Novel In-Line Inspection Tool for Deposit Characterization in Pipelines

Deposition formation inside pipelines is a major and growing problem in the oil and gas industry. The optimal use of prevention and remediation tools such as chemical inhibitors and cleaning processes could lead to major savings due to minimized production problems and optimized pipe cleaning costs. This requires characterization and quantification of the actual deposits inside pipelines and downholes. Recently, a novel deposition inline inspection sensor moving inside the pipeline has been proposed based on "inside-out" electrical tomography. In this sensor, the distribution of electrical properties between the sensor and the pipe wall are estimated based on measurements carried out using electrodes around the sensor. In this study, the next generation sensor moving inside the pipeline is described and a deep neural network based approach to deposit estimation is introduced. Test results from a 70 m long semi-industrial scale flow loop containing paraffin wax and calcium carbonate deposits of different thicknesses are shown. Challenges include the changing position and orientation of the sensor during the low. The results show that the sensor is able to measure both deposit thickness and type with good accuracy which indicates that the sensor is suitable for industrial use. Accurate knowledge about deposits allows future blockage prevention, detecting build-up locations in the early phase, increasing accuracy of multi-phase flow and deposition models, optimization of chemical use and validation of deposit cleaning tools before integrity campaigns leading to overall reduced pipeline operation costs.

Application of Electrical Tomography Imaging Using Machine Learning Methods for the Monitoring of Flood Embankments Leaks

This paper presents an application for the monitoring of leaks in flood embankments by reconstructing images in electrical tomography using logistic regression machine learning methods with elastic net regularisation, PCA and wave preprocessing. The main advantage of this solution is to obtain a more accurate spatial conductivity distribution inside the studied object. The described method assumes a learning system consisting of multiple equations working in parallel, where each equation creates a single point in the output image. This enables the efficient reconstruction of spatial images. The research focused on preparing, developing, and comparing algorithms and models for data analysis and reconstruction using a proprietary electrical tomography solution. A reliable measurement solution with sensors and machine learning methods makes it possible to analyse damage and leaks, leading to effective information and the eventual prevention of risks. The applied methods enable the improved resolution of the reconstructed images and the possibility to obtain them in real-time, which is their distinguishing feature compared to other methods. The use of electrical tomography in combination with specific methods for image reconstruction allows for an accurate spatial assessment of leaks and damage to dikes.

Comparison of Machine Learning Methods for Image Reconstruction Using the LSTM Classifier in Industrial Electrical Tomography

Electrical tomography is a non-invasive method of monitoring the interior of objects, which is used in various industries. In particular, it is possible to monitor industrial processes inside reactors and tanks using tomography. Tomography enables real-time observation of crystals or gas bubbles growing in a liquid. However, obtaining high-resolution tomographic images is problematic because it involves solving the so-called ill-posed inverse problem. Noisy input data cause problems, too. Therefore, the use of appropriate hardware solutions to eliminate this phenomenon is necessary. An important cause of obtaining accurate tomographic images may also be the incorrect selection of algorithmic methods used to convert the measurements into the output images. In a dynamically changing environment of a tank reactor, selecting the optimal algorithmic method used to create a tomographic image becomes an optimization problem. This article presents the machine learning method’s original concept of intelligent selection depending on the reconstructed case. The long short-term memory network was used to classify the methods to choose one of the five homogenous methods—elastic net, linear regression with the least-squares learner, linear regression with support vector machine learner, support vector machine model, or artificial neural networks. In the presented research, tomographic images of selected measurement cases, reconstructed using five methods, were compared. Then, the selection methods’ accuracy was verified thanks to the long short-term memory network used as a classifier. The results proved that the new concept of long short-term memory classification ensures better tomographic reconstructions efficiency than imaging all measurement cases with single homogeneous methods.

Reliable Detection of Water Cut in Inlet Piping Under Stratified Flow Conditions for Mature Fields and WIO and OIW Emulsion Layers in Crude Oil Settling Tanks with Novel Sensor Technology

Objectives/Scope Mature fields operations, which are almost 70% of today's production have a high water cut content. For each barrel of oil produced there can be 3 or more barrels of water. This means that operational conditions are challenging and might not be ideal for the facilities. In crude oil tanks, one of the most crucial operation parameter is the settling time of oil and water. Especially with heavy oils and high water cut, the operational conditions can be challenging with the existence of emulsion/rag layers in the tank. Most common level detection instrumentation struggle with detecting the proper interface levels leading to faulty control that can cause costly remediation and loss of revenue. This paper presents a novel solution by applying electric tomography pipe and probe sensors. Methods, Procedures, Process In electrical tomography, multiple electrodes are attached on the surface of the sensor and excitations are applied to some electrodes and responses are measured from other electrodes. Assuming a fast separation in the following crude oil tank, the operator expects the flow being stratified already in incoming trunk line. In real life this is not often the case: The incoming flow is turbulent meaning that there is no clear water/oil interface. To overcome this a pipe sensor is needed to monitor the flow regime and hence there is a possibility to control the a) chemical feed and b) flow speed to get the flow stratified. As soon as the flow is stratified in a trunk line it will be guided to a crude oil settling tank for an additional separation. In this tank there is a possibility to apply a probe sensor to monitor reliably the emulsion layer between water and oil. This allows settling time, process parameters and chemicals to be optimized to get a clear separation and hence improving the oil and water quality for a further processing. Results, Observations, Conclusions Results from pipe sensor operation in crude oil pipelines will be shared. The results will show an accurate water cut profile across the pipe cross section even under stratified flow conditions. Additionally, probe sensor results in a crude oil tank operation will be shared and hence confirming the reliability and robustness of the probe sensor operation in tanks. One of the key features of the pipe and probe sensors is the full functionality even under severe contamination with deposits on the sensor surfaces. The operational principle of this method will be shared and verified by experimental results. Novel/Additive Information The sensor technology for the tank inspection and piping uses novel electrical tomography with compact electronic and fast-acting computation with high resolution. This type of technology for settling tank application is new.

Probabilistic cracking prediction via deep learned electrical tomography

In recent years, electrical tomography, namely, electrical resistance tomography (ERT), has emerged as a viable approach to detecting, localizing and reconstructing structural cracking patterns in concrete structures. High-fidelity ERT reconstructions, however, often require computationally expensive optimization regimes and complex constraining and regularization schemes, which impedes pragmatic implementation in Structural Health Monitoring frameworks. To address this challenge, this article proposes the use of predictive deep neural networks to directly and rapidly solve an analogous ERT inverse problem. Specifically, the use of cross-entropy loss is used in optimizing networks forming a nonlinear mapping from ERT voltage measurements to binary probabilistic spatial crack distributions (cracked/not cracked). In this effort, artificial neural networks and convolutional neural networks are first trained using simulated electrical data. Following, the feasibility of the predictive networks is tested and affirmed using experimental and simulated data considering flexural and shear cracking patterns observed from reinforced concrete elements.