Stochastic modelling of wind speed over northern states in Nigeria

Abstract

This paper presents a stochastic modeling of wind speed over sixteen (16) Northern states in Nigeria using thirty-seven years (1984-2020) wind speed real time data. A Markov Chain Model was developed for the monthly wind speed state across study locations. In order to obtain the Markov chain transitional probabilities, the wind speed data was categories into various states using the Beaufort wind scale. It was observed that only the first four description of wind speed state A, B, C and D exist in the study locations. Uniform random states were also formed by generating uniform random number. The comparison of monthly simulated and actual wind speed state clearly shows that the model simulated over six months correctly across study locations except Niger. Given a current wind speed state conditions, the stochastic models available in this paper can be adapted to generate future wind speed state condition. The understanding of wind speed state helps in wind turbine design and selection of wind farm sites for wind energy generation.