Patent Application
20160238547
The present invention relates a sensor skin and a system usable for direct and/or active detection and monitoring of hydrocarbon leakage in oil and gas pipelines, storage structures, and/or transportation structures. More particularly, the present invention relates to sensor skin comprising a network of sensing elements capable of detecting the presence of hydrocarbons, stress, and/or temperature which can be used to monitor the integrity of any hydrocarbon transportation or storage systems and/or structures, including pipelines, pressure vessels, tanks, etc. The present invention also relates to a method of fabricating the sensor skin.
Disruptions, structural deficiencies, and wear and tear in any hydrocarbon transportation or storage systems can cause leakage, which often leads to significant financial losses, environmental contamination, and endangerment to public safety. In order to maintain structural integrity of these systems, the monitoring of structures for defects and leakage is critical. The occurrence of leaks involves the uncontrolled outflow of hydrocarbons. Since hydrocarbon transportation systems such as pipelines are usually in kilometric lengths, structural monitoring for leakage is an issue. To detect pipeline leakage, the systems currently employed range from simple, visual checking by walking along the pipe line to checking using complex arrangements of hardware and software. Some examples of known leak detection methods are Mass Balance, Pressure Point Analysis, Flow Deviation, Acoustic Emission systems, Chemical based systems, Temperature profiles, neural networks, Fiber optic sensing technologies, etc.
The abovementioned leakage detection methods have various limitations. For example, visual monitoring is limited by a lack of accessibility of pipelines because many pipelines are buried or underwater. Acoustic methods suffer from a limited detection range. While fiber optic systems are difficult to implement on existing pipeline structures and such a system is prone to malfunction if there is damage to the sensing cable. Microwave reflectometry systems have been used to localize water leakage from water pipelines. However, such systems require a microwave source and user station close to the pipeline. Apart from above mentioned limitations, most existing leakage detection methods measure leakage indirectly or passively. They also fail to detect or measure small leaks which are the precursor to bigger leaks.
For direct or active detection of hydrocarbon leakage through chemi-resistor materials have been developed for the monitoring volatile chemicals or hydrocarbon leakage. For example, U.S. Pat. No. 5,498,372, entitled, “Oil leakage sensor”, describes fabrication of an oil-leakage sensor based on porous water repellent resin admixed with electrically conductive carbon particles. Similarly, U.S. Pat. No. 5,498,372, entitled, “Electrically conductive polymeric compositions” cites many polymers with electrically conductive carbon black particles as filler materials. A conductive polymer is normally formed when the concentration of the electric conductive fillers reach a critical value, which is known as the “percolation threshold”. The other fillers such as carbon black and metallic particulates which are also used for producing conductive polymers exhibit much higher percolation thresholds which also adversely affects the mechanical stability of polymers.
U.S. Pat. No. 8,012,420, entitled, “Robust low resistance vapor sensor materials” describes compositions of polymer and a plurality of conductive particles for volatile vapor sensing. They used axial geometry conductive particles such as carbon nanotube and carbon black as conductive fillers inside polymers. However, they did not incorporate graphene nano platelets. Due to their much smaller size and planer structure, it facilitates the electron transport increasing conductivity considerably at lower concentration compared to carbon black. It also provides higher strength to weight ratio compared to carbon black providing higher mechanical and thermal stability to the polymer composite.
Apart from carbon based nanoparticulates, U. S. Pat. Pub. 2006/0249384, entitled, “Chemical sensor”, describes a chemical sensor for detection of a gaseous chemical species which incorporates metal oxide nanoparticles as sensing element. The electric conductivity of the sensor varies depending on adsorption of the chemical species on a large surface area of the crystalline metal oxide nanoparticles on the sensor surface.
E. P. Pat. No. 2,572,187, entitled, “Metal and metal oxide co-functionalized single walled carbon nanotube for high performance gas sensors”, proposed a tin oxide nanoparticles synthesized onto the single-walled carbon nanotube coated with metal nanoparticles to fabricate high performing gas sensor. The patent describes the use of the carbon nanotubes and metal oxide, simultaneously to improve the performance of the gas sensors.
Most of the patents and work mentioned above primarily disclose various compositions and use of conductive nano materials and/or metal oxide nanoparticles, but they do not describe any specific composition suitable for use with oil and gas pipelines and storage/transportation structures. They also provide very little information on producing cost effective and high performance prototype sensor systems that can be installed on such structures for hydrocarbon leakage detection and monitoring.
To implement sensing elements skin discussed above on the vast pipeline system, a multi array sensing element network distributed over the pipeline surface is required. U.S. Pat. No. 8,414,489, entitled “Fabrication of multi-sensor arrays”, discloses the methods for fabricating multi-sensor array for a long-term analyte sensor where the analyte is glucose. This patent provides a method for fabricating a large scale multi-sensor array system based on low cost roll-to-roll manufacturing method incorporating air spray.
E. P. Pat. No. 2,433,716, entitled, “Surface spraying device with a nozzle control mechanism and corresponding method”, discloses a handheld, dynamically movable surface sputtering device comprising of one or more nozzles and a nozzle control mechanism to sputter materials on to a substrate. It reports that the area that can be sputtered on the target surface is dependent on the set of nozzle characteristics such as distance and inclination of the nozzle from the target surface. However, due to lack of computer control and algorithm, it is limited for the small surface sputtering.
