Wind Speed Predictions via ARIMAX-ANN Hybrid Model for Brazilian Northeast

A hybrid model predicts monthly and hourly average wind speeds for regions of the Brazilian Northeast. The model combines ARIMAX and ANN to incorporate linear and nonlinear features, correlating pressure, temperature, and precipitation with wind speed. The author, Henrique do Nascimento Camelo, reports a mean percentage error of approximately 5.0% and a Nash-Sutcliffe efficiency coefficient of 0.96.

Use Cases

• Forecasting wind speed for energy production planning based on the described meteorological correlations.
• Evaluating hybrid linear-nonlinear modeling techniques based on the ARIMAX-ANN architecture described.
• Analyzing seasonal wind patterns in the Brazilian Northeast based on the model's reported capacity to represent seasonality.

Strengths

• Model performance is quantified with a reported mean percentage error of approximately 5.0%.
• Model accuracy is validated with a reported Nash-Sutcliffe efficiency coefficient of 0.96.
• The hybrid approach is designed to capture both linear and nonlinear features common in time series.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect geographic bias inherent to the specific focus on the Brazilian Northeast.

Provenance

Source

Henrique do Nascimento Camelo via paperswithcode

Collection Method

Model development and evaluation described in a research paper abstract.

Geography

Brazilian Northeast