.png){kind=icon}
.png){kind=icon}
Return
GLDM Algorithm for Big Data (SCADA) Wind Speed Modelling
.png){kind=icon}
PRE
AI
DOI：10.31803/tg-20240419205218.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This study enhances wind speed forecasting by implementing the second-order Generalized Least Deviation Method (GLDM), focusing on wind turbines in Turkey. The research aims to improve predictive accuracy and operational efficiency in renewable energy systems through advanced mathematical modeling in meteorology. The GLDM, utilizing a quasilinear recurrence equation, addresses the inherent non-linearity and variability of wind speed data. By applying the method to extensive SCADA data, this study minimizes residuals in nonlinear big data environments, integrating both linear and nonlinear components to refine predictions. A critical aspect of this research is the comparison between the second-order GLDM and traditional forecasting models, including statistical methods and machine learning approaches. The results demonstrate the superior performance of GLDM, as indicated by lower prediction errors and greater accuracy across key metrics. The study also underscores the importance of GLDM coefficients, alpha i, in improving predictive capabilities. The findings advocate for the adoption of GLDM in wind speed forecasting, highlighting its potential to significantly enhance wind energy management through increased accuracy. This study also sets a precedent for broader applications of advanced mathematical models in environmental science, illustrating the effectiveness of GLDM in optimizing renewable energy resources.
Keywords:
atmospheric dynamics
Generalized Least Deviation Method (GLDM)
renewable energy optimization
SCADA Data Analysis
statistical model validation
wind speed forecasting
wind turbine efficiency
Journal
T
IF:
0.7
Papers:
48
Citations:
336