Non-stationary statistical modelling of wind speed: A case study in eastern Canada

The assessment of wind energy potential is generally based on the analysis statistical distribution observed speed short time resolution. Record periods observational data used in practice at sites interest are often very short, ranging from a few months to years. Predictions such small record likely be biased as it recognized that subject important interannual variability and long-term trends. Large-scale atmospheric circulation patterns have an influence speed. Their predictable nature can make them useful for prediction during lifetime farm projects. This feature not exploited practice. It proposed this study introduce predictors non-stationary models. approach allows development predictions conditionally state predictors. here indices account temporal index trend. was applied hourly selected meteorological stations province Québec (Canada). 20 with long over 30 years were used. most identified area North-Atlantic Oscillation (NAO) winter season Pacific North American (PNA) spring season. Results indicate annual goodness-of-fit case improved average when model compared stationary model. potentially projected farms using forecasts