Before a wind energy system is installed, modelling and prediction of wind speed plays an important role during wind energy planning stage. The short term wind speed of Katsina was measured at 5 m elevation at different locations using hand held digital rotating cup. In order to obtain long term wind speed, a prediction model using artificial neural network (ANN) was developed. The model was implemented using Matlab/Simulink 2019, which consists of 3 layers with back propagation algorithm. The model has 5 inputs which were selected using trial and error techniques. The model was trained, tested and validated using available 1 year data obtained from Katsina state agricultural and rural development agency KATARDA in 34 ground stations situated in each local government. After, the development of the ANN wind speed prediction model, the best architecture has root mean square error (RMSE) of 8.9%, mean biased error (MBE) 4.3% while the optimum correlation value of 0.9380 was realized. The predicted annual wind speeds are in the range of 0.9-13.1 m/s and  the yearly evaluated  wind speed are 3.2, 2.6 and 4.7 m/s for Katsina central, Katsina south and Katsina north respectively. The wind speed map of the study area was developed; the map shows the locations wind high, moderate and low wind energy potential. The map will serve as a reference for wind energy development in the state and will provide the policy makers and potential investors as a blue print for wind energy development.