Performance assessment of five MCP models proposed for the estimation of long-term wind turbine power outputs at a target site using three machine learning techniques

•Five models based on MCP methods to estimate WTPOs at a target site are assessed.•Models are proposed which take into account air density variability.•Three machine learning techniques implemented in the models are analysed.•The models use wind turbines with blade pitch control and stall-regulated wind turbines.•Statistical hypothesis tests are used to compare the ML techniques in the best model and to compare some of the models.