E. P. Pat. No. 1,740,313, entitled, “Cost effective automated preparation and coating methodology for large surfaces”, proposes an apparatus and method of coating on large surface such as a ship's hull using plurality of spray guns disposed in array positioned using a robotic arm. The array of the spray guns travels linearly down and makes strips. They provide an algorithm to paint large surface area. But except for the control over positions of different spray nozzles, the method lacks control over individual spray gun characteristics such as air pressure, mass flow rate and thickness control.
U.S. Pat. No. 4,614,300, entitled, “Computerized spray machine”, has put forward the improved design of spray machine with computer control. The spray apparatus provides the control over the speed and acceleration of horizontal and vertical movements of the spray gun. It also proposes the method to control over dwell time during spraying using a computer system. The computer system also provides assistance in actuating, monitoring and correcting parameters of the spray procedure to insure a desired spray pattern on the substrate. However, the system requires adjustable spray nozzle and feedback parameters from every components of the system such as air pump, motor controller.
In multi-array and/or network systems, identification of sensing element electrodes is important to identify the location of the hydrocarbon leakage as well as the flow pattern and size of the structural failure. There are various methods presented for electrode identification. U.S. Pat. No. 5,813,404, entitled, “Electrode connector system” taught electrode identification system based on providing a specific code to each electrode which could be recognized when connected to the digital signal converter. While U.S. Pat. No. 7,339,580, entitled, “Method and apparatus for integrating manual input”, presents a proximity transduction circuit which is placed under each electrode to construct flexible multi-touch surface system.
U.S. Pat. No. 8,384,136, entitled, “Demultiplexed nanowire sensor array for detection of chemical and biological species” describes a nanowire sensor array for detecting chemical and biological species containing a demultiplexer array.
There are various methods for finding the location of a disturbance. For example, U. S. Pat. Pub. No. 2007/0247303, entitled “Event localization within a distributed sensor array,” proposes a method for event localization based on the signal received from the multiple sensor nodes. The localization of a disturbance is performed based on receiving signals that corresponds to the relative proximity of the sensor nodes to the disturbance. The patent discloses an algorithm which determines the location of disturbance by activating only nearby sensor nodes consuming less power. However, due to the lack of a multiplexing system, it is difficult for such systems to handle large amounts of data and to be monitored remotely.
Adaptive Neuro-Fuzzy Interference System (ANFIS) is used in analyzing large amounts of data coming from the sensor network system. ANFIS is combination of artificial neural network and fuzzy inference system. A fuzzy inference systems consists of fuzzy rules (if-then) and based on the past training defines the membership functions associated with the fuzzy sets used in fuzzy rules and infers a decision. In order to come up with the membership functions, the ANFIS utilizes an artificial neural network which is based on the biological nervous system which consists of multiple layers of processing elements called neurons that collect inputs and generate output [Jang et al. 1993].
The ANFIS model provides the means for fusing various sensing element signals and gives the decisions under different conditions. The ANFIS algorithm is trained previously, using a set of sensitivity factors obtained through calibration tests, to determine the sensor response and generate the set of fuzzy rules that transforms the inputs, i.e. the electrical signals from electrodes using the sensitivity parameters (i.e. hydrocarbon detection, stress, and/or temperature).
There are many inventions proposed which uses ANFIS algorithm for various different applications. E. P. Pat. No. 2551798 A1, entitled, “Genetic Algorithm-based training of an ANFIS for electric energy consumption forecasting” provides the method of predicting the electric consumptions in various industrial production systems using ANFIS algorithm. The real time inference algorithm, which determines the selection criteria based on input signals consists of range of operating parameters of system. The input data is fed in to fusion hierarchy and fuzzy inference logic is applied to predict the output.
According to a broad aspect of the present invention, there is provided a composition for use in a sensing element for one or more of: (i) monitoring leakage, (ii) a structural change, and (iii) temperature change in a hydrocarbon transportation or storage structure, the composition comprising: a polymer; and conductive and/or semi conductive nanoparticles admixed with and dispersed within the polymer
According to another broad aspect of the present invention, there is provided a sensor system for one or more of: (i) monitoring leakage, (ii) a structural change, and (iii) temperature change in a hydrocarbon transportation or storage structure, the sensor system comprising: one or more sensing element comprising a polymer and conductive and/or semi conductive nanoparticles admixed with and dispersed within the polymer, each sensing element having a positive electrode and a passive electrode, the passive electrode being connected to ground; and a data acquisition system in communication with the positive electrode, for receiving electrical signals from the one or more sensing elements.
According to yet another broad aspect of the present invention, there is provided a system for collecting and processing signals from a sensor network comprising a plurality of sensing elements, each of the sensing elements exhibiting a change in resistance in response to one or more of: presence of hydrocarbon, a change in vibration signature, and temperature change, each of the sensing elements having an electrode, the system comprising: a voltage divider circuit connected to each of the sensing elements for converting the change in resistance therein into a voltage signal, the voltage divider circuit comprising a current limiting resistance connected in series to sensing element; a multiplexer microswitch in communication with the voltage divider circuit, for detecting and recording the voltage signals; and an adaptive neuro-fuzzy inference system (ANFIS) in communication with the multiplexer microswitch, the ANFIS calibrated to process the voltage signals and provide information relating to the status of the sensor network's surroundings.
Referring to the drawings, several aspects of the present invention are illustrated by way of example, and not by way of limitation, in detail in the figures, wherein:
The description that follows and the embodiments described therein, are provided by way of illustration of an example, or examples, of particular embodiments of the principles of various aspects of the present invention. These examples are provided for the purposes of explanation, and not of limitation, of those principles and of the invention in its various aspects. In the description, similar parts are marked throughout the specification and the drawings with the same respective reference numerals. The drawings are not necessarily to scale and in some instances proportions may have been exaggerated in order more clearly to depict certain features.
