Search

Demand-side management (DSM) increases the complexity of forecasting environment, which makes traditional forecasting methods difficult to meet the firm's need for predictive accuracy. Since deep learning can comprehensively consider various factors to improve prediction results, this paper improves the deep belief network from three aspects of input data, model and performance, and uses it to solve the short-term load forecasting problem in DSM. In the data optimization stage, the Hankel matrix is constructed to increase the input weight of DSM data, and the gray relational analysis is used to select strongly correlated data from the data set. In the model optimization stage, the Gauss-Bernoulli restricted Boltzmann machine is used as the first restricted Boltzmann machine of the deep network to convert the continuity feature of input data into binomial distribution feature. In the performance optimization stage, a pre-training method combining error constraint and unsupervised learning is proposed to provide good initial parameters, and the global fine-tuning of network parameters is realized based on the genetic algorithm. Based on the actual data of Tianjin Power Grid in China, the experimental results show that the proposed method is superior to other methods. [ABSTRACT FROM AUTHOR]

With the largest installed capacity in the world, wind power in China is experiencing a ∼20% curtailment. The inflexible combined heat and power (CHP) has been recognized as the major barrier for integrating the wind source. The approach to reconcile the conflict between inflexible CHP units and variable wind power in Chinese energy system is yet unclear. This paper explores the technical and economic feasibility of deploying the heat storage tanks and electric boilers under typical power grids and practical operational regulations. A mixed integer linear optimization model is proposed to simulate an integrated power and heating energy systems, including a CHP model capable of accounting for the commitment decisions and nonconvex energy generation constraints. The model is applied to simulate a regional energy system (Jing–Jin–Tang) covering 100-million population, with hourly resolution over a year, incorporating actual data, and operational regulations. The results project an accelerating increase in wind curtailment rate at elevated wind penetration. Investment for wind breaks even at 14% wind penetration. At such penetration, the electric boiler (with heat storage) is effective in reducing wind curtailment. The investment in electric boilers is justified on a social economic basis, but the revenues for different stakeholders are not distributed evenly. [ABSTRACT FROM AUTHOR]