The performance evaluation and design optimisation of multiple fractured horizontal wells in tight reservoirs

Multiple fractured horizontal wells (MFHWs) are recognised as the most efficient stimulation technique to improve recovery from unconventional gas assets. Although multistage fracture treatment has been very successful in stimulating these reservoirs, very little work has been done on multi-stage design optimisation. In most of the published works, the improved MFHWs design is recommended to be determined by sensitivity analysis of one variable while keeping all the other variables fixed. Several researches suggested that this optimisation should be typically performed based on economic objectives such as Net Present Value (NPV). This paper initially describes the results of an exercise that uses statistical algorithms coupled with numerical reservoir simulations to evaluate the simultaneous impacts of important pertinent parameters on the performances of different MFHW designs at various production periods. It is shown that the impact of the individual parameter, quantified by Spearman's rank correlation coefficients technique, on different objective functions e. g. total gas production during the production period, varies depending on the governing flow regimes. For example, it is demonstrated that the impact of fracture length on the performance of MFHWs decreases over the production time while the number of fractures exhibits almost a fixed effect. It was also shown that the general trend of the importance of parameters on productivity index (PI) is similar to those observed for some of other objective functions including total gas production and NPV. In addition, these results confirm the applicability of available well productivity models developed for the early, middle and boundary dominated flow conditions to optimise the design of MFHWs in tight reservoirs. The result of the study confirms provided maximising a desired objective in the long term (longer than the time to reaching the compound linear flow) is targeted; the pseudo-steady state productivity indices models are appropriate to be used for the design optimisation of MFHWs. Otherwise, if a shorter-term objective is targeted, this optimisation could be performed based on appropriate productivity index models available for the early or middle production periods. These results are also confirmed by performing reservoir simulation-based optimisation of the MFHWs design using the genetic algorithm approach for various cases. This work provides a general, fit for purpose set of guidelines, suitable for an improved well design of MFHWs in tight reservoirs. In addition, a new and easily to use workflow based on the productivity index equations is developed to optimise MFHWs design in tight gas reservoirs for a chosen targeted time while considering the practical limits and economics.