Dataset Information

Dataset of distribution transformers for predictive maintenance.

ABSTRACT: In electricity sector is possible to collect large quantities of data that contain information on relevant processes and events that occur in a given period. It gives a knowledge of the different operation conditions of the electrical network and its components. Through the treatment and analysis of these data is possible to propose market, cost reduction, reduction of failures and repairs in machines and inventory decrease strategies. Grid operator can implement strategies to improve indicators of reliability and quality of service. From a maintenance point of view, the equipment operating time is a relevant aspect to identify and solve failures without service suspensions. This paper aims to show distribution transformers failures characteristics data using historical data collected by the grid operator (Compañia Energética de Occidente) at Cauca Department (Colombia), under the cooperation of the Universidad del Cauca and Universidad del Valle. The dataset could be helpful to researchers and data scientists who use machine learning to develop applications that help engineers in predictive maintenance.

SUBMITTER: Bravo M DA

PROVIDER: S-EPMC8521452 | biostudies-literature | 2021 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Dataset of distribution transformers for predictive maintenance.

Bravo M Diego-A DA Alvarez Q Laura-I LI Lozano M Carlos-A CA

Data in brief 20211012

In electricity sector is possible to collect large quantities of data that contain information on relevant processes and events that occur in a given period. It gives a knowledge of the different operation conditions of the electrical network and its components. Through the treatment and analysis of these data is possible to propose market, cost reduction, reduction of failures and repairs in machines and inventory decrease strategies. Grid operator can implement strategies to improve indicators ...[more]

PMID: 34703850

Similar Datasets

Project description:BackgroundPredictive analytic models, including machine learning (ML) models, are increasingly integrated into electronic health record (EHR)-based decision support tools for clinicians. These models have the potential to improve care, but are challenging to internally validate, implement, and maintain over the long term. Principles of ML operations (MLOps) may inform development of infrastructure to support the entire ML lifecycle, from feature selection to long-term model deployment and retraining.ObjectivesThis study aimed to present the conceptual prototypes for a novel predictive model management system and to evaluate the acceptability of the system among three groups of end users.MethodsBased on principles of user-centered software design, human-computer interaction, and ethical design, we created graphical prototypes of a web-based MLOps interface to support the construction, deployment, and maintenance of models using EHR data. To assess the acceptability of the interface, we conducted semistructured user interviews with three groups of users (health informaticians, clinical and data stakeholders, chief information officers) and evaluated preliminary usability using the System Usability Scale (SUS). We subsequently revised prototypes based on user input and developed user case studies.ResultsOur prototypes include design frameworks for feature selection, model training, deployment, long-term maintenance, visualization over time, and cross-functional collaboration. Users were able to complete 71% of prompted tasks without assistance. The average SUS score of the initial prototype was 75.8 out of 100, translating to a percentile range of 70 to 79, a letter grade of B, and an adjective rating of "good." We reviewed persona-based case studies that illustrate functionalities of this novel prototype.ConclusionThe initial graphical prototypes of this MLOps system are preliminarily usable and demonstrate an unmet need within the clinical informatics landscape.

Project description:Shallow groundwater (GW), defined as the water table of unconfined or perched aquifers that is near enough to the land surface to influence the vadose zone and the surface soil moisture, impacts land surface water, energy, and carbon cycles by providing additional moisture to the root zone via capillary fluxes. Although the interactions of shallow GW and the terrestrial land surface are widely recognized, incorporating shallow GW into the land surface, climate, and agroecosystem models is not yet possible due to the lack of groundwater data. Groundwater systems are affected by various factors, including climate, land use/land cover, ecosystems, GW extractions, and lithology. Although GW wells are the most direct and accurate way of monitoring water table depths at point scales, upscaling GW levels from point scale to areal or regional scale poses significant challenges. Here, we provide high spatiotemporal resolution global maps of the terrestrial land surface areas influenced by shallow GW from mid-2015 to 2021 (a separate NetCDF file for each year) in a 9 km spatial and daily temporal resolution. We derived this data from NASA's Soil Moisture Active Passive (SMAP) mission spaceborne soil moisture observations with a temporal resolution of 3 days and approximately 9 km grid resolution. This spatial scale corresponds to SMAP's "Equal Area Scalable Earth" (EASE) grids. The central assumption is that the monthly moving average of soil moisture observations and their coefficient of variation are sensitive to shallow GW regardless of the prevailing climate. We process the Level-2 enhanced passive soil moisture SMAP (SPL2SMP_E) product to detect shallow GW signals. The presence of shallow GW data is calculated by an ensemble machine learning model, which is trained using simulations from a variably saturated soil moisture flow model (Hydrus-1D). The simulations span various climates, soil textures, and lower boundary conditions. The spatiotemporal distribution of shallow GW data based on SMAP soil moisture observations is provided for the first time with this dataset. The data are of value in a wide variety of applications. The most direct use is in climate and land surface models as lower boundary conditions or as a diagnostic tool to verify model results. Some other applications may include flood risk analyses and regulation, identifying geotechnical issues such as shallow GW-triggered liquefaction, global food security, ecosystem services, watershed management, crop yield, vegetation health, water storage trends, and tracking mosquito-borne diseases by identifying wetlands, among other applications.

Dataset Information

Dataset of distribution transformers for predictive maintenance.

Publications

Dataset of distribution transformers for predictive maintenance.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets