Spatio-temporal probabilistic forecasting of wind power output
A method for forecasting wind power output of a target wind farm. The method includes normalizing, wind power output data for each wind farm of a group of wind farms, based, at least in part, on a respective installed capacity; transforming, the normalized power output data to yield transformed normalized wind power output data. Fitting, by the temporal module, each temporal model of at least one temporal model to model input data for each wind farm. The model input data corresponds to normalized wind power output data or transformed normalized wind power output data. The method further includes fitting, by a spatial module, a DVINE copula model for the group of wind farms, based, at least in part, on at least one residual value. Each residual value is determined based, at least in part on a selected fitted temporal model for each wind farm in the group.
1. A method for forecasting wind power output of a target wind farm, the method comprising:
normalizing, by a preprocessor module, wind power output data for each wind farm of a group of wind farms, based, at least in part, on a respective installed capacity;
transforming, by a temporal module, the normalized wind power output data to yield transformed normalized wind power output data;
fitting, by the temporal module, each temporal model of at least one temporal model to model input data for each wind farm, the model input data corresponding to normalized wind power output data or transformed normalized wind power output data;
fitting, by a spatial module, a DVINE copula model for the group of wind farms, based, at least in part, on at least one residual value, each residual value determined based, at least in part on a selected fitted temporal model for each wind farm in the group;
validating, by a validation module, each model of the at least one model, the validating comprising at least one of a univariate validation technique and a multivariate validation technique;
selecting, by the validation module, at least one selected temporal model based, at least in part, on a comparison of validation results; and
predicting a wind power output for the target wind farm utilizing the selected model.
2. The method of claim 1 , further comprising adjusting, by the temporal module, at least one model parameter based, at least in part, on a model residual and based, at least in part on a support vector regression (SVR) hybridization.
3. The method of claim 1 , wherein the temporal model is selected from the group comprising Naïve—ARIMAX, Transformed ARIMAX, Transformed dynamic autoregressive (AR), quantile AR, dummy and persistence, the ARIMAX models corresponding to an autoregressive integrated moving average and comprising an additive cyclic feature and the transformed models corresponding to application of a υ-logit transform.
4. The method of claim 1 , wherein the univariate validation technique is selected from the group comprising a probability integral transformation, an exceedance calibration, a marginal calibration, a continuous-ranked probability score and the multivariate validation technique is selected from the group comprising evaluating a pre-rank function and determining an energy score.
5. The method according to claim 1 , further comprising acquiring, by the preprocessor module, wind farm data, the wind farm data comprising the power output data, geographic location data, and weather data, the weather data comprising one or more of wind direction, wind speed, air temperature, air density and air pressure.
6. A wind power output forecasting system for a target wind farm, the system comprising:
a preprocessor module configured to normalize wind power output data for each wind farm of a group of wind farms, based, at least in part, on a respective installed capacity;
a temporal module configured to transform the normalized wind power output data to yield transformed normalized wind power output data and to fit each temporal model of at least one temporal model to model input data for each wind farm, the model input data corresponding to normalized wind power output data or transformed normalized wind power output data;
a spatial module configured to fit a DVINE copula model for the group of wind farms, based, at least in part, on at least one residual value, each residual value determined based, at least in part on a selected fitted temporal model for each wind farm in the group; and
a validation module configured to validate each model of the at least one model, the validating comprising at least one of a univariate validation technique and a multivariate validation technique, configured to select at least one selected temporal model based, at least in part, on a comparison of validation results, and configured to predict a wind power output for the target wind farm utilizing the selected model.
7. The system of claim 6 , wherein the temporal module is further configured to adjust at least one model parameter based, at least in part, on a model residual and based, at least in part on a support vector regression (SVR) hybridization.
8. The system of claim 6 , wherein the temporal model is selected from the group comprising Naïve—ARIMAX, Transformed ARIMAX, Transformed dynamic autoregressive (AR), quantile AR, dummy and persistence, the ARIMAX models corresponding to an autoregressive integrated moving average and comprising an additive cyclic feature and the transformed models corresponding to application of a υ-logit transform.
9. The system according to claim 6 , wherein the univariate validation technique is selected from the group comprising a probability integral transformation, an exceedance calibration, a marginal calibration, a continuous-ranked probability score and the multivariate validation technique is selected from the group comprising evaluating a pre-rank function and determining an energy score.
10. The system of claim 6 , wherein the preprocessor module is further configured to acquire wind farm data, the wind farm data comprising the power output data, geographic location data, and weather data, the weather data comprising one or more of wind direction, wind speed, air temperature, air density and air pressure.
11. A computer readable storage device having stored thereon instructions that when executed by one or more processors result in the following operations comprising:
normalizing wind power output data for each wind farm of a group of wind farms, based, at least in part, on a respective installed capacity;
transforming the normalized wind power output data to yield transformed normalized wind power output data;
fitting each temporal model of at least one temporal model to model input data for each wind farm, the model input data corresponding to normalized wind power output data or transformed normalized wind power output data;
fitting a DVINE copula model for the group of wind farms, based, at least in part, on at least one residual value, each residual value determined based, at least in part on a selected fitted temporal model for each wind farm in the group;
validating each model of the at least one model, the validating comprising at least one of a univariate validation technique and a multivariate validation technique;
selecting at least one selected temporal model based, at least in part, on a comparison of validation results; and
predicting a wind power output for a target wind farm from the group of wind farms utilizing the selected model.
12. The device of claim 11 , wherein the instructions that when executed by one or more processors result in the following additional operations comprising: adjusting at least one model parameter based, at least in part, on a model residual and based, at least in part on a support vector regression (SVR) hybridization.
13. The device of claim 11 , wherein the temporal model is selected from the group comprising Naïve—ARIMAX, Transformed ARIMAX, Transformed dynamic autoregressive (AR), quantile AR, dummy and persistence, the ARIMAX models corresponding to an autoregressive integrated moving average and comprising an additive cyclic feature and the transformed models corresponding to application of a υ-logit transform.
14. The device of claim 11 , wherein the univariate validation technique is selected from the group comprising a probability integral transformation, an exceedance calibration, a marginal calibration, a continuous-ranked probability score and the multivariate validation technique is selected from the group comprising evaluating a pre-rank function and determining an energy score.
15. The device of claim 11 , wherein the instructions that when executed by one or more processors result in the following additional operations comprising: acquiring wind farm data, the wind farm data comprising the power output data, geographic location data, and weather data, the weather data comprising one or more of wind direction, wind speed, air temperature, air density and air pressure.