Browsing by Author "Mahmud, Shahriar"

Now showing 1 - 6 of 6

Analysis of wellbore instability while drilling exploratory wells in Bangladesh
(Department of Petroleum and Mineral Resources Engineering, (PMRE), BUET, 2018-03-13) Evana Tinny; Mahmud, Shahriar
Wellbore instability is one of the most prevalent downhole challenges since oil well drilling began. The difficulties become more profound and unpredictable when the wells are exploratory. Wellbore Instability issues related to exploratory wells causes significant amount of non-productive time (NPT). These problems could lead to fishing operation and sidetrack which may cause greater financial loss. In Bangladesh wellbore related issues especially stuck pipe cases were experienced from time to time. This work analyzes the wellbore instability incidents of two exploratory wells, Mubarakpur-1 and Sunetro-1. Based on available data like Geological Technical Order (GTO), Daily Drilling Reports, Well Logging Reports, Mud Reports, XRD (X-ray Diffraction) reports, Cavings samples with pictures, Well Completion Report etc. this work presents the sequence of events leading to the stuck pipe incident on Mubarakpur-1 and Sunetro-1 wells. Based on available information the root cause(s) of wellbore instability issues of the said wells have been analyzed and suggestions have been made on how to drill similar wells while tackling such issues in future. Mubarakpur-1 and Sunetro-1 are different in geological aspects. Formation pressure and wellbore stresses were higher in Mubarakpur-1 than Sunetro-1. Shale formation showed non-reactive character in Mubarakpur-1, whereas Sunetro-1 suffered from shale swelling effect. Both wells encountered stuck pipe incidents but the sequences of events were different. Mubarakpur-1 faced compressive failure due to stress concentration around the wellbore. And Sunetro-1 crossed the reactive shale instability which resulted in hole pack-off. Some operational mishaps were also responsible for both stuck pipe incidents. The analysis showed that focus should be given on understanding the in-situ stress condition while drilling, identifying formation type. So that appropriate mud weight selection, BHA design can be implemented for trouble free operation.
Cricket shot analysis using machine learning
(Daffodil International University, 2024-07-13) Mahmud, Shahriar
Cricket, being one of the most popular sports globally, relies heavily on the skills and techniques of players. With the advent of technology, there is a growing interest in leveraging machine learning techniques to analyze and enhance players' performance. This research focuses on the development of a Cricket Shot Analysis system using advanced machine learning algorithms. The proposed system utilizes computer vision and machine learning models to automatically recognize and classify cricket shots batsmen play. Highspeed cameras capture the footage of the cricket match, and the frames are processed to extract relevant features such as bat speed, shot angle, and player stance. My dataset contains 8000 images of different shot. Here sweep image data 2000, pullshot images data 2000, legglance-flick images data 2000, drive images data 2000. These features are then fed into a machine learning model, trained on a diverse dataset of cricket shots, to classify each shot into predefined categories like cover drive, square cut, pull shot, etc. The system aims to provide valuable insights into a player's shot selection, technique, and overall performance. Coaches and players can benefit from the detailed shot analysis, identifying strengths and weaknesses to tailor training regimens accordingly. In this research I use Xception, InceptionV3, and DensNet. Anlysis shows that the highest accuracy achieved by DenseNet121 is 98.26% and the lowest accuracy is 96.38% which is achieved by the InceptionV3 model. The development of this Cricket Shot Analysis system not only addresses the growing demand for data-driven insights in sports but also showcases the potential of machine learning in refining and revolutionizing traditional coaching methodologies. Through continuous refinement and validation, this technology has the potential to become an indispensable tool for cricket professionals, contributing to the evolution of player training and game strategy in the modern era of cricket.
Determination of the critical drawdown pressure to predict potential risk of sand production of a gas well in Bangladesh
(Department of Petroleum & Mineral Resources Engineering (PMRE) , BUET, 2016-06) Kaoser Hasan, Md.; Mahmud, Shahriar
Sub-surface formations are always in a stressed state, mostly due to overburden and tectonic stresses. When an oil or gas well is drilled into a formation, stressed solid material is removed. The fluid pressure of the borehole supports the borehole wall. As this fluid pressure generally does not match with the in situ formation stresses, there will be stress redistribution around the wellbore. The new stress may be greater than the stress that the formation can support and subsequently failure may occur. Knowledge of the stresses around a well is therefore imperative to determine the wellbore stability and subsequently any potential risk of sand production. Due to lack of proper knowledge for calculating Critical Drawdown Pressure and predicting sand production, some gas wells in Bangladesh have started producing sand. For the same reason, production rate from some gas wells have been kept limited assuming that sand production may trigger although the well formation has sufficient strength to prevent sand production. The main objective of this project is to determine the Critical Drawdown Pressure of a gas well which is the maximum difference between reservoir pressure and bottom-hole flowing pressure that the formation can withstand without sand being produced along with the formation fluid. Different existing Models have been used for this purpose and comparative analysis of those models have been performed. Based on this study, the Drucker-Prager model result meets and explain the current operating condition of the sample gas well. Hence with all its limitation, Drucker-Prager failure criterion has been considered a reliable criterion to predict sanding production when the well is in production.
