Low-permeability, coal seam gas (CSG) wells have been the subject of laboratory research and modelling studies over the last decade, particularly focusing on the pressure-dependent permeability (PDP) behaviour of coals. These research efforts have progressed diagnostic methods to identify and quantify PDP and provide practical technologies to counter these effects.
Firstly, machine learning methods based on drilling and historical well-test data can provide insight into the range of coal permeability during drilling. Next, the process of history-matching the after-closure pressures from diagnostic fracture injection tests (DFITs) using reservoir simulators can determine best-fit values for fracture compressibility, a key parameter for reservoir models. Finally, these data along with DFIT reservoir pressure and permeability data can inform the decision-making process as to the most applicable completion strategy and aid developmental planning.
For areas where vertical or surface-to-inseam (SIS) wells have been unsuccessful, new hydraulic fracturing technologies have been developed to enhance the stimulated reservoir volume (SRV) in coals, using horizontal wells with multi-stage hydraulic fracturing in excess of 20 stages. Recent laboratory and modelling of micro-proppants has extended prior laboratory and modelling studies and provided insight into proppant transport, embedment, and screen-out behaviour. These well stimulation technologies can be co-applied in new or existing CSG fields and are suitable for areas where overlapping tenements limit conventional, steel-based completion strategies.
In conclusion, this paper will bring the key findings of these studies together in a cohesive framework and provide the workflows to implement these technologies for better productivity in low-permeability coals.

In this study, we present a novel inversion model to infer the subsurface fractures from the ground surface tilt measurements. A machine-learned surrogate forward model based on conditional Generative Adversarial Networks (cGAN) is developed, tested and utilised to predict ground surface tilts (displacement gradients) induced by pressurised fractures in the subsurface. The results show that the surrogate forward model satisfactorily predicts the tilt vector at the ground surface induced by the prescribed pressurised fracture. The model prediction was satisfactorily for complex cases of multiple fractures located at different depths, while the model itself was originally trained for a set of single fractures located at a prescribed constant depth. Then, an inversion algorithm based on Bayesian method was implemented to find the optimised solution (pressurised fracture) for a given ground surface data (an array of tilt data) using the trained surrogate forward model. The results show how the inversion using the surrogate model is reliable and much faster than the finite element model used for creating the training data.

Welspun Corp Limited (WCL) is engaged in manufacturing and supply of SAWL carbon steel pipes for oil and gas transportation since more than 2 decades. The line pipe dimensions are very critical for offshore pipes particularly ovality (out-of-roundness) at pipe ends, skew (out-of-straightness) and diameter at pipe ends, as these are key contributors to Hi-Lo and fit-up of pipes onboard offshore lay vessels.
WCL has installed APDMS & RPEMS at our Anjar SAWL, which can measure 19 & 9 parameters with the help of 72 & 6 measurement sensors respectively. APDMS & RPEMS is an integrated optical measurement system, using laser triangulation, coupled with SAP integration, its own calibration modulus and provides an accurate solution to all pipe measurements. At each pipe end, 1000 readings minimum are generated at a particular location which is shared with Allseas, helping them during fit-up, resulting in excellent weld quality, increased productivity, reduction in cost and meeting timelines. Also, taking into the equipment accuracy for each parameter to be measured by APDMS or RPEMS, the negative tolerance was deducted from the acceptance criteria for each parameter to be met as per specification requirements.
WCL has currently executed offshore projects with close dimensional tolerances measured using the APDMS for one project and RPEMS for other project for Allseas, who is going to lay the pipe offshore. Present paper aims to show the significance of using APDMS/ RPEMS based on the data from the executed project for SAWL pipes complimenting it with experience of Allseas.