Journal of Modern Power Systems and Clean Energy

ISSN 2196-5625 CN 32-1884/TK

Recurrent Neural Network for Nonconvex Economic Emission Dispatch
Author:
Affiliation:

1.Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing, College of Electronic and Information Engineering, Southwest University, Chongqing 400715, China;2.Key Laboratory of Machine Perception and Children’s Intelligence Development, Chongqing University of Education, Chongqing 400067, China;3.School of Electrical Engineering and Computer Science, University of Newcastle, Newcastle, NSW 2308, Australia

Fund Project:

This work was supported by the Fundamental Research Funds for the Central Universities (No. XDJK2019B010), the Natural Science Foundation of China (No. 61773320), the Natural Science Foundation Project of Chongqing Science and Technology Commission (CSTC) (No. cstc2018jcyjAX0583, No. cstc2018 jcyjAX0810), the Research Foundation of Key Laboratory of Machine Perception and Children’s Intelligence Development funded by Chongqing University of Education (CQUE) (No. 16xjpt07), and the Foundation of Chongqing University of Education (No. KY201702A ).

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
    Abstract:

    In this paper, an economic emission dispatch (EED) model is developed to reduce fuel cost and environmental pollution emissions. Considering the development of new energy sources in recent years, the EED problem involves thermal units with the valve point effect and WTs. Meanwhile, it complies with demand constraint and generator capacity constraints. A recurrent neural network (RNN) is proposed to search for local optimal solution of the introduced nonconvex EED problem. The optimality and convergence of the proposed dynamic model are given. The RNN algorithm is verified on a power generation system for the optimization of scheduling and minimization of total cost. Moreover, a particle swarm optimization (PSO) algorithm is compared with RNN under the same problematic frame. Numerical simulation results demonstrate that the optimal scheduling given by RNN is more precise and has lower total cost than PSO. In addition, the dynamic variation of power load demand is considered and the power distribution of eight generators during 12 time periods is depicted.

    表 1 Table 1
    表 2 Table 2
    表 5 Table 5
    表 3 Table 3
    图1 Architecture of proposed EED model.Fig.1
    图2 Effect caused by VPE.Fig.2
    图3 Network architecture of proposed RNN.Fig.3
    图7 Load variations in 12 time periods.Fig.7
    图8 Power distribution of four generators in 12 time periods.Fig.8
    图9 Power distribution of four generators in 12 time periods.Fig.9
    图10 Optimal total cost by PSO.Fig.10
    表 4 Table 4
    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:December 14,2018
  • Online: January 22,2021