Health, safety and environment (HSE) culture in Bangladesh challenges and recommendation
(Department of Petroleum and Mineral Resources Engineering (PMRE), 2016-10) Enamul Kibria, Md.; Mahmud, Shahriar
With rapid growth of national economy and strong social needs, energy has become an important strategic resource for national development. Although Bangladesh government is making great efforts to develop sustainable energy such as nuclear, coal mining, hydro power, wind and solar power etc, undoubtedly, petroleum and natural gas still dominate the national energy demands. Although technological improvements have reduced Petroleum and other industry related deaths, accidents are still too common. That’s why health, safety, and the environment culture remain among the top priorities for oil and gas industries, reflecting an attitude of zero tolerance for accidents. Petroleum (Drilling) activities can take place in diverse geological and geophysical settings, each posing unique type of challenges. However proper planning, hazard control, inspection, incident investigation and reporting can reduces the number of accidents approximately to zero. Health, Safety and Environment (HSE) should be the first priority in petroleum and mining industry for ensuring incident free operation with profitable growth in the business. Industry without practicing HSE (regulation) is very dangerous for personnel, equipment, material, property and environment. To ensure safety (incident free operation) regular inspection of project plant, rig site and necessary control measure should always be carried out by trained manpower, HSE personnel along with concerned special authorities to meet the demand of Occupational Safety and Health Administration (OSHA) standard and regulation (or OSH). Physical inspection, incident investigation and reporting, emergency planning, hazard identification, hazard control, Lockout/ Tag out (LOTO), Manual Materials Handling (MMH), hazard analysis, job safety analysis, risk analysis, stop work condition and authorities, training, employee orientation, tool safety, toolbox talk, are essential element for maintaining safety of industries. That’s why this paper about safety culture has prepared by working on all of those. Attitudes, personal education and experience, organizational training, communication mechanism and so on, all of them affect the development of a safety culture in any organization. The environment in which people work and the systems and processes in the industry also influence the safety culture. Therefore, each industry needs to consider all of these aspects in developing and nurturing a safety culture that suits the organization and the individuals within it. In any Petroleum project, concern authority need to maintain standard quality of implementation of the HSE programmed with due consideration to standing Safety rules and regulations. The project may be considered viable from the Health, Safety and Environmental point of view and therefore be considered for implementation by appropriate authorities. It has been recommended that, Health, Safety and Environmental clearance would be issued in favor of petroleum project for execution of the project as incident free operation in future. The research and conclusions and recommendations of this report could provide useful organizational safety culture concept to promote safety performance and safety management in Bangladesh.
Leveraging RLHF with Instruction Fine-tuning for Improving LLM Response in Bangla Conversations
(Department of Computer Science and Engineering(CSE), Islamic University of Technology(IUT), Board Bazar, Gazipur-1704, Bangladesh, 2024-11-30) Rahman, Tahmid; Mahmud, Shahriar; Nasrum, Nur
In the realm of Bangla conversational agents, this research endeavors to elevate the responsiveness of Large Language Models (LLMs) through the synergistic ap plication of Reinforcement Learning from Human Feedback (RLHF) and instruc tion fine-tuning. The primary objectives encompass the creation of a domain specific Bangla conversational dataset, an evaluation of existing LLMs using an instruction-tuned dataset, and the introduction of a novel human-centric bench marking framework. This work uses a multi-step process to improve the effi cacy of Large Language Models (LLMs) in the context of Bangla Conversational Agents. The technique consists of many key steps, each of which contributes to the refining and optimization of model performance. To address the shortage of domain-specific datasets for Bangla Conversational Agents, we start by construct ing a Bangla Conversational Dataset. The dataset is then fine-tuned with the use of an Instruction-Tuned Format. This structuring makes the data more suitable for training language models, allowing them to better comprehend and respond to precise commands in the Bangla conversational environment. Existing LLMs go through the next phase of our procedure, Supervised Fine-Tuning (SFT), us ing the instruction-tuned dataset. This fine-tuning procedure guarantees that the models are tailored to the variety and complexity of Bangla talks, maximizing their performance in accordance with the dataset’s unique instructions. Following fine-tuning, we do a detailed examination and comparison of the LLM models. This stage gives insight into the effectiveness of the fine-tuned models and enables the selection of the most promising candidate. This iterative procedure entails modifying the model with human feedback to improve its performance in a more dynamic and sophisticated way.
Print Feasibility study of intelligent pigging in the ring main transmission line of Chattogram
(Department of Petroleum and Mineral Resources Engineering, BUET 2019, 2019-03-09) Refat Noushad Bhuiya, Muhammad; Mahmud, Shahriar
The project work is intended to establish the appropriateness of intelligent pigging as a single integrity assessment tool for the ring main natural gas transmission line of Chattogram city. For the purpose of the project work, previous cleaning pigging operations of the ring main line were reviewed. Significant portions where important structures/installations had grown up were identified. Records and information regarding the physical attributes and operational issues were collected from site visits, official documents and personal communications. Calculations were done to identify the limit of gas withdrawal rate at different operating pressures within the practiced speed limit of an intelligent pig. Calculations done are indicative of the ability of the ring main to allow an intelligent pig run. Alternates of intelligent pigging were checked for their appropriateness for adopting as an assessment tool for the ring main line. Review of previous cleaning piggings indicates existence of metal loss issue within the ring main line. This phenomenon is widely addressed by magnetic flux leakage (MFL) technology based metal loss intelligent pigging operation. An MFL based intelligent pigging will cost Tk 1,939.61 /meter on an average. Immediate rehabilitation of the ring main line would require Tk 39,518.76/meter, Tk 75,813.13/meter, Tk 46,839.52/meter and Tk 40,822.91/meter for segment 1, segment 2, segment 3 and segment 4 respectively. Assuming yearly 10% flat price escalation, these costs would be Tk 63,645.36/meter, Tk 1,22,097.80/meter, Tk 75,435.52/meter and Tk 65,745.71/meter for the segments mentioned above after five years. A qualitative risk analysis indicates that the intelligent pigging of the ring main line is a medium risk project.