It is important in the industry to be able to pinpoint the location of a leak and/or a structural defect in structures that are used to store and/or transport hydrocarbons, since a large amount of hydrocarbons may be lost due to leaks and defects. Rapid detection of leaks and structural defects may help prevent and/or mitigate spills. Oil spills have been a major issue for the industry not only economically, but also environmentally. With increasing environmental awareness, there is a need to develop a low cost and reliable system to help protect the environment from oil spills.
The present invention aims to overcome the challenges outlined above by providing a sensor system for direct and/or active hydrocarbons leakage detection and monitoring. In one embodiment, direct and/or active detection and monitoring is accomplished by installing a smart skin, having a network of sensing elements disposed thereon, on to hydrocarbon storage and transporting structures. The present invention aims to provide an overall solution for the hydrocarbon leakage detection and monitoring system for pipelines, storage tanks and other industrial bodies containing hydrocarbons.
In one embodiment, one or more sensing elements are disposed on a sheet of material to provide a sensor skin (also sometimes referred to as “smart skin”) for detecting and monitoring hydrocarbon leakage. The sensing element comprises a first polymer capable of swelling in the presence of a fluid containing hydrocarbons. The sensing element may further comprise electrically conductive carbon nanoparticles and semi-conductive metal oxide nanoparticles. The conductive carbon nanoparticles incorporated into the sensor material may include, for example, carbon nanotubes (CNTs) and graphene nanoplatelets (GNPs), both of which have excellent electrical properties.
In one embodiment, the detection of hydrocarbon leakage is performed by detecting a change of the electrical conductivity of the sensing element due to adsorption and/or absorption and desorption of the chemical species on the surface of sensing element. Further, the sensitivity of the sensing element can be tailored by incorporating metal oxide nanoparticles.
Alternatively or additionally, one or more of the sensing elements comprise a second polymer admixed with CNTs for monitoring the structural integrity of surfaces, including for example surfaces of pipelines, pressure vessels, storage tanks, and other structures for storing and/or transporting hydrocarbons. These sensing elements are sensitive to changes in the external vibration (i.e. stress or strain) and temperature, which helps to determine the dynamics of the structure. Based on the dynamics information, the failure of a structure due to corrosion and/or other external factors may be predicted and/or detected before any substantial damage to the structure.
In a preferred embodiment, the sheet of material is a flexible material, such as flexible polymer, including for example polyimide, polyethylene terephthalate (PET), polycarbonate (PC), fluorene polyester, and the like, thereby allowing the smart skin to be applied to most surfaces. In one embodiment, the sheet of material is a polyimide sheet based printed circuit board having a non-invasive electrode network system fabricated therein for electronic communication with a data acquisition system.
a) Sensing Element
According to one embodiment, the smart skin comprises a plurality of chemical sensing elements (sometimes also referred to herein as “SE-A”), each containing a polymer admixed with electrically conductive nanoparticles and metal oxide nanoparticles. The smart skin is deployed to cover at least part of a surface, such as a pipeline outer surface. The sensing elements in the smart skin are comprised of a swellable polymer admixed with electrically conductive particles such as carbon based nanoparticles, including for example carbon fibre, graphite nano powder, carbon nanotube (CNT), and graphene nanoplatelets (GNP). In one embodiment, the SE-A contains a polymer and conductive particles, especially carbon nanotubes (CNT), exfoliated graphene nanoplatelets (xGNP), and metal oxides, in any combination exceeding the percolation threshold.
Chemical sensing elements are configured to have a great affinity towards hydrocarbons. The presence of hydrocarbons can therefore be directly detected by detecting a change in the electrical conductivity of the sensing elements resulting from the adsorption and desorption of chemical species on the surface of sensing elements. Further, the sensitivity of the sensing elements may be tailored by incorporating metal oxide nanoparticles.
SE-A comprises an admixture of conductive particles and a swellable polymer, i.e., a polymer material capable of swelling when exposed to hydrocarbon fluid. The swellable polymer has an ability to absorb and swell upon exposure to hydrocarbon without degrading or dissolving, and is hydrophobic and reversible, i.e., capable of recovering to its original size and conductivity once the hydrocarbon is removed therefrom. The swellable polymer for use with the present invention is preferably low cost and easy to process, and has low curing temperatures. Examples of suitable polymer materials include synthetic rubber, polyvinyl chloride, polymetheyacrylate, silicone based polymers, and the like. Silicone based polymers are preferred, given their water and corrosion resistant properties.
The conductive particles selected for use with the present invention preferably have excellent electrical conductivity, and are inert to hydrocarbons and to water or moisture. Examples of suitable conductive particles include gold, platinum, silver, copper, nickel, ferrite, electrically conductive carbon, and the like. Carbon based conductive particles such as carbon nanotubes and graphene nanoplatelets are preferred due to their excellent electrical conductivity. The mechanical and thermal properties of carbon nanotubes and graphene nanoplatelets may provide thermal stability and durability to the polymer in the SE-A.
In a sample embodiment, the SE-A comprises a polysiloxane copolymer (a silicone based polymer) and a plurality of nanoparticles homogeneously dispersed in the polymer using ultasonication technique. The nanoparticles comprise an equally weighted carbon nanotubes and semi conductive metal oxide nanoparticles. The SE-A shows different sensing characteristics associated with different oxide nanoparticles which may be useful in different applications.
