Proceedings of the 3rd International Conference on Public Management and Big Data Analysis, PMBDA 2023, December 15–17, 2023, Nanjing, China

Research Article

Intelligent Analysis of Enterprise Power Trading Based on Stochastic Stability Particle Swarm Optimization

Download18 downloads
  • @INPROCEEDINGS{10.4108/eai.15-12-2023.2345326,
        author={Xinhui  Wang},
        title={Intelligent Analysis of Enterprise Power Trading Based on Stochastic Stability Particle Swarm Optimization},
        proceedings={Proceedings of the 3rd International Conference on Public Management and Big Data Analysis, PMBDA 2023, December 15--17, 2023, Nanjing, China},
        publisher={EAI},
        proceedings_a={PMBDA},
        year={2024},
        month={5},
        keywords={particle swarm; electricity trading; intelligent analysis},
        doi={10.4108/eai.15-12-2023.2345326}
    }
    
  • Xinhui Wang
    Year: 2024
    Intelligent Analysis of Enterprise Power Trading Based on Stochastic Stability Particle Swarm Optimization
    PMBDA
    EAI
    DOI: 10.4108/eai.15-12-2023.2345326
Xinhui Wang1,*
  • 1: Information Technology Department of Beijing Agricultural Vocational College
*Contact email: wangxh12252023@163.com

Abstract

In order to understand the application of particle swarm optimization algorithm in enterprise power trading, research on intelligent analysis of enterprise power trading based on stochastic stability particle swarm optimization has been proposed. The particle swarm algorithm involves three random variables, namely initial position, initial velocity, and learning constant correction coefficient. The distribution function in the algorithm is a uniform distribution. In this paper, the uniform distribution is generalized and transformed into a probability distribution function that meets the corresponding conditions. Based on the 2022 electricity market trading rules of a certain province, formulate a "thermal power conventional" variety purchase strategy in the annual trading. In the case study, uniform distribution and normal distribution were selected as the research objects, and the convergence of the two distributions in the annual strategy formulation of particle swarm optimization was compared. Table 3 shows that selecting a normal distribution (corrected) for only the initial position can significantly improve global convergence.