Discovering hidden geothermal signatures using non-negative matrix factorization with customized k-means clustering
نویسندگان
چکیده
Discovery of hidden geothermal resources is challenging. It requires the mining large datasets with diverse data attributes representing subsurface hydrogeological and conditions. The commonly used play fairway analysis approach typically incorporates subject-matter expertise to analyze regional estimate characteristics favorability. We demonstrate an alternative based on machine learning (ML) process a dataset from southwest New Mexico (SWNM). study region includes low- medium-temperature hydrothermal systems. Several these systems are not well characterized because insufficient existing limited past explorative work. This discovers patterns relations in SWNM improve our understanding conditions energy-production obtained by applying unsupervised ML algorithm non-negative matrix factorization coupled customized k-means clustering (NMFk). NMFk can automatically identify (1) signatures characterizing analyzed datasets, (2) optimal number signatures, (3) dominant associated each signature, (4) spatial distribution extracted signatures. Here, applied 18 geological, geophysical, hydrogeological, at 44 locations SWNM. Using NMFk, we find associations within two physiographic provinces (Colorado Plateau Basin Range) sub-regions (the Mogollon-Datil volcanic field Rio Grande rift) five that differentiate between low (<90°C) medium (90-150°C)-temperature also suggests rift northern most favorable regions for future resource discovery. identified critical area. resulting model be predict their uncertainties new unexplored regions. code execute performed analyses as corresponding found https://github.com/SmartTensors/GeoThermalCloud.jl.
منابع مشابه
Discovering hierarchical speech features using convolutional non-negative matrix factorization
Discovering a representation that reflects the structure of a dataset is a first step for many inference and learning methods. This paper aims at finding a hierarchy of localized speech features that can be interpreted as parts. Non-negative matrix factorization (NMF) has been proposed recently for the discovery of parts-based localized additive representations. Here, I propose a variant of thi...
متن کاملClinical Document Clustering using Multi-view Non-Negative Matrix Factorization
Clinical document contains vital information like symptom names, medication names, age, gender and some demographical information. These information can be used for giving quick relief from a disease. In existing system, they had built a system for clustering symptom names and medication names using Multi-View Non-Negative Matrix Factorization. While considering the clinical documents the facto...
متن کاملMulti-Task Clustering using Constrained Symmetric Non-Negative Matrix Factorization
Researchers have attempted to improve the quality of clustering solutions through various mechanisms. A promising new approach to improve clustering quality is to combine data from multiple related datasets (tasks) and apply multi-task clustering. In this paper, we present a novel framework that can simultaneously cluster multiple tasks through balanced Intra-Task (within-task) and Inter-Task (...
متن کاملMusical Onset Detection by means of Non-Negative Matrix Factorization
In this paper, we propose a musical onset detection method, with reference to polyphonic piano music. This method operates on a frame-by-frame basis and exploits a suitable time-frequency representation of the audio signal. The solution proposed consists of an onset detection algorithm based on Short-Time Fourier Transform (STFT) and Non-Negative Matrix Factorization (NMF). To validate this met...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Geothermics
سال: 2022
ISSN: ['1879-3576', '0375-6505']
DOI: https://doi.org/10.1016/j.geothermics.2022.102576