Debotyam Maity

Novel hybrid artificial neural network based autopicking workflow for passive seismic data

Microseismic monitoring is an increasingly common geophysical tool to monitor the changes in the subsurface. Autopicking involving phase arrival detection is a common element in microseismic data processing schemes and is necessary for accurate estimation of event locations as well as other workflows such as tomographic or moment tensor inversion, etc. The quality of first arrival picking is dependent on the actual seismic waveform, which in turn is related to the near surface and subsurface structure, source type, noise conditions, environmental factors, and monitoring array design, etc. We have developed a new hybrid autopicking workflow which makes use of multiple derived attributes from the seismic data and combines them within an artificial neural network framework. An evolutionary algorithm scheme is used as the network training algorithm. The autopicker has been tested and its applicability has been validated using a synthetically modelled seismic source, with promising results. In this work, we share the basic workflow and different attributes that have been tested with this algorithm for a synthetic data set to provide a framework for independent implementation, use and validation. We also compare the results obtained using the new neural network based autopicking routine with very robust contemporary autopicking algorithms in use within the industry. Key words: automatic picking, neural network, data processing.

An integrated methodology for sub-surface fracture characterization using microseismic data: A case study at the NW Geysers

Geothermal and unconventional hydrocarbon reservoirs are often characterized by low permeability and porosity. So, they are difficult to produce and require stimulation techniques, such as thermal shear deactivation and hydraulic fracturing. Fractures provide porosity for fluid storage and permeability for fluid movement and play an important role in production from this kind of reservoirs. Hence, characterization of fractures has become a vitally important consideration in every aspect of exploration, development and production so as to provide additional energy resources for the world. During the injection or production of fluid, induced seismicity (micro-seismic events) can be caused by reactivated shears created fractures or the natural fractures in shear zones and faults. Monitoring these events can help visualize fracture growth during injection stimulation. Although the locations of microseismic events can be a useful characterization tool and have been used by many authors, we go beyond these locations to characterize fractures more reliably. Tomographic inversion, fuzzy clustering, and shear wave splitting are three methods that can be applied to microseismic data to obtain reliable characteristics about fractured areas. In this article, we show how each method can help us in the characterization process. In addition, we demonstrate how they can be integrated with each other or with other data for a more holistic approach. The knowledge gained might be used to optimize drilling targets or stimulation jobs to reduce costs and maximize production.

Artificial Neural Network Based Autopicker For Micro-earthquake Data

A large number of autopickers are in use today to detect phase arrivals for microseismograms. The most widely used method is the Short term averaging / Long term averaging (STA/LTA) algorithm (Allen, 1978) and its modifications (Baer, et al., 1987). Methods based on abrupt changes in different attributes such as energy as used in Coppens method (Coppens, 1985) and modified Coppens method (Sabbione, et al., 2010) , the Modified Energy Ratio (MER) method (Wong, et al., 2009), statistical attributes such as frequency and higher order statistics such as skewness & kurtosis (Saragiotis, et al., 2002), or combining energy and frequency characteristics of the local maxima distribution (LMD) (Panagiotakis, et al., 2008) have all been proposed over time. Most of this procedure is effective with high signal-to-noise ratio (SNR). However, with low SNR, there is always a possibility of the arrival not being detected by these methods. The quality of the first arrival picking is related to the nearand subsurface structure, source type, and SNR conditions. Any of these methods or other available methods could be applicable in some specific areas and not in others based on the mentioned conditions. Finding a robust method to work under most circumstances is the challenge.

Neuro-evolutionary event detection technique for downhole microseismic surveys

Abstract Recent years have seen a significant increase in borehole microseismic data acquisition programs associated with unconventional reservoir developments such as hydraulic fracturing programs for shale oil and gas. The data so acquired is used for hydraulic fracture monitoring and diagnostics and therefore, the quality of the data in terms of resolution and accuracy has a significant impact on its value to the industry. Borehole microseismic data acquired in such environments typically suffer from propagation effects due to the presence of thin interbedded shale layers as well as noise and interference effects. Moreover, acquisition geometry has significant impact on detectability across portions of the sensor array. Our work focuses on developing robust first arrival detection and pick selection workflow for both P and S waves specifically designed for such environments. We introduce a novel workflow for refinement of picks with immunity towards significant noise artifacts and applicability over data with very low signal-to-noise ratio provided some accurate picks have already been made. This workflow utilizes multi-step hybrid detection and classification routine which makes use of a neural network based autopicker for initial picking and an evolutionary algorithm for pick refinement. We highlight the results from an actual field case study including multiple examples demonstrating immunity towards noise and compare the effectiveness of the workflow with two contemporary autopicking routines without the application of the shared detection/refinement procedure. Finally, we use a windowed waveform cross-correlation based uncertainty estimation method for potential quality control purposes. While the workflow was developed to work with the neural network based autopicker, it can be used with any other traditional autopicker and provides significant improvements in pick detection across seismic gathers.

