomalies like crack, corrosion, pitting and erosion, etc. will take place.
[0025] In contrast to the convention available solutions for pipeline inspection mechanisms, the Autonomous Ground Vehicle (AGV) as per the present invention utilizes Artificial Intelligence (AI) and Machine Learning (ML) for inspection and management of the buried metallic pipelines that enables predictive maintenance capability and prediction of future anomalies in the pipeline based on data collected in the past. Further, the AGV as per the present invention provides an alternative method for external non-contact inspection of pipelines as a screening tool instead of always depending on expensive, tedious and time-consuming ILI mechanisms. Furthermore, the AGV as per the present invention comprises various sensors fitted thereon to detect presence of any hazardous gas thereby enabling gas leakage detection mechanism. Also, the AGV as per the present invention comprises various sensors (including GPS) that not only detects the location of the pipeline based on data collected and share the same at remote location but also captures the exact location of the defect in the pipeline based on data collected and shares the same at remote location thereby achieve geotagging functionality. Further the the AGV as per the present invention can also detect the exact location of ay toll performing the Inline Inspection and share its precise location to the remote devices. The AGV as per the present invention has a Gas-leak detection sensor is used to detect and map the gas leak density and GPS tag the gas density profile.
BRIEF DESCRIPTION OF THE DRAWINGS
[0026] FIG. 1A-1C illustrates a non-contact condition monitoring (NCM) autonomous ground vehicle (AGV) for inspecting condition of one or more metallic pipelines buried underground, in accordance with the present invention.
[0027] FIG. 2 illustrates a non-contact condition monitoring (NCM) method for inspecting condition of one or more metallic pipelines buried underground, in accordance with the present invention.
[0028] FIGs. 3A-3I shows an experimental validation of the a non-contact condition monitoring (NCM) autonomous ground vehicle (AGV) for inspecting condition of one or more metallic pipelines buried underground, in accordance with the present invention.
DETAILED DESCRIPTION
[0029] Embodiments of the present disclosure include various steps, which will be described below. The steps may be performed by hardware components or may be embodied in machine-executable instructions, which may be used to cause a general-purpose or special-purpose processor programmed with the instructions to perform the steps. Alternatively, steps may be performed by a combination of hardware, software, and firmware or by human operators.
[0030] Various terms as used herein are shown below. To the extent a term used in a claim is not defined below, it should be given the broadest definition persons in the pertinent art have given that term as reflected in printed publications and issued patents at the time of filing.
[0031] The term “non-contact sensor” used in the present invention indicates the sensor that uses wear-free technology, the sensor contains no friction on the moving parts, and this eliminates wear and tear and mechanical failure. TO elaborate, the sensor is using a technology that uses non-touching parts. This term non-contact or non-contacting usually refers to a position sensor that can measure movement or displacement in a rotary or linear fashion. A non-contact sensor uses technology which doesn't come into physical contact. The technology within a contacting position sensor will consist of a track and a slider. The slider will move along the track to measure position of an object. This method of measuring position often causes wear and tear within the sensor and although contacting sensors are lower cost, they can cost more long term as they require more frequent replacement. The term “non-contact sensor” used in the present invention indicates monitoring of a particular object without actually coming into physical contact.
[0032] FIGs. 1A-1C illustrates a non-contact condition monitoring (NCM) autonomous ground vehicle (AGV) (102) for inspecting condition of one or more metallic pipelines buried underground, in accordance with the present invention.
[0033] In an embodiment, the NCM AGV (102) includes a first non-contact sensor (106) configured at a front side of the AGV that detects at least one metallic pipeline (104) from the one or more metallic pipelines buried underground, and a second non-contact sensor (108) configured at a back side of the AGV for detecting a magnetic field leakage from at least one metallic pipeline selected from the one or more metallic pipelines buried underground. The first non-contact sensor and the second non-contact sensor are communicably coupled to one or more wheels (110) of the AGV to align said one or more wheels for navigating the AGV over the at least one detected metallic pipeline (so as to travel exactly and precisely over the one or more metallic pipelines buried underground on the surface of ground (500)).
