The purpose of this research is to apply a new approach to identify natural fractures in wells in a hydrocarbon reservoir using resistive image logs, fractal dimension and support vector machines (SVMs). The stratigraphic sequence investigated by each well is composed of Cretaceous calcareous rocks from the Catatumbo Basin, Colombia. The box counting method was applied to image logs in order to generate a curve representing variations of fractal dimension in these images throughout each well. The arithmetic mean of fractal dimension showed values ranging from 1,70 to 1,72 at the mineralized fracture intervals, and from 1,72 to 1,76 at the open fracture intervals. Morphological classification between open and mineralized natural fractures is performed using corelogs integration in a pilot well. Fractal dimension of images along with gamma rays and resistivity logs were employed as the input dataset of a SVM model identifying intervals with natural open fractures automatically, shortly after logs acquisition and previous to its interpretation by specialists. Although final results were affected by borehole conditions and logs quality, the SVM model showed accuracy between 72,3 % and 82,2 % in 5 wells evaluated in the studied field. Keywords:Fractal dimension, resistive image logs, box counting method, natural fractures, hydrocarbon reservoir, Catatumbo basin, support vector machines (svms). RESUMENEl propósito de esta investigación es aplicar un nuevo enfoque para identificar fracturas naturales en pozos de un yacimiento de hidrocarburo utilizando registros de imágenes resistivas, dimensión fractal y máquinas de soporte vectorial (MSV). La secuencia estratigráfica alcanzada por cada pozo está compuesta por rocas calcáreas cretácicas de la Cuenca del Catatumbo, Colombia. El método del conteo de cajas se aplicó a registros de imágenes, generando una curva que representa variaciones de dimensión fractal en las imágenes a lo largo de cada pozo. La media aritmética de dimensión fractal mostró valores desde 1,70 a 1,72 en intervalos con fracturas mineralizadas y desde 1,72 a 1,76 en intervalos con fracturas abiertas. La clasificación morfológica entre fracturas naturales abiertas y mineralizadas es realizada utilizando integración núcleo-registro de un pozo piloto. La dimensión fractal de las imágenes junto con registros de rayos gamma y resistividad son empleados como datos de entrada a un modelo de MSV identificando intervalos con fracturas naturales abiertas automáticamente, poco después de adquirir los registros y previo a su interpretación por especialistas. Aunque los resultados finales están afectados por condiciones del hoyo y calidad de registros, el modelo de MSV mostró exactitud entre 72,3 % y 82,2 % en 5 pozos evaluados del campo estudiado.Palabras clave: Dimensión fractal, registros de imágenes resistivas, método del conteo de cajas, fracturas naturales, yacimiento de hidrocarburos, cuenca del Catatumbo, máquinas de soporte vectorial.

In this research two support vector machines (SVMs) and a logic function were applied to identify calcareous sections automatically in wells located in the former Barco Concession, Catatumbo Basin -Colombia. During training stages the SVMs used nuclear logs, such as neutron, photoelectric factor and gamma ray in order to differentiate calcareous from clastic sections; additionally, in this stage the fractal dimension of resistive images along with mean and variance of resistivity acquired with imaging tool (of high resolution) are employed to identify textural features of the rocks. The first SVM also employed in the training stage intervals manually interpreted of fossiliferous limestone, performed by a specialized geologist integrating core and logs information of a pilot well; during classification stage, this SVM automatically recognized intervals with fossiliferous limestone by using only logs data of any well of the field. The second SVM was also trained with nuclear logs, resistivity and fractal dimension, but in this case, with information of intervals composed of calcareous shales interbedded with limestone, recognizing automatically these rock associations during classification stage without interpretation of a geologist as input data. Finally, a logic function was applied to intervals with photoelectric factor ≥ 4 and all sections not classified by the SVMs were grouped as laminated calcareous rocks. The SVMs and logic function show accuracy of 98.76 %, 94.02 % and 94.60 % respectively in five evaluated wells and can be applied to other wells in the field that have the same dataset conditions. This methodology is dependent of the data quality and all intervals affected by poor borehole conditions should be removed in order to avoid erratic interpretations. This model must to be recalibrated in case to be applied in other fields of the