Silicone based polymers (also referred to as “siloxane polymers”) are the preferred co-monomers for the SE-A. The siloxane polymer comprises “[—SiRR′O—]” functional group with various alkyl and aryl R and R′ side groups. The polysiloxane polymers are preferred as the base polymers for the SE-A due to their ability to facilitate selective sensing of chemical vapors [Huang et al. 2010].
A benefit of CNTs is their high electrical conductivity and aspect ratio, which help reduce the concentration required to achieve the percolation threshold. The percolation threshold is a critical concentration of conductive particle required in insulating the polymer to form a conductive network. Thus, a low concentration of CNTs is needed to achieve same conductivity as that of other conductive particles, which may assist in reducing fabrication and material costs [McLachlan et al. 2005]. Further, due to their tubular geometry, CNTs may facilitate the elastic deformation or buckling of the SE-A, which may improve the robustness and stability of the SE-A. CNTs have a unique nanostructure, which gives CNTs a large surface area having high electrical conductivity, and thermal and chemical inertness properties. These properties make CNTs a good choice for use in the SE-A.
The polymer film reinforced with carbon nanotubes offers a unique approach to pipeline leak detection, where the sensing mechanism is attributed to the effect on the inter-CNT conductivity from physical absorption of hydrocarbon molecules between CNTs [Li et al. 2011]. The change in the resistance in the SE-A is due to the charge transfer between chemical molecules and CNTs, which varies based on the concentration of the chemicals absorbed in the nanocomposite. The quantum mechanical tunnelling effect controls the electrical response of CNT composites. As the polymer swells upon absorption of hydrocarbons, the increased volume of the polymer increases the distance between adjacent nanotubes, thereby decreasing the chance of electron transfer. Consequently, with fewer conducting paths available, the resistance of the SE-A increases. High sensitivity and reversibility of CNTs to hydrocarbons are reported in the literature, thereby affirming the applicability of the CNT nanocomposite in detecting hydrocarbons [Li et al. 2011].
xGNP nanoparticles have a smaller size and planer structure compared to materials disclosed in the prior art. These features facilitate electron transport and increase conductivity at lower concentrations, as compared to, for example, carbon black. The xGNP also has a higher strength to weight ratio compared to carbon black, which may provide higher mechanical and thermal stability to the polymer composite. In a preferred embodiment, xGNP is used in combination with CNT and/or the other particle types in the SE-A, to help expand the range and sensitivity of the sensing element. xGNPs do not react with the polymer and/or other chemicals, but present increased sensitivity with respect to CNTs.
Additional characteristics and sensitivity of the SE-A may be achieved by incorporating metal oxides into the polymer. Metal oxides may be admixed with the CNT to impart different sensing behavior to the sensing elements depending on the intended environment in which the SE-A is to operate. Metal oxides form wide-ranging nanostructures, which may provide a larger surface area for chemical absorption and stable performance over a large range of temperatures to which the SE-A may be exposed. Examples of metal oxides that may be used with the polymer include titanium dioxide, zinc oxide, iron (III) oxide, and any combination thereof.
Alternatively or additionally, the smart skin may comprise structural monitoring sensing elements (also sometimes referred to herein as “SE-B”) that contain CNTs admixed with other hydrophobic and oleophobic polymers that are capable of monitoring the structural health of the pipeline and other devices. In one embodiment, SE-B comprises a thermoplastic polymer admixed with CNTs homogeneously dispersed using ultrasonication. The CNTs form a conductive network inside the polymer. The conductivity of this type of sensing elements is affected by external stress, external strain, and/or forces applied to the surface on which the smart skin is deployed, and changes in surrounding temperature. The structural dynamics of the pipeline including external stress/strain measurements are computed as a function of recorded changes in the electrical conductivity of the sensing elements. This may assist in predicting potential failure of the pipeline structure on or near which the smart skin is applied, as well as any external and/or internal corrosion and physical damage to the pipeline surface.
Apart from hydrocarbon sensing and structural monitoring, both types of the sensing elements (SE-A and SE-B) can be used to passively sense insulating coating wear on pipeline surface.
The present invention provides a nanoparticle-reinforced polymer coating for direct hydrocarbon leak detection and monitoring of pipelines. The two types of sensing elements, SE-A and SE-B, provide four major functions to the sensor skin, i.e.:
b) Fabrication
Carbon nanoparticles and other nanoparticles discussed above may be admixed with swellable polymer resin in an organic solvent such as chloroform, toluene, benzene and the like to form a coating composition for the printing of the sensing elements. This coating composition is referred hereinafter as the “ink”. Examples of suitable ink based fabrication processes include spray coating, silk screening, spin coating, and the like.
In one embodiment, a fabrication process using a spray coating technique comprising compressed air nozzle is developed to manufacture smart skin. In a preferred embodiment, the smart skin layer is composed of both SE-A and SE-B elements, thereby providing multi-functionality such as detecting hydrocarbon leakage, predicting structural failure of pipelines, generating patterns of hydrocarbon leakage, an locating surface coating wear on pipeline surfaces.
In additions to the favourable compositions suitable specifically to the hydrocarbon oil and gas materials that pipelines carry, a surface spray system is provided herein. The surface spray system has the capability of dynamic movements. In one embodiment, the surface spray system comprises a single nozzle for spray painting on to a target substrate and a control mechanism for nozzle and substrate stage to produce different arrangements of spray coated films.
The use of a roll-to-roll manufacturing process to incorporate the spray system for fabrication of sensing elements is also described herein. The roll-to-roll manufacturing process may provide the benefit of fabricating sensing elements in large rolls, which may be wrapped around the outer surface of a tubular structure, such as a pipeline.
