Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Edwin Campos Applicant's institution: ANL Applicant's division: Environmental Sciences Project Name: Solar_Forecasting Project title: Multi-Model Machine-Learning Solar Forecasting Technology Associated funding: DOE EERE SunShot Other Systems: None. Science: The objectives of this 3-year project (starting Aug. 2013) are as follows: -Step 1: Perform seminal research to improve modeling and forecasting of solar irradiance. -Step 2: Apply research results obtained from Step 1 (previous) to develop solar power forecasts. -Step 3: Validate developed solar forecasts to meet target values of pre-existing standard metrics. Project description: Current weather forecast inaccuracies often result in substantial economic losses, as the electric grid operators must continuously balance supply and demand to maintain the reliability of the power grid. Failures to predict solar energy generation accurately will undoubtedly constrain the national expansion of renewable energy. The relevant weather variable to forecast here is solar-irradiance, and its rapid variations are primarily due to cloud movement, cloud formation and dissipation. There are several technology barriers – such as forecast inputs integration, forecasts output integration, scalability of forecasting tools, and numerical representation of relevant cloud processes – that must be overcome for improving solar forecast accuracies to a useful level, if solar energy were to be become a dependable source of energy This project proposes to overcome these barriers by using novel machine-learning algorithms and hi-performance computers. The project will demonstrate that transformational improvements are possible on solar forecasting – for periods between 0 to 6 hours (including power ramps) and the day ahead – by determining optimal combinations for observations of thermodynamic profiles measured over a solar power plant, measured cloud motions in the area (e.g., from remote sensing observations at high resolution) and routine forecast outputs from numerical weather models. We are a team of 6 Argonne researchers, working in coordination with a new Research Consortium lead by IBM. We will develop advanced cloud and radiative transfer computations, as well as machine learning algorithms based on the self-adjusting voting algorithms of the Watson computer (http://en.wikipedia.org/wiki/Watson_%28computer%29). Project URL: http://solarhighpen.energy.gov/project/ibm_thomas_j_watson_research_center_w... Requested allocation: 200000 Q1: 50000 Q2: 50000 Q3: 50000 Q4: 50000 Justification: The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System