An intelligent demand response (DR) program is developed for multi-energy industrial micro-grid consisting of manufacturing facilities, photovoltaic (PV) panels, and battery energy storage system (BESS). The proposed DR program tackles the practical challenges of components in the micro-grid including industrial process represented by a discrete manufacturing production model, uncertainty of PV generation, and operational cost of the BESS. The proposed DR program optimizes day-ahead production scheduling for manufacturing facilities and operation regime for the BESS in response to time of use electricity price and probabilistic forecasting of PV power. To capture the uncertainty of PV power, a data-driven PV power probabilistic forecasting model is developed and a copula-based approach is deployed for the sampling of temporally correlated scenarios of PV power over the scheduling horizon from probabilistic forecasts. The multi-energy management optimization problem is formulated as a scenario-based stochastic nonconvex mixed integer nonlinear programming (MINLP). A hybrid optimization method integrating the evolutionary algorithm and the branch-and-bound algorithm for mixed integer liner programming is proposed to solve the nonconvex MINLP. Simulation studies illustrate that the proposed DR program efficiently reduces the operational cost for manufacturing production and releases the stress of the main grid by making full use of flexibility of all the components in the micro-grid.

Short-term forecasting of solar irradiance at a targeted site with consideration of its time-series and measurements at neighboring sites is studied in this paper. A data-driven framework for forecasting solar irradiance based on fusing spatial and temporal information is proposed. In the framework, data-driven approaches including boosted regression trees (BRT), artificial neural network (ANN), support vector machine (SVM), and least absolute shrinkage and selection operator (LASSO) are applied to model spatial dependence among solar irradiance time-series thereby to generate forecasts of solar irradiance at the targeted site. A comprehensive comparison among data-driven forecasting models is performed using 30-min averaged data of recent two years. Moreover, benchmarking models including scaled persistence (S-PER) model, autoregressive (AR) model, and autoregressive exogenous (ARX) model are employed to further validate the effectiveness of data-driven forecasting models. Computational results of multiple-steps ahead forecasting demonstrate that the BRT model gives the best performance with the lowest normalized root mean squared error (nRMSE) of 18.4%, 24.3%, 27.9%, and 30.6% for forecasting horizons of 30-min, 60-min, 90-min, and 120-min, respectively.

To address foreseeable challenges during the penetration of wind energy into the power grid, including accurate wind power forecasting and smart power generation scheduling, this study proposes a novel short-term wind speed forecasting model, named EMD-KM-SXL, which is based on empirical mode decomposition (EMD), K-means clustering and machine learning, and a new two-stage short-term wind power forecasting model based on wind speed forecasting and wind power curve modeling. The former wind speed forecasting model regards historical wind speed observations as model inputs, and the latter power forecasting model utilizes knowledge augmentation, introducing wind power conversion relationship, environmental factors as well as wind power system status parameters. In the proposed wind speed forecasting model, three machine learning models, including support vector regressor, XGBoost regressor, and Lasso regressor, are employed to forecast three types of frequency components that are generated via EMD and K-means clustering. Then, the wind power curve model is utilized to compute potential outpower based on the predicted wind speed, which is regarded as the first stage of the proposed wind power forecasting model. In the second stage, environmental factors and wind power system status parameters are introduced and an artificial neural network model, considering preliminary predicted power, environmental factors, and wind power system status parameters as model inputs, is built to make final power prediction. Computational results show that proposed models achieve the best performance in terms of wind speed and power forecasting over different forecasting horizons ranging from 10 to 40 minutes, compared with benchmarking methods.