Two installation methods for the sensor skin system are provided herein. The first method comprises wrapping a long roll of smart sensor skin around long pipelines. The second method comprises a smaller sensor skin system which may be deployed on an accident prone area of the pipeline.
c) Electronics
In one embodiment, the non-invasive electrode network system uses electrically conductive electrodes to support smart skin system, which is described in more detail herein below. Possible electrode materials include for example, silver, copper, gold, platinum, etc. An analog electrode tagging technique may be used for electrode identification in the sensor skin system. Such technique allows identification of the position of the sensing element electrodes in the sensor network without a large amount of electronic infrastructure.
d) Data Processing & Monitoring
An algorithm is described herein for pinpointing the location of a leakage and/or structural defect based on the signals received from various sensing elements in the sensor skin. At the leakage location, signals from the sensing elements closest to the leakage reflect higher deviation as compared to the elements that are further away. The algorithm includes multiplexing of a large number of signals coming from the sensing elements and analyzing same using an adaptive neuro-fuzzy inference system (ANFIS). The set of fuzzy rules that transform the input signals to the presence of hydrocarbon based on sensitivity parameters (i.e. hydrocarbon affinity) of the sensing elements are also described herein. In one embodiment, the location of hydrocarbon leakage is determined by detecting the relative discontinuity in the reflected signals.
In one embodiment, the ANFIS analyzes a large amount of data coming from the plurality of sensor elements. The ANFIS combines artificial neural network and fuzzy inference computer-based technologies. A fuzzy inference system consists of fuzzy rules (if-then) and based on past training defines the membership functions associated with the fuzzy sets used in fuzzy rules and infers a decision.
The ANFIS model provides ways for fusing various sensing element signals and gives the decisions under different conditions. The ANFIS algorithm is pre-trained, using a set of sensitivity factors obtained through calibration tests, to determine the sensor response and generate the set of fuzzy rules that transform the inputs, i.e. the electrical signals from the electrodes and the sensitivity parameters (i.e. hydrocarbon detection) of the sample. According to one embodiment, several electrode outputs are used to detect the condition of the specimen through the utilization of the neuro-fuzzy method. Through the use of the ANFIS algorithm, it is possible to determine the location of disturbances (e.g. presence of hydrocarbon, applied force, stress, strain, vibration, etc.) on the sensor skin
The smart skin system as illustrated in
Several experiments are performed to investigate the capabilities of the chemical sensing elements SE-A. To test the capabilities in various aspects, the following criteria have been developed:
To study the above criteria, experiments are divided into three main categories. Each experiment is designed to test one or more of the criteria described above. The three experiment categories are:
In chemical sensing elements SE-A, different compositions of sensing elements affect the abilities of the sensor skin and allow the sensor skin to be applied for various applications under different circumstances. In this case, by varying the sensing elements in the sensor skin, the sensor skin can be configured and tailored to specific applications and/or environments in the pipeline industry. To test the sensor skin for use in different applications and/or environments, experiments have been conducted with different compositions of nanoparticles mentioned above.
Experiments with a Single-Electrode Sensor Skin
The first category of experiments are designed and performed to examine the capabilities of the sensing elements. A sensor skin with a single sensing element electrode is fabricated and tested. Six compositions have been tested to show the functionality of the different nanoparticles.
With reference to
The data indicates that a polymer admixed with a mixture of carbon nanoparticulates has the most hydrocarbon sensitivity when the sensing element is exposed the crude oil vapor. For example, with reference to
With reference to
Table 1 below provides a comparison between the hydrocarbon sensing properties of the sensing elements comprising the different compositions mentioned above.
The data shows that various sensing elements comprising different nanoparticles may be used for hydrocarbon detection. Further, the sensing elements display a substantially immediate response (i.e. an increase in resistance) to the presence of crude oil vapour, when the elements are in the vicinity of crude oil. Therefore, the data suggest that direct contact with crude oil is not necessary for hydrocarbon detection and that the sensing elements are capable of detecting even a very small amount of hydrocarbon leakage, almost instantaneously.
The response sensitivity towards hydrocarbons can be altered by varying the size and/or surface area of the sensing elements as size and/or surface area affects the electrical conductivity of the sensing elements. An increase in size decreases the resistance of the sensing elements, which improves the percentage sensitivity in the sensing elements' response.
The single sensing elements have also been tested with various liquid hydrocarbon products that the pipelines may carry. Such hydrocarbon products comprise mainly bitumen mixed with diluents to facilitate transportation. The diluents are generally low viscosity hydrocarbons such as pentane, hexane, octane, etc. The sensing elements are tested by exposing them to small amounts of the liquid hydrocarbon products. Table 2 below shows the sensitivity of each composition towards the diluents after 10 seconds of exposure. The data in Table 2 suggests that the abovementioned compositions are highly sensitive and react quickly to the diluents. Table 3 below shows the sensitivity of the different compositions to the diluted bitumen after one minute of 100 μL of exposure.
The single sensing elements have also been tested against temperature, in the absence of hydrocarbons.
With reference to
A sensing element is pressurized in the absence of hydrocarbons (
Further, the sensing elements are tested for their stability over long time periods. Each sensing element is kept on (i.e. with current passing therethrough) for more than three months, and the change in its resistance is observed. The sensing elements show a maximum of approximately 3% variation in resistance over the time period. This test suggests that the sensing elements of the present invention have good stability and are reliable over a long period of time.
The various tests on the abovementioned single sensing elements show that the elements are sensitive to the presence of hydrocarbons and are able to detect trace amounts of hydrocarbon whether in air or soil. The sensing elements are also capable of functioning under varying temperatures and compacted soil. They also show sufficient stability over long periods of time.