basin. En esta investigación dos máquinas de vector de soporte (MVS) y una función lógica fueron aplicadas para identificar automáticamente secciones calcáreas en pozos ubicados en la antigua Concesión Barco, Cuenca de Catatumbo -Colombia. Durante etapas de entrenamiento las MVS utilizaron registros nucleares, tales como neutrón, factor fotoeléctrico y rayos gamma para diferenciar secciones calcáreas de clásticas; adicionalmente, en esta etapa la dimensión fractal de las imágenes resistivas junto con la media y varianza de la resistividad adquirida con la herramienta de imágenes (de alta resolución), son empleadas para identificar rasgos texturales de las rocas. La primera MVS también empleó durante el entrenamiento intervalos manualmente interpretados de calizas fosilíferas, realizado por un geólogo especialista integrando información de núcleo y registro de un pozo piloto; posteriormente, durante la clasificación, esta MVS automáticamente reconoció intervalos con calizas fosilíferas utilizando solamente datos de registros de cualquier pozo del campo. La segunda MVS fue entrenada con registros nucleares, resistivos y dimensión fractal, pero en este caso, también con inf...

Scale control has been a significant challenge for the management of well performance for several years in the Cusiana field in Colombia. Physical evidence of scale has been found in very few wells, but circumstantial evidence, including increases in gamma ray across productive zones and production declines inconsistent with reservoir properties has been widespread. Scale prediction, from water analysis, and the actual analysis of the limited scale samples suggested that carbonate scales were dominant, along with small quantities of barium sulphate scale. However, there was significant variation in the scale analysis, with both calcium and iron carbonates being identified as the major component. Furthermore, stimulation treatments to remove carbonate scale were not always successful and gamma ray readings often continued to increase after scale inhibition treatments designed for carbonate scales. Several theories were developed based on this background information. These included the presence of mixed scales, with the sulphate scales not being removed during stimulation treatments and theories of near wellbore evaporation due to the high GOR of the produced fluids. This evaporation of connate water could lead to higher scaling potentials for conventional scales, or even the deposition of halite and other salts, and could induce the inhibition of further connate water into the evaporated zone. In order to obtain direct evidence of the downhole scales, a multi-stage stimulation treatment was designed for a selected well. This paper describes this treatment, including operational details and data analysis. Clear evidence of both carbonate and sulphate scales downhole was found, however there was no evidence of halite deposition. It was not possible to conclude the presence of downhole iron carbonate scale, as corrosion products appeared to dominate the observed iron in flowback samples. Lessons learnt for future stimulation treatments in wells with suspected multiple scales are also discussed. Background The Cusiana field has three productive horizons, the Mirador, Barco and Guadalupe reservoirs. The majority of the production is from the Mirador formation, a quartz arenite sandstone. Scale was first suspected as a damage mechanism in the field when it was observed that an otherwise unexplained decline in production coincided with an increase in gamma ray in the producing intervals. The evidence for scale deposition was not conclusive, but there was significant circumstantial evidence that suggested that it could still be the cause of decline. Scale predictions from water analyses in each reservoir, using various models,1–2 showed that scale deposition was expected. The water analyses for each reservoir are shown in Table 1, with the associated scale predictions in Table 2. Calcium carbonate was the predominant scale predicted for all three reservoirs, with both iron carbonate and barium sulphate also being predicted in varying amounts. The quantity of scale actually deposited is clearly dependent upon both the scaling tendency and the volume of water produced. A common characteristic of the Cusiana wells was that the decline in production occurred at very low water cuts, including wells with less than 0.5% water cut. The characteristics of the quartz arenite sandstone in the Mirador formation have previously been described 3. Due to the porosity-permeability relationship for quartz arenite sandstones, see Figure 1, any reduction in effective porosity can result in a significant reduction in permeability. Consequently, small amounts of scale deposited in the near wellbore region could cause a significant production loss.