[0034] The NCM AGV (102) further includes a processor (112) coupled to the second non-contact sensor, the processor detects one or more parameters associated with the at least one detected metallic pipeline based at least on the detected magnetic field leakage while the AGV is aligned and navigated over the at least one detected metallic pipeline to thereby inspect condition of the at least one detected metallic pipeline buried underground based on the one or more detected parameters.
[0035] In an exemplary embodiment, the first non-contact sensor (106) is a general-purpose pipe locator or an underground pipe detector manufactured by VIVAX-METROTECH Corporation.
[0036] In an exemplary embodiment, the second non-contact sensor (108) is a magnetic sensor, an Electromagnetic (EM) sensor, a Metal Magnetic Memory (MMM) sensor, or a combination thereof.
[0037] In an exemplary embodiment, the one or more detected parameters are transmitted by a transmitter (114) coupled to the processor (112) of the AGV to a remote computing device (116) for analysis of the at least one detected metallic pipeline (104) to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline (104) and the one or more detected parameters.
[0038] In an exemplary embodiment, the NCM AGV (102) further includes a camera (118) for capturing real-time visual data of the at least one detected metallic pipeline while the AGV (102) is aligned and navigated over the at least one detected metallic pipeline, and wherein the captured real-time visual data is transmitted to a remote computing device for analysis of the at least one detected metallic pipeline to forecast a performance related information in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline, the captured real-time visual data, and the one or more detected parameters.
[0039] In an exemplary embodiment, the one or more detected parameters are analyzed by at least one of an artificial intelligence (AI) technique pre-stored in the remote computing device (116), a machine learning (ML) technique pre-stored in the remote computing device, and a signal processing technique pre-stored in the remote computing device for analysis of the at least one detected metallic pipeline to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline and the one or more detected parameters.
[0040] In an exemplary embodiment, the one or more detected parameters are analyzed by a neural network receiving the one or more parameters as a multimodal information input to the neural network for analysis of the at least one detected metallic pipeline to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline and the one or more detected parameters.
[0041] In an exemplary embodiment, the autonomous ground vehicle (AGV) (102) is an automated guided vehicle or an automatic guided vehicle (AGV).
[0042] In an exemplary embodiment, the second non-contact sensor (108) is selected from laser micrometers, vision measuring machines, profile projectors and microscopes, Vision Systems, CT Scanners, Laser Scanners, Photogrammetry, Articulating Laser Scanning Arms, Structured Light Scanners, and CFS Sensors (Confocal White Light Sensor) devices, and Charge-Coupled Device sensors (CCD) CMOS sensors (Complementary Metal-Oxide Semiconductor).
[0043] In an exemplary embodiment, the one or more parameters associated with the at least one detected metallic pipeline (104) are selected from a magnetic field intensity emanating from the at least one detected metallic pipeline and a magnetic field intensity emanating from the underground environment.
[0044] In an exemplary embodiment, the condition of the at least one detected metallic pipeline (104) is selected from a normal condition or a defect condition due to at least one of cracks present in the at least one detected metallic pipeline, a wall thinning of the at least one detected metallic pipeline, pitting of the at least one detected metallic pipeline, erosion of the at least one detected metallic pipeline, and stress led corrosion of the at least one detected metallic pipeline.
[0045] In an exemplary embodiment, the AGV (102) comprises of a GPS to transmit a current location of the AGV to a remote computing device, and acoustic sensor or a surface acoustic wave sensor to sense one or more signal associated with the the at least one detected metallic pipeline and transmit the sensed signal to the remote computing device.
[0046] In an exemplary embodiment, the first non-contact sensor upon locating the exact location of the pipes underground may send a signal (information) to the processor which then processes the information to identify the current position of the wheels and then directs the electronic control unit (ECU) of the vehicle to move the one or more wheels (110) and align said one or more wheels for navigating the AGV over the at least one detected metallic pipeline.
[0047] In another embodiment, the AGV comprises a navigation sensor coupled to the first non-contact sensor and the navigation sensor is used for navigation of the vehicle.