These characteristics make the sensing elements of the present invention suitable for a variety of applications, including pipelines, storage tanks, and other industrial structures for hydrocarbons.
Experiments with Nine-Electrode Sensor Skin (Vapor Sensing)
As pipeline leakages may happen at almost any location along a pipeline, it is desirable to use a multi-array and/or network of sensing elements electrodes that can cover all or at least a portion of the pipeline outer surface. With reference to
It is desirable to be able to locate a leakage of on a pipeline surface. For testing purposes and with reference to
With reference to
Further, as shown in
Another characteristic of the sensor skin is its reversibility. The reversibility of the sensor can be observed by measuring the time it takes for the resistance or voltage of the sensor to revert back to its normal (i.e. pre-hydrocarbon exposure) values after the hydrocarbon source has been removed. With reference to
With reference to
With respect to hydrocarbon sensitivity, sensing elements containing composition 2 show similar characteristics as those containing composition 1. The experiments suggest that both compositions react to the presence of hydrocarbons by increases in voltage, the magnitude of which is substantially proportional to the proximity to the hydrocarbon source. However, the addition of xGNP to the polymer nanocomposite gives the sensing element different characteristics in terms of sensitivity and retainability. For example, composition 2 shows a maximum change of 15 mV after 10 minutes of hydrocarbon exposure, while composition 1 shows a maximum change of 7 mV after 15 minutes of exposure. This suggests that the composition with GNP has higher sensitivity toward crude oil and takes less time to achieve higher voltage differences. The sensor skin containing GNP also exhibits higher retainability. As shown in
Experiments with Nine-Electrode Sensor Skin on a Pipe (Direct Crude Oil Contact)
For leakage detection, it is desirable to place the sensor skin directly on to the pipeline outer surface above the insulating coating layer of the pipeline. With reference to
To simulate a small hydrocarbon leakage, a hole with a diameter of approximately 1 mm is drilled on one side of the pipe covered by at least a portion of the sensor skin. After collecting data of the initial voltage values of the sensor skin for 5 min, crude oil H is poured into the pipe. After that, the voltage changes of the sensor skin resulting from the leakage of hydrocarbon through the hole are observed.
As shown in
The SE-B of the present invention comprises a thermoplastic polymer, such as for example polycarbonate (PC), polystyrene (PS), and polyvinylidene fluoride (PVDF), etc. and conductive particles, such as for example carbon nanotubes and the like. SE-B may be used in a pipeline monitoring system to predict pipeline surface failures such as corrosion and indirect leakage. The vibration signature of the pipeline structure, caused by externally applied stress, strain, and forces, changes as the pipeline structure ages and deteriorates. The variation in the dynamic parameters (e.g. vibration signatures) is used to predict the location of corrosion and/or leakage in the pipeline system.
A conductive CNT/polymer filler composite is formed when the concentration of CNTs reaches a critical value, which is known as the “percolation threshold”. As external strain, stress and/or forces are applied to the conductive filler composite, it contributes to the breakup of the network between the fillers and increases the inter-filler distance, which promotes a tunneling effect that increases the resistivity of the composite. It has been established that CNTs embedded in a polymer can be used as potential strain sensors using the piezoresistive behavior of composites [Zhang et al. 2006].
With reference to
It can be seen from the experimental data that SE-B has a high affinity towards applied stress, strain, and/or force sensitivity. The vibration signature (i.e. response to external stress, strain and/or force) of the pipeline structure varies with changes to the pipeline structure. SE-B responds to any changes in the vibration signature of the pipeline, which may be a result of physical changes in the pipeline such as, for example, internal and/or external corrosion to the pipeline surface. SE-B may also be sensitive towards changes in the vibration signature of the pipeline resulting from any structural damage to the pipeline, which may be caused by vandalism, weather, wear and tear, natural disaster, etc.
Apart from sensing changes in the vibration signature of the pipeline structure, SE-B may also be used to measure the surrounding temperature of the pipeline. An increase in the resistance of SE-B is attributed to the tunnelling effect, whereby electrons are transported from CNT to CNT inside the polymer. An increase in temperature results in an increase in the tunnelling distances between two adjacent CNTs, thereby causing an increase in resistance of the sensing element.
With reference to
In general, the outer surfaces of steel pipes carrying liquid and gas hydrocarbons are coated with an insulating coating comprising anti-corrosive layers of insulating materials such as epoxy. The smart skin may be wrapped over the insulating coating, thereby placing its sensing elements in physical contact with the insulating coating. The placement of the sensing elements on the insulating coating allows indirect monitoring of the coating. More specifically, if coating wears off at a particular location, the sensing elements at that location are then in direct contact with conductive steel surface of the pipeline, which causes disruptions in the response of the sensing elements, thereby indicating the location of the pipeline coating wear.
In additions to the abovementioned compositions suitable for detecting the presence of hydrocarbon products commonly carried by pipelines, a spray system capable of dynamic movements for fabricating the sensor skin is also provided herein. The surface spray system comprises a single nozzle for spray painting composition ink on to a target substrate and a control mechanism for controlling the nozzle and substrate stage to produce different arrangements of sensing elements on the sensor skin.
In a further embodiment, the spray system comprises a plurality of inks of different or the same compositions. For example, a first ink of the spray system comprises a swellable polymer and a plurality of nanoparticles distributed within the polymer. The first ink is for fabricating SE-A. A second ink and a third ink each comprises thermoplastic polymer admixed with carbon nanotubes. In one embodiment, the second ink comprises 5 wt. % MWCNT/PVDF and the third ink comprises 5 wt. % MWCNT/PC. The sensing elements (i.e. SE-B) fabricated using the second or third ink have a higher sensitivity towards stress, strain, and/or force (i.e. changes in vibration signature), and surrounding temperature. A fourth ink is for spraying an insulating layer around the nanocomposite sensing elements SE-A and SE-B.
