The irradiation forecasting technology is important for the effective utilization of solar power. Existing irradiation forecasting methods have achieved excellent performance with a massive amount of data in a centralized way. However, concerns about privacy protection and data security, which may arise in the process of data collection and transmission from distributed points to the centralized server, pose challenges to current forecasting methods. In this paper, a novel federated probabilistic forecasting scheme of solar irradiation is proposed based on deep learning, variational Bayesian inference, and federated learning. In this scheme, the training data are stored and computed in local IoT devices, only forecasting models are shared. Two real-world datasets from SolarGIS and NSRDB, and one benchmark dataset of Folsom are used to verify the feasibility and performance of the federated-based scheme. Comprehensive case studies are conducted to analyze the performance of the proposed scheme in multi-horizon. And the effects of its components are evaluated. Compared with other state-of-the-art probabilistic centralized models, when data can be shared, the proposed scheme achieves competitive forecasting performance on the basis of data privacy protection; when data-sharing is unavailable, due to the cooperative nature inherent (model-sharing) of federated learning, the performance advantage of the proposed scheme is more obvious.

Energy imbalance market (EIM) provides an opportunity that allows larger shares of variable renewable energy sources in the grid. Under highly volatile weather conditions, an accurate forecasting of photovoltaic (PV) power is necessary for grid stability and market operation. Most of existing forecasting methods strongly rely on the accuracy of measurements, and the adaptability of these methods to complex weather conditions is rarely discussed. In this paper, a weather classification multivariate adaptive regression spline (MARS) forecasting model is introduced for complex weather conditions in all seasons. It can be updated incrementally and its high computational efficiency satisfies EIM operations. A data set that consists of the historical power and meteorological parameters produced by a small-scale PV platform is classified and used to train MARS models with forecast horizons ranging from 15 min to 24 h in different seasons. The tests and analyses results indicate higher accuracy, adaptability, and efficiency of the novel model.

The unprecedented level of flexibility in energy management is required to ensure the balance of real-time energy production and consumption. Accurate short-term load forecasting (STLF) is vital for making the intelligent operation scheme. However, conventional forecasting techniques may not meet the increasingly demanding precision in load forecasting. This paper presents a novel energy load forecasting methodology based on Recurrent Neural Network (RNN), specifically Long Short-term Memory (LSTM) algorithms. The proposed LSTM-based model was trained and tested on a benchmark dataset which contained electricity consumption data for different kinds of buildings in America with onehour resolution. The comparative models including multi-layer perceptron neural network (MLP), random forest (RF), and kernelized support vector machine (SVM) was also tested on the same dataset. The week-ahead forecasting results have shown that the proposed LSTM-based model outdoes the three comparative models in nine of twelve months.

For distributed energy system, multi-energy sources and their comprehensive utilization are popularly concerned nowadays. In this paper, optimal operation of a multi-energy complementary distributed combined cooling, heating and power (CCHP) system is deeply studied. This system is applied on a commercial building in Beijing and it has complex configuration including natural gas energy source, renewable energy source and different energy conversion devices. Meanwhile, a separated cooling, heating and power (SCHP) supply system is also proposed. Their optimal operation strategies and performance are compared to show the advantages of multi-energy complementary system. Through simulation, the results show that the multi-energy complementary system operates with great performance and the optimal operation strategy is effective.