[0048] As shown in FIG. 1C, the AGV of the present invention uses a non-contact MMM sensor type 11-6W probe, which is a specialized highly sensitive six-channel scanning device with an analog-digital analyzer for non-contact inspecting the self-magnetic leakage field (SMLF) of oil and gas pipelines buried at the depth. This probe can be used with a measuring wheel or independently. The wheel is installed with an encoder on it so that it can transfer the wheel's displacement into electrical impulse signal. When working with the measuring wheel (wheel mode), the probe records the data at the same time as the encoder transmits an impulse signal to it. The wheel not only works as an odometer but also a generator. Also, the probes can work in time mode with the analyzer from where the probes get impulse signal at a specified time interval. Besides, the probe contains two transducers — upper transducer and lower transducer. The lower transducer is on the downside to inspect the SMLF of the pipeline while the upper one is to inspect the environmental magnetic field for reference.
[0049] FIG. 2 illustrates a non-contact condition monitoring (NCM) method for inspecting condition of one or more metallic pipelines buried underground, in accordance with the present invention.
[0050] In an embodiment, the method is performed by an autonomous ground vehicle (AGV) as disclosed in FIG. 1A or 1B.
[0051] At step 202, at least one metallic pipeline from the one or more metallic pipelines buried underground is detected by a first non-contact sensor of the autonomous ground vehicle (AGV). The first non-contact sensor is a general-purpose pipe locator or an underground pipe detector manufactured by VIVAX-METROTECH Corporation.
[0052] In an exemplary embodiment, Pipe locator, uses radar signal or electromagnetic wave to detect buried objects in the subsurface. The pipe locator used in this work has two components: transmitter and receiver. By connecting the transmitter to the Cathodic Protection (CP) pole or uncovered portion of the pipeline, an Alternating Current (AC) with optional frequency would be injected into the pipeline and generates an electromagnetic signal along the pipeline. This electromagnetic signal radiates from the pipe and is known as the signal that received by the receiver. Based on the received signal strength, the receiver sends a digitalized output to the onboard computer, which indicates:
[0053] Angle: Angle between the pipe locator and the pipeline direction. The negative sign indicates the clockwise direction and positive sign indicates the anti-clockwise deviation
[0054] Position: Indicates the position of pipe locator with respect to the pipeline. -1 indicates the pipe locator is on the right side, +1 indicates the pipe locator is on the left side and 0 indicates the pipe locator is directly above the pipeline.
[0055] Relative distance: This number represents the relative distance between the pipe locator and the pipeline. The relationship between relative distance and actual lateral distance.
[0056] Depth: Indicates the distance between the pipe locator and the pipeline structure. When the transmitter is connected properly, this value indicates the actual distance or else it gives some random numbers.
[0057] In the direct connection mode, the transmitter connects to the pipe using cable and inputs an AC with an optional frequency. The receiver, setting the same frequency, will receive the electromagnetic signal from the pipe with information shows pipe located on the left or right, pipe's depth and pipe's direction. The pipe locator can send out a digitized message per second indicating the pipeline location preferably via Bluetooth.
[0058] Since the pipe locator cannot explicitly express the true lateral distance of the pipe locator to the pipe and the relationship between relative-distance and lateral distance is still unknown to us, a trajectory controller is designed to steer the AGV following a trajectory to approach the pipeline and then to track the pipeline based on the relative-distance value. In the trajectory design, it has three level controls. The first level is when the AGV is far the relative-distance is large, and then AGV will follow a 60-degree line to approach the pipe. The second level is when the AGV 1s close enough or the relative-distance is small enough, the AGV would follow a 15-degree line to approach the pipe, which could avoid the AGV running to the pipe too fast in a short distance and oscillating around the pipe. After that, the third level control 1s to keep the AGV running on the top of the pipe. This controller aims at making the AGV can quickly and smoothly track the pipe and at the same time can avoid the AGV purely doing rotating that would damage the cover soil of the pipe and affects the MMM sensor working.
[0059] The working principle and other details about pipe locator are explained in “V. SUDEVAN, A. SHUKLA, and H. KARKI et al Magnetic Memory Method, Prague, CzechRepublic, June 2019.” & X. Zhang, A. Shukla, A. A. Ali, and H. Karki, "A Smart Robotic System for Non-ContactCondition Monitoring and Fault Detection in Buried Pipelines," in Abu Dhabi InternationalPetroleum Exhibition & Conference, 2018: Society of Petroleum Engineers. Pipe locator is attached on the front side of AGV and the digitalized output of Pipe locator such as side indicator, angle, relative distance and depth of Pipe locator with respect to the buried pipeline is then fed to the onboard computer of the platform.