A roll-to-roll manufacturing process is also provided herein. The process incorporates the spray system for fabricating the sensing elements on the substrate. The process may allow fabrication of sensing elements in large rolls that can be wrapped around large sections of the pipeline outer surface.
Two installation techniques for the sensor system are provided herein. The first technique comprises wrapping a long continuous roll of sensor skin around a length of pipes. The second technique comprises applying a smaller sensor skin on specific areas of a pipeline.
With reference to
In the spray coating system 50, there are a number of adjustable parameters including for example ink flow rate, nozzle diameter, geometry of atomizers, and various liquid ink parameters (i.e. liquid dynamic viscosity (μ), liquid density (φ, and liquid surface tension (σ)) that can affect the performance of the system.
To obtain substantially homogenous distribution of various nanoparticles inside the polymer, the polymer and CNTs are dissolved into an organic solvent, such as chloroform, and the resulting solution is ultrasonicated for about half an hour. A substantially uniform dispersion of nanoparticles in the polymer gives the composite consistent electric properties in all the direction. In one embodiment, the solvent is added into the polymer until the viscosity of the mixture is around 30-50 cSt., which is the preferred range for spray painting through a compressed air nozzle.
The mixture of polymer and nanoparticulates in the chloroform solvent may be spray painted using spray coating system shown in
In one embodiment, the nozzle of the spray system focuses on one point (“focal point”). The performance of the spray nozzle depends on two parameters: focus length Lg and focus diameter. Focus length is the distance between the nozzle and the focal point, while focus diameter is the spray diameter at the focal point. Beyond the focal length, the ink diverges based on the rate at which the ink is being released from the nozzle. Thus, it is important to determine the location of the focal point and area to be covered by experimentation so as to ensure uniform dispersion and thickness throughout the substrate. The focus diameter is preferably as small as possible, depending on variables such as air pressure and ink viscosity. The focus diameter, for example, is preferably between about 2 mm to about 5 mm, to help ensure substantial uniform thickness on the substrate. Optimum values for the focus length and focus diameter, which may depend on the viscosity of the ink, may be determined through routine experimentation.
In one embodiment, liquid polymeric nanocomposite (“nano ink”) is provided in a glass bottle and the nano ink is siphoned through a tube into an airbrush assembly 58, where the ink is atomized into fine droplets. The ink is delivered to the nozzle via a delivery line 60. Further, pressurized air A enters into the nozzle 54 from a pressure inlet 62, thereby increasing the pressure inside the nozzle. The increase in pressure is accompanied by the atomization of the ink, which causes the ink to exit through the nozzle. In a sample embodiment, the distance between the nozzle and the substrate (i.e., the spray distance) is adjustable via stage 53.
The spray system further comprises a drive system for controlling various parameters such as the three dimensional movement of the nozzle 54, the mass flow rate of the ink coming out from the nozzle, and the spray stroke. These parameters are adjusted based on the type of ink and printing patterns. The drive system may be manually controlled or computer automated.
The spray system is controlled by a computer system. Stages 52 and 53 provide the system with three degrees of freedom. Stage 52, where the substrate is to be placed for spraying, has two degrees of freedom on a plane, which in one embodiment may be substantially horizontal, and stage 52 is movable in the x and y directions on the plane. The nozzle 54 is attached to stage 53, which movable in a direction z substantially orthogonal to the plane of stage 52, thereby allowing adjustment of the distance between nozzle and the open surface of the substrate.
The computer system also controls a solenoid valve 64 which in turn controls the dwelling time and spraying time of the nozzle. As shown for example in
The spray coating system shown in the
The roll-to-roll process can deposit large numbers of nanocomposite elements on a polyimide substrate. It can also control the thickness and geometry of the nanocomposite elements, thereby allowing any desired configuration of the sensing element to be produced. Further, the computer-controlled system may allow the process to be carried out accurately and efficiently.
The sensing elements in the sensor skin of the present invention are connected to an electrode network. In one embodiment, the electrode network is printed using electrically conductive silver ink.
An analog tagging process may be used for electrode identification for the sensor system of the present invention. The analog tagging process is based on a voltage divide circuit and the voltage responses coming from each electrode, which are used to identify the position of the electrode in the sensor network. In one embodiment, a multiplexer microswitch is used to read the responses coming from the sensing element electrodes to overcome the extensive wiring associated with a large number of sensing elements in the smart skin.
With reference to
To keep the sensor skin system flexible, electrodes are printed on the substrate surface by, for example, inkjet printing, line patterning, etc. Conductive silver ink can be used for the electrodes since it can be annealed as lower temperatures and can yield conductivity close to bulk silver [Walker et al. 2012]. Therefore, according to an embodiment of the present invention, an electrode network of conductive silver ink can be printed on polyimide sheets (DuPont™ Kapton® FPC). After printing the electrode network, nanocomposite materials are applied on the electrode network through spray painting technique described above to form the sensing elements of the sensor system.
If direct sensing from individual sensing elements is employed for a large number of sensing elements, the sensor system would require crowded electric wiring to communicate with all the elements and large computer power is required to monitor all the sensing elements. This approach is infeasible for the monitoring of pipelines. To overcome this problem, the sensor system of the present invention uses multiplexing micro switches.
