In the wind industry, a power curve refers to the functional relationship between the power output generated by a wind turbine and the wind speed at the time of power generation. Power curves are used in practice for a number of important tasks including predicting wind power production and assessing a turbine’s energy production efficiency. Nevertheless, actual wind power data indicate that the power output is affected by multiple covariates such as wind direction, air density, humidity,and wind shears. Furthermore, the power density changes drastically throughout the covariate space. We present a Bayesian conditional density estimation method which can accommodate multiple covariates as well as sharp changes of the density function.The method is based on Bayesian partition model framework and utilize logistic Gaussian process within each partition. The partition is created using a Voronoi tessellation and is learned from the data using a reversible jump Markov chain Monte Carlo algorithm.he method has desirable consistency properties.In application to windmill data, the model successfully estimates the partition structure and the conditional distribution of the power output.