[0060] At step 204, a second non-contact sensor of the AGV detects a magnetic field leakage from at least one metallic pipeline selected from the one or more metallic pipelines buried underground. The second non-contact sensor is a magnetic sensor, an Electromagnetic (EM) sensor, a Metal Magnetic Memory (MMM) sensor, or a combination thereof. The first non-contact sensor and the second non-contact sensor are communicably coupled to one or more wheels of the AGV to align said one or more wheels for navigating the AGV over the at least one detected metallic pipeline.
[0061] In an exemplary embodiment, MMM method is newly proposed NDT technique in 1990 by Russian researcher Dr. Dubov. This method has already been applied to the ferromagnetic fault diagnosis in the power industry, o1l and gas industry, and manufacture factory.
[0062] The present invention uses a non-contact MMM sensor type 11-6W probe, which is a specialized highly sensitive six-channel scanning device with an analog-digital analyzer. The scanning device is manufactured as a telescopic bar with a changeable length. It is designed for non-contact inspecting the self-magnetic leakage field (SMLF) of oil and gas pipelines buried at the depth up to 3m. This probe can be used with a measuring wheel or independently. The wheel is installed with an encoder on it so that it can transfer the wheel's displacement into electrical impulse signal. When working with the measuring wheel (wheel mode), the probe records the data at the same time as the encoder transmits an impulse signal to it. The wheel not only works as an odometer but also a generator. Also, the probes can work in time mode with the analyzer from where the probes get impulse signal at a specified time interval. Besides, the probe contains two transducers — upper transducer and lower transducer. The lower transducer is on the downside to inspect the SMLF of the pipeline while the upper one is to inspect the environmental magnetic field for reference.
[0063] At step 206, a processor of the AGV detects one or more parameters associated with the at least one detected metallic pipeline based at least on the detected magnetic field leakage while the AGV is aligned and navigated over the at least one detected metallic pipeline to thereby inspect condition of the at least one detected metallic pipeline buried underground based on the one or more detected parameters.
[0064] At step 208, a transmitter coupled to the processor of the AGV transmits the one or more detected parameters to a remote computing device for analysis of the at least one detected metallic pipeline to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline and the one or more detected parameters.
[0065] FIGs. 3A-3I shows an experimental validation of the a non-contact condition monitoring (NCM) autonomous ground vehicle (AGV) for inspecting condition of one or more metallic pipelines buried underground, in accordance with the present invention.
[0066] A calibration and validation test of the AGV was performed at test pipeline in Mussafah, Abu Dhabi, UAE (with all the possible confidentiality). This pipeline was laid as experimental test bed for testing of long-range UT tool.
[0067] FIG. 3A shows a pipeline structure used for experimentation; FIG. 3B shows Overall magnetic profile of the pipeline; FIG. 3C shows welding locations are shown bounded in vertical bars; FIG. 3D shows welding locations marked at pipeline schematic diagram; FIG. 3E shows a zoomed section of magnetic inspection data; FIG. 3F shows locations of welding in the pipeline; FIG. 3G shows all the anomalies of pipeline in same frame and marked with vertical rectangles; FIG. 3H shows Location of anomalies at non-welding locations; FIG. 3I shows anomalies in second section of the pipeline.
[0068] The parameters of this pipeline were as follows: Total length: 150 m, Diameter: 12’’, Material: Carbon Steel, Number of Welding: 15, Nature of Defects: Dent and Grinding, Number of Defects: 46. All the defects were created artificially, and it was not due to natural operation cycle of flow in loading conditions.
[0069] The first task in data analysis of long-pipeline is to figure out the welding locations based on magnetic field intensity components of the lower channels. Among all the x-y-z channels, y-component of the magnetic field is the most critical data channel for detecting the location of wildings.