The sensor system of the present invention uses surface in-plane resistance measurement with an analog tagging system to identify electrodes in a sensor network having a large number of sensing elements. To analogically tag the electrode, a voltage divide circuit is used.
After tagging the each electrode, a multiplexing microswitch 60 is used to record the output from each electrode. The output from the positive electrodes 66 is provided to microswitches 62, while all the passive electrodes 68 are connected together to make a common reference ground for all the electrodes. The multiplexing of the electrode signals allow truncating large numbers of input data into a small amount of data, thereby reducing the need for extensive wiring. The signal from multiplexing microswitch 60 can be de-multiplexed at a later stage and the data from each electrode can be recorded on to a computer system. This data can be processed and analyzed further for determining leakage and finding the source thereof.
In one embodiment, ANFIS is used for analyzing the data obtained from the sensing elements in the sensor network. The ANFIS model allows the fusing of different voltage signals coming from the sensing element electrodes and provides decisions under various conditions.
Leakage detection is possible without rigorous ANFIS analysis for a sensor skin having a small number of sensing element electrodes. ANFIS is desirable for large and/or long pipelines where a greater amount of surface area needs to be covered with the sensor skin and/or where both SE-A and SE-B are used simultaneously in the sensor skin. In a preferred embodiment, ANFIS analysis is used to monitor pipeline surface corrosion and/or hydrocarbon leakage.
By employing a sensor fusion methodology wherein various types of sensors with different functionalities (i.e. hydrocarbon detection, vibration signature change detection, temperature change detection, etc.) are included in the sensor system, reliability, redundancy, and robustness of the sensor system may be improved.
An ANFIS model is preferred for fusing various sensor signals from SE-A and SE-B elements in the sensor skin. In one embodiment, the fusion of the signals is performed by a set of fuzzy rules that transforms the inputs, i.e. any signals of voltage change acquired from the different sensors into a possible output which determines the status of the surface covered by the sensing elements. The amalgamation of different signals may increase the reliability of leakage detection, pipeline failure prediction, and failure and/or leakage localization.
In one embodiment, the development of the ANFIS-based pipeline monitoring scheme is completed in two phases, i.e., training phase and verification phase. In the training phase, input variables include voltage responses related to hydrocarbon sensing, vibration signature and temperature signals from the two types of sensors. The training datasets comprising voltage responses from a hydrocarbon leakage, or vibration signature changes due to corrosion or temperature change, all based on laboratory experiments, are provided to ANFIS. The training of the ANFIS-based scheme is an iterative process. Once the training phase is completed, the verification of the trained ANFIS scheme is performed using multiple verification datasets of the input variables obtained from the experiments on the pipeline in different conditions, such as pipeline in normal operation, corroded, and with hydrocarbon leakage. The number of membership functions and types are then defined for each of the input variables. “If and then” rules are generated using first-order Sugeno fuzzy model which is represented by following Equation (1) for n inputs and m membership functions.
If (x1 is A11) and . . . (xn is An1) then f1=p11x1+ . . . +pinxn+r1 (1)
The input membership functions are preferably normalized to the range from 0 to 1. Each of the input variables is mapped by three membership functions, for example, as shown in
As an alternative to sensor signal fusion, in some embodiments it may not be desirable or feasible to have individual electrodes under each sensing element. It is possible to spray paint sensing element material over a larger surface with a particular sensing solution as described above, to form an integrated sensor skin. For example, the pipeline outer surface may be painted with either SE-A or SE-B material. A calculation technique of Electrical Resistance Tomography (ERT) based upon the configuration shown in
ERT is a method that calculates the surface distribution of electrical resistivity from a large number of resistance measurements M made from electrodes 166 which are placed at the sides of an integrated sensor skin 130. In this method, electric current is fed through two adjacent electrodes and the voltage change between the remaining electrodes is measured. Based on the type of sensor solution, when there is a disturbance (i.e. hydrocarbon presence, a change in vibration signature, or a change in temperature) on the structure where the integrated sensor skin is applied, the surface resistivity of the sensor skin 130 changes. This change in resistivity may be measured through the electrodes 166 and the electrode system helps determine the exact location of the disturbance. For in-situ application, ERT provides an efficient way to measure the response from a large number of surface resistance measurements.
Yet another sample sensor system 430 is shown in
For new pipelines, the sensor system may be incorporated during the construction process of pipe itself. Before coating the pipelines with an insulation layer, the sensor system may be installed first and the insulating layer may be applied on top of the sensor system. In this embodiment, the sensor system is insulated and protected by the insulation layer, which may help prevent damage to the sensor system from the environment and/or vandalism. This embodiment may minimize the need for regular checks and maintenance of the sensor system.
The following documents are incorporated herein by reference to the same extent as though fully replicated herein.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to those embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein, but is to be accorded the full scope consistent with the claims, wherein reference to an element in the singular, such as by use of the article “a” or “an” is not intended to mean “one and only one” unless specifically so stated, but rather “one or more”. All structural and functional equivalents to the elements of the various embodiments described throughout the disclosure that are known or later come to be known to those of ordinary skill in the art are intended to be encompassed by the elements of the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims. No claim element is to be construed under the provisions of 35 USC 112, sixth paragraph, unless the element is expressly recited using the phrase “means for” or “step for”.
This application claims priority from U.S. provisional patent application No. 61/891,248, filed Oct. 15, 2013.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/CA2014/050992 | 10/14/2014 | WO | 00 |
| Number | Date | Country | |
|---|---|---|---|
| 61891248 | Oct 2013 | US |
