No abstract

During the summer months at the U.S. Army Test and Evaluation Command's (ATEC) White Sands Missile Range (WSMR), forecasting thunderstorm activity is one of the primary duties of the range forecasters. The safety of personnel working on the range and the protection of expensive test equipment depend critically on the quality of forecasts of thunderstorms and associated hazards, including cloud-toground lightning, hail, strong winds, heavy rainfall, flash flooding, and tornadoes. The National Center for Atmospheric Research (NCAR) Auto-Nowcast (ANC) system is one of the key forecast tools in the ATEC Four-Dimensional Weather System (4DWX) at WSMR, where its purpose is to aid WSMR meteorologists in their mission of very short term thunderstorm forecasting. Besides monitoring the weather activity throughout the region and warning personnel of potentially hazardous thunderstorms, forecasters play a key role in assisting with the day-to-day planning of test operations on the range by providing guidance with regard to weather conditions favorable to testing. Moreover, based on climatological information about the local weather conditions, forecasters advise their range customers about scheduling tests at WSMR months in advance. This paper reviews the NCAR ANC system, provides examples of the ANC system's use in thunderstorm forecasting, and describes climatological analyses of WSMR summertime thunderstorm activity relevant for long-range planning of tests. The climatological analysis illustrates that radar-detected convective cells with reflectivity of Ն35 dBZ at WSMR are 1) short lived, with 76% having lifetimes of less than 30 min; 2) small, with 67% occupying areas of less than 25 km 2 ; 3) slow moving, with 79% exhibiting speeds of less than 4 m s Ϫ1 ; 4) moderately intense, with 80% showing reflectivities in excess of 40 dBZ; and 5) deep, with 80% of the storms reaching far enough above the freezing level to be capable of generating lightning.

A modern renewable energy forecasting system blends physical models with artificial intelligence to aid in system operation and grid integration. This paper describes such a system being developed for the Shagaya Renewable Energy Park, which is being developed by the State of Kuwait. The park contains wind turbines, photovoltaic panels, and concentrated solar renewable energy technologies with storage capabilities. The fully operational Kuwait Renewable Energy Prediction System (KREPS) employs artificial intelligence (AI) in multiple portions of the forecasting structure and processes, both for short-range forecasting (i.e., the next six hours) as well as for forecasts several days out. These AI methods work synergistically with the dynamical/physical models employed. This paper briefly describes the methodology used for each of the AI methods, how they are blended, and provides a preliminary assessment of their relative value to the prediction system. Each operational AI component adds value to the system. KREPS is an example of a fully integrated state-of-the-science forecasting system for renewable energy.

This work compares the solar power forecasting performance of tree-based methods that include implicit regime-based models to explicit regime separation methods that utilize both unsupervised and supervised machine learning techniques. Previous studies have shown an improvement utilizing a regime-based machine learning approach in a climate with diverse cloud conditions. This study compares the machine learning approaches for solar power prediction at the Shagaya Renewable Energy Park in Kuwait, which is in an arid desert climate characterized by abundant sunshine. The regime-dependent artificial neural network models undergo a comprehensive parameter and hyperparameter tuning analysis to minimize the prediction errors on a test dataset. The final results that compare the different methods are computed on an independent validation dataset. The results show that the tree-based methods, the regression model tree approach, performs better than the explicit regime-dependent approach. These results appear to be a function of the predominantly sunny conditions that limit the ability of an unsupervised technique to separate regimes for which the relationship between the predictors and the predictand would differ for the supervised learning technique.