[0070] In FIG. 3C, y-component of magnetic field is represented by red-colour data points. Variation of other components x and z are also observed in correlation to change in y-component along with gradients of all the components. By thoroughly analysing the all three channels data, welding locations are marked as dotted vertical rectangles. It can be clearly observed that, pipeline has total 15 welding, and all the 15 welding locations are correctly detected by non-contact magnetic inspection procedure, making its welding detection success rate 100%. Data also shows few irregular locations of welding reflecting true reality of the structure as well. A zoomed section of magnetic profile is presented in FIG. 3D for closer observation. Since welding has some lateral dimension but in general its centre location be localized at peak location of magnetic data from y-channel, though there are many other parameters like gradients and correlation with other channels can also be used to determine exact localization of these welding.
[0071] FIG. 3G presents zoomed magnetic profile of the first section of total pipeline. In this section there are total 12 artificial defects out of which 11 defects are detected in magnetic data of lower channels, so detection accuracy is almost 92%. All the anomaly locations are marked with vertical rectangles.
[0072] The second section of the pipeline has total 24 artificial defect out of which 15 are detected clearly so detection accuracy is almost 62.5%. In the remaining section there are total 13 defects out of which 11 defects are detected in our analysis which lead to accuracy of 84.6%. Therefore, there are total 47 defects in complete pipeline and out of which total 34 defects are detected in our analysis making success rate of 74%.
[0073] It has been observed that total number of detected defects is varying in different section of the pipeline. Inspection sensor records magnetic field vector from the six channels and presence of natural stress of pipelines leads to leakage of additional stray magnetic field from the surface of pipeline. These stray magnetic field get superimposed over almost uniform but weak magnetic of earth and thus creating overall varying magnetic field profile of the pipeline. Any abnormal changes in the overall magnetic field like peaks, gradients and component correlation show different signature pattern of various kinds of structural stress of the pipelines like welding and anomalies. These stress levels are clearly visible in operational section of pipelines where natural stresses are created due to industrial operation of the pipelines with pressure, temperature, vibration, weldings and various other loading conditions. While this particular pipeline under inspection is a dead pipeline without any real loading conditions of field operations.
[0074] While some embodiments of the present disclosure have been illustrated and described, those are completely exemplary in nature. The disclosure is not limited to the embodiments as elaborated herein only and it would be apparent to those skilled in the art that numerous modifications besides those already described are possible without departing from the inventive concepts herein. All such modifications, changes, variations, substitutions, and equivalents are completely within the scope of the present disclosure. The inventive subject matter, therefore, is not to be restricted except in the protection scope of the appended claims.

WE CLAIM:

1. A non-contact condition monitoring (NCM) autonomous ground vehicle (AGV) (102) for inspecting condition of one or more metallic pipelines buried underground, the NCM AGV (102) comprising:
a first non-contact sensor (106), configured at a front side of the AGV, to detect at least one metallic pipeline (104) from the one or more metallic pipelines buried underground;
a second non-contact sensor (108), configured at a back side of the AGV, for detecting a magnetic field leakage from at least one metallic pipeline selected from the one or more metallic pipelines buried underground, wherein the first non-contact sensor and the second non-contact sensor are communicably coupled to one or more wheels (110) of the AGV to align said one or more wheels for navigating the AGV over the at least one detected metallic pipeline; and
a processor (112) coupled to the second non-contact sensor, the processor detects one or more parameters associated with the at least one detected metallic pipeline based at least on the detected magnetic field leakage while the AGV is aligned and navigated over the at least one detected metallic pipeline to thereby inspect condition of the at least one detected metallic pipeline buried underground based on the one or more detected parameters.
2. The NCM AGV as claimed in claim 1, wherein:
the first non-contact sensor (106) is a general-purpose pipe locator or an underground pipe detector manufactured by VIVAX-METROTECH Corporation;
the second non-contact sensor (108) is a magnetic sensor, an Electromagnetic (EM) sensor, a Metal Magnetic Memory (MMM) sensor, or a combination thereof.
3. The NCM AGV as claimed in claim 1, wherein the one or more detected parameters are transmitted, by a transmitter (114) coupled to the processor (112) of the AGV, to a remote computing device (116) for analysis of the at least one detected metallic pipeline (104) to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline (104) and the one or more detected parameters.