Reservoir Characterization of an Unconventional Reservoir by Integrating Microseismic, Seismic, and Well Log Data

Varied data types including geophysical data as well as well logs have been used frequently in the past to characterize reservoirs. However, the use of microseismic data as a potential source of useful information and its integration with conventional seismic data for reservoir characterization is an area of opportunity where properties predicted from the microseismic data can be used as a vital source of information which can then be tied with the overall characterization scheme in a seamless manner. In this paper we discuss the characterization scheme followed for an unconventional reservoir associated with a promising geothermal prospect. The field involves microseismic data acquisition being done continually as part of the field monitoring operations and extensive well control due to the presence of large number of injection and production wells and finally a 3D conventional seismic survey done to try and better define the reservoir. We have applied an integrated approach with these data sources in order to better characterize the reservoir in question using novel data analysis schemes where necessary to get optimum results. The approach shared in this paper can be applied to any type of reservoir setting with microseismic, seismic and well log data being available. What we present is a workflow to integrate these data types to generate useful property predictions including important rock property estimates with the aim of obtaining useable reservoir property maps to aid in reservoir development. This approach shows how a modest data acquisition program can still lead to useful characterization of reservoirs particularly with the inclusion of microseismic data in the workflow. We have used novel methods as part of our workflow including multi-attribute analysis, geostatistical techniques and soft computing techniques such as ANN based property prediction and mapping which are discussed in brief.

Novel fracture zone identifier attribute using geophysical and well log data for unconventional reservoirs

AbstractWe have characterized a promising geothermal prospect using an integrated approach involving microseismic monitoring data, well logs, and 3D surface seismic data. We have used seismic as well as microseismic data along with well logs to better predict the reservoir properties to try and analyze the reservoir for improved mapping of natural and induced fractures. We used microseismic-derived velocity models for geomechanical modeling and combined these geomechanical attributes with seismic and log-derived attributes for improved fracture characterization of an unconventional reservoir. We have developed a workflow to integrate these data to generate rock property estimates and identification of fracture zones within the reservoir. Specifically, we introduce a new “meta-attribute” that we call the hybrid-fracture zone-identifier attribute (FZI). The FZI makes use of elastic properties derived from microseismic as well as log-derived properties within an artificial neural network framework. Temporal ...

Fracture Characterization in Unconventional Reservoirs Using Active and Passive Seismic Data With Uncertainty Analysis Through Geostatistical Simulation

This study discusses a new workflow for fracture characterization and modeling using geophysical (microseismic and 3D surface seismic) data along with independent reservoir information (such as well logs). The framework is ideally suited for unconventional environments such as shale and tight reservoirs where modern technologies such as the use of hydraulic fracturing and passive seismic monitoring allow application of the proposed workflow. The workflow involves generating geomechanical property estimates (including stress and weakness estimates) as derived from passive seismic data analysis and relevant seismic attributes derived from 3D seismic data combined using ANN based reservoir property modeling framework. The training information for the networks is generated based on a-priori information through image logs. Resolution of passive seismic derived velocity models is improved by using sequential Gaussian co-simulation by combining low resolution velocity maps high resolution seismic impedance data for phase velocity estimation. Uncertainty estimates are quantified by adequate number of realizations and associated probability density functions for fracture properties within study volume. In this paper, different properties estimated through ANN modeling have been shared. New fracture identifier (FZI) properties have been defined and the models have been used to characterize fracture zones and major discontinuities for a representative unconventional reservoir (geothermal setting) used in our study. We also share uncertainty estimates for the identified fracture zones for improved characterization. Finally fracture property estimates for the study area (derived using FZI and other properties) have been generated for future reservoir simulation studies. The proposed method allows for improved understanding of shale and other unconventional reservoirs through fracture mapping and provides a workflow for improved volumetrics of the reservoir by making use of identified properties for fracture modeling. This work validates the potential for using relatively low resolution passive seismic data for improved reservoir characterization using Geostatistical tools. It also provides a valuable framework for pseudo 4D characterization where a single 3D seismic survey can be used as the basis to characterize the reservoir in a time lapse fashion using new information collected in time through passive seismic arrays as well as new well logs being obtained within the area of interest. Introduction Use of geophysical tools for reservoir characterization and monitoring is fairly well understood and has been used extensively in recent years. Passive seismic monitoring has found applications in development of unconventional reservoirs such as geothermal systems involving injection and production of hot water or steam from the reservoir, tight gas and oil systems which require hydraulic fracturing and finally monitoring of injection wells (waste water, CO2, etc.) among others. In the field of conventional seismic, techniques such as multi-attribute analysis and integrated analysis techniques are being used extensively for reservoir characterization. While conventional seismic data is rarely available for small unconventional reservoir developments, the use of microseismic data is limited to better understanding of the fracturing process including diagnosis and volumetrics but is seldom used