Research
Real-Time Gas Flow Leakage Detection: A Machine Learning Approach to Sensitivity and Uncertainty Analysis
Leakage monitoring in flow lines and pipelines is highly important in gas plants due to the relevance of such a system to safety and efficiency. This work will, therefore, attempt to resolve the uncertainties in flow monitoring by integrating machine learning techniques in conducting sensitivity tests on real-time detection mechanisms. In this paper, the effectiveness of pressure-based indicators compared with volume changes has been considered with variations in flow rate and lifting processes. The findings obtained showed that the conventional assumption of the leakage being represented by the difference between initial and final gas volumes is unsatisfactory, especially during the initial pumping phase where inflow rates may appear to be less than outflow rates because of the purging of residual gases. In addition, the ramp-up and plateau stages exhibited a fair amount of variation in inflow and outflow pressure readings, further adding to the leak detection uncertainties. It has, therefore, been deduced that a variable tolerance window will be effective for leak detection based on the differential pressure data analysis between the inlet and outlet gauges. According to the result of the data analysis, the population variance is 5.38; the sample variance varies across different stages of operation, while the maximum tolerance and pressure are 0.166 and 7.9 bars, respectively. The work automates leak detection and simulates the range of variations, showing the potentiality of AI-ML modeling in enhancing real-world applications. In this work, we are pointing out how machine learning integration may enable a completely new way to define the variable tolerance windows that dramatically improve conventional leak detection.
Automation of Gas Leak Detection: AI and Machine Learning Approaches for Gas Plant Safety
Safety and protection of the environment involve real-time gas leak detection. The paper discusses the improvement in the accuracy and speed of gas leak detection using AI based on pressure-based monitoring. The model will be performing a flow consistency check using machine learning techniques for instantaneous detection at distinct stages in flows. Extensive exploratory data analysis was performed to assess the data and to choose the right machine learning models. The findings showed a significant evolution of pressure differences over time; hence, refining the tolerance level for leakage detection down to a fractional ΓΒ±0.166 window was necessary. The gas flow data was divided into training and testing datasets, which consisted of 80% and 20%, respectively. Several AI models were tested, such as linear regression, logistic regression, SVM, and Random Forest-all had a test accuracy of over 99%. This AI-powered monitoring system could trigger an alarm or immediate notification in the case of a pressure drop beyond the defined tolerance window, improving upon the traditional methods of inspection. All of these contribute to improved safety, operational efficiency, and even cost savings. Furthermore, the scalability of the model holds great promise for other industrial scenarios. The animated simulation of the proposed solution was demonstrated.
Modeling the Constant Composition Expansion Test of Black Oil Using Pressure, Volume, and Temperature (PVT) Calculations
The black oil pressure, volume, and temperature (PVT) properties of well X were measured in the laboratory in a PT cell with subsurface and surface recombination samples. Four sets of black oil samples were collected for the analysis. The black oil standard test is a constant composition expansion test (CCE) separator flash test for volatile oils and rich oil gas condensate and a constant volume depletion test (CVD). The PVT analysis was carried out at Reservoir Fluid Laboratory, Port Harcourt. Oil samples were collected from the Q oil field. The PVT analysis results were correlated to validate the bubble point pressure (Pb), oil isothermal combustibility, (Co), oil formation volume factor (Bo), and the oil viscosity (Γ―ΒΒo). The PVT report gives Po = 2000 psi while the standing correlation gives Pb = 1934.271 psi a difference of 65.7 psi, i.e. 3.3% and solution gas/oil ratio 647.3 SCF/STB while the standing correlation gives 671.03 SCF/STB a difference of 3.5%, oil formation volume factor (Bo) of 1.456 res. Bbl/STB while standing correlations give (Bo) of 1.0675 res bbl/STB a difference of 3.6%. The isothermal compressibility of the oil ranges from 10.12 x 10-6 psi-1 at P < Pb (at 4500 psi) to 4.1309 x 1018 cp at 15 psi. The conclusion is that Gas began evolving at 2000 psig and increased as the pressure decreased. Also, it was noticed that at high pressure of 4500 psig the black oil viscosity was as low as 0.54 cp while at a lower pressure of 15 psiag the viscosity recorded was 1.38 cp. The crude is of high viscosity, with an average absolute error = 3.5% (0.035). The reservoir contains heavy crude oil with an API rating of 30.