4. The NCM AGV as claimed in claim 1, wherein the NCM AGV (102) further includes:
a camera (118) for capturing real-time visual data of the at least one detected metallic pipeline while the AGV (102) is aligned and navigated over the at least one detected metallic pipeline, and wherein the captured real-time visual data is transmitted to a remote computing device for analysis of the at least one detected metallic pipeline to forecast a performance related information in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline, the captured real-time visual data, and the one or more detected parameters; and
a GPS to transmit a current location of the AGV to a remote computing device, and acoustic sensor or a surface acoustic wave sensor to sense one or more signal associated with the the at least one detected metallic pipeline and transmit the sensed signal to the remote computing device, wherein GPS is configured to perform geotagging based on the inspected condition of the at least one detected metallic pipeline to tag at least a location of the at least one detected metallic pipeline or a location of an defect present in the at least one detected metallic pipeline.
5. The NCM AGV as claimed in claim 1, wherein the one or more detected parameters are analyzed:
by at least one of an artificial intelligence (AI) technique pre-stored in the remote computing device (116), a machine learning (ML) technique pre-stored in the remote computing device, and a signal processing technique pre-stored in the remote computing device for analysis of the at least one detected metallic pipeline to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline and the one or more detected parameters; or
by a neural network receiving the one or more parameters as a multimodal information input to the neural network for analysis of the at least one detected metallic pipeline to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline and the one or more detected parameters.
6. The NCM AGV as claimed in claim 1, wherein the autonomous ground vehicle (AGV) (102) is an automated guided vehicle or an automatic guided vehicle (AGV).
7. The NCM AGV as claimed in claim 1, wherein the second non-contact sensor (108) is selected from laser micrometers, vision measuring machines, profile projectors and microscopes, Vision Systems, CT Scanners, Laser Scanners, Photogrammetry, Articulating Laser Scanning Arms, Structured Light Scanners, and CFS Sensors (Confocal White Light Sensor) devices, Charge-Coupled Device sensors (CCD) CMOS sensors (Complementary Metal-Oxide Semiconductor) and Ultrasonic sensor.
8. The NCM AGV as claimed in claim 1, wherein the one or more parameters associated with the at least one detected metallic pipeline (104) are selected from a magnetic field intensity emanating from the at least one detected metallic pipeline and a magnetic field intensity emanating from the underground environment.
9. The NCM AGV as claimed in claim 1, wherein the condition of the at least one detected metallic pipeline (104) is selected from a normal condition or a defect condition due to at least one of cracks present in the at least one detected metallic pipeline, a wall thinning of the at least one detected metallic pipeline, pitting of the at least one detected metallic pipeline, erosion of the at least one detected metallic pipeline, and stress led corrosion of the at least one detected metallic pipeline.
10. A non-contact condition monitoring (NCM) method for inspecting condition of one or more metallic pipelines buried underground, the method comprising:
detecting (202), by a first non-contact sensor of an autonomous ground vehicle (AGV), at least one metallic pipeline from the one or more metallic pipelines buried underground, the first non-contact sensor is a general-purpose pipe locator or an underground pipe detector manufactured by VIVAX-METROTECH Corporation;
detecting (204), by a second non-contact sensor of the AGV, a magnetic field leakage from at least one metallic pipeline selected from the one or more metallic pipelines buried underground, the second non-contact sensor is a magnetic sensor, an Electromagnetic (EM) sensor, a Metal Magnetic Memory (MMM) sensor, or a combination thereof, wherein the first non-contact sensor and the second non-contact sensor are communicably coupled to one or more wheels of the AGV to align said one or more wheels for navigating the AGV over the at least one detected metallic pipeline;
detecting (206), by a processor of the AGV, one or more parameters associated with the at least one detected metallic pipeline based at least on the detected magnetic field leakage while the AGV is aligned and navigated over the at least one detected metallic pipeline to thereby inspect condition of the at least one detected metallic pipeline buried underground based on the one or more detected parameters; and
transmitting (208), by a transmitter coupled to the processor of the AGV, the one or more detected parameters to a remote computing device for analysis of the at least one detected metallic pipeline to forecast a performance related information, in future by utilizing past pre-stored performance value of the at least one detected metallic pipeline and the one or more detected parameters.