Abstract

Different energy systems are closely connected with each other in industrial-park integrated energy system (IES). The energy demand forecasting has important impact on IES dispatching and planning. This paper proposes an approach of short-term energy forecasting for electricity, heat, and gas by employing deep multitask learning whose structure is constructed by deep belief network (DBN) and multitask regression layer. The DBN can extract abstract and effective characteristics in an unsupervised fashion, and the multitask regression layer above the DBN is used for supervised prediction. Then, subject to condition of practical demand and model integrity, the whole energy forecasting model is introduced, including preprocessing, normalization, input properties, training stage, and evaluating indicator. Finally, the validity of the algorithm and the accuracy of the energy forecasts for an industrial-park IES system are verified through the simulations using actual operating data from load system. The positive results turn out that the deep multitask learning has great prospects for load forecast.

Highlights

  • To date, massive use of fossil energy has caused serious harm for society, environment, and economy.At the same time, renewable energy has received widespread attention and become an important part of modern energy due to its good cleanliness and low carbon property

  • Since a deep learning algorithm requires considerable calculations and storage resources because of its special structures, which contain enormous numbers of nodes and layers compared with a conventional shallow algorithm, services related to this portion of the framework could be provided by industrial-park integrated energy system (IES) service providers or the operators of the energy distribution timeliness, each training dataset used in off-line training needs to span the same appropriate fixed time range

  • Experiments and Results of Goldwind Technology Co., Ltd., in Daxing District, Beijing, China, consisting by electricity, The experimental dataset is collected from the industrial-park IES demonstration project of natural gas, and heat

Read more

Summary

Introduction

Massive use of fossil energy has caused serious harm for society, environment, and economy. Traditional research mainly focuses on the planning, design, and operation of individual energy systems, while ignoring the coupling between different types of energy source, resulting in greatly reduced flexibility in system scheduling. On the basis, integrated energy system (IES) is proposed, which involved electricity, natural gas, and heat. Researches and works on IES can be divided into two characteristics based on application scenarios as [2,3] wide-area IES and industrial-park IES. Wide-area IES mainly includes the electricity transmission network and gas transmission network. Industrial-park IES involve the coordination of multiple regional energy systems on distribution network and can be seen as upgraded microgrids.

Wide-area
Comparison
Deep Belief Network
Multitask Learning
Preprocessing for loads Data and Normalization
Input Properties Setting
Evaluating Indicator
Experiments and Results
The interactionsamong amongvarious various energy industrial-park
Parameters
Various Types of Load Forecasts Results
Comparison of Multiple Forecasting Methods
Comparison of Different Training Models
12. Comparison
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call