One of the newest alternatives recommended as suitable substitution for the conventional perforating methods, is laser perforation. This is because of its superiorities over the current shaped charge methods that the most important one is increasing the permeability considerably and no need to have costly post-perforation operations to reduce the effect of formation damage caused by perforation. To improve the efficiency of laser perforation in sandstone, the effectiveness of the important parameters should be taken into account. In this paper, a neural network approach has been used for backward elimination sensitivity analysis of the effective parameters during laser perforation in sandstone. For this purpose firstly a feed-forward with back propagation neural network was developed that predicted the volume removed due to the effective parameters like laser power, lasing time, purging system, saturation and confining pressure. The data is related to around 110 laser perforation laboratory tests on sandstone core samples. Then sensitivity analysis was done and finally the effectiveness of each parameter was determined. This sensitivity analysis can be used to optimize the laser perforation in sandstone in both technical and economical aspects.
Perforating oil and gas wells is a process of creating tunnels through the cemented casing to let the formation fluid flow into the well. The current explosive shaped charge perforation methods that are applied, is accompanied with formation damage and reducing permeability. It has been found out by the petroleum industry that laser technology has the potentiality for being an efficient tool to be applied as a prominent substitution for conventional oil and gas production. Due to the mechanism of laser rock removal which is thermal and without any kind of destruction, applying high power lasers for perforating oil and gas wells leads to the remarkable permeability increase even up to 171% for sandstone samples [2]. In this way, there is a term that has a direct effect on the efficiency of laser perforation in rocks which is specific energy (SE). SE is defined as the amount of energy required to remove a unit volume of rock. So, for having a successful laser perforation with the maximum efficiency, SE value should be minimal [3]. In recent years, there has been a steady growth in applying Artificial Neural Network (ANN). ANNs have the potential for solving many of the challenging and complex problems in many fields.
Laser/sandstone interaction is due to a phase transformation of quartz from α to β phase at 600°C that leads to 0.8% of expansion in volume which helps the spallation process although the produced gaseous materials originated from some mineral vaporization, is also effective [4]. laser/rock interaction is affected by several parameters such as laser power, lasing time, purging system, rock saturation and confining pressure [5]. These parameters influence the SE value in different ways, hence all of them should be taken into account for getting a viewpoint about the removed volume while SE and removed volume are proportional.
