Casual Relationship Identification Between Power System Load and Meteorological Data Based on Improved FCI Algorithm

ZHANG Guanglun; ZHONG Haiwang

doi:10.13334/j.0258-8013.pcsee.240407

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Casual Relationship Identification Between Power System Load and Meteorological Data Based on Improved FCI Algorithm

更新时间：2026-07-14

- Casual Relationship Identification Between Power System Load and Meteorological Data Based on Improved FCI Algorithm
- Vol. 45, Issue 17, Pages: 6612-6624(2025)
- 作者机构：
  
  清华大学电机工程与应用电子技术系, 北京市海淀区,100084
- 作者简介：
- 基金信息：
- DOI：10.13334/j.0258-8013.pcsee.240407
  CLC：
- Published：2025
- 稿件说明：
移动端阅览
ZHANG Guanglun, ZHONG Haiwang. Casual Relationship Identification Between Power System Load and Meteorological Data Based on Improved FCI Algorithm[J]. 2025, 45(17): 6612-6624.
DOI：

ZHANG Guanglun, ZHONG Haiwang. Casual Relationship Identification Between Power System Load and Meteorological Data Based on Improved FCI Algorithm[J]. 2025, 45(17): 6612-6624. DOI： 10.13334/j.0258-8013.pcsee.240407.

摘要

电力系统中存在着大量不同时间尺度的时间序列变量，如运行场景下的功率负荷曲线、暂态场景下的电压电流波形等。其中，负荷与其他时序变量之间的关联关系在负荷预测、调度运行等场景下均被广泛地研究。作为关联关系的一种表述形式，因果性具有明确的物理意义，所描述的关联性信息冗余度小，能够在一定程度上反映电力系统的物理运行规律，在电力大数据的背景下具有广泛的应用前景。该文提出一种基于多种因果辨识原理的改进快速因果推断(fast casual inference，FCI)算法，并将其用于电力负荷-气象数据因果关系辨识中，建立负荷及多种气象时序数据间的因果关系图。基于某地2019—2022年的实际数据开展案例分析，并与基于相关性分析建立关联关系图的方法进行对比，验证所提方法的有效性。

Abstract

In power systems

many variables appear as time series under different scales

such as voltage/current waveforms in transient analysis or power/load data. Among these variables

the relationship between load and other time series has been widely investigated in applications such as load forecasting or system dispatch. Analysis of relationship between time series can not only provide reference for research on unknown variables

but also can reveal the underlying physical laws of power system operation to some extent. However

in most existing research

the concept of relationship stays as correlation

rather than causality

which leads to the weakness of the physical meaning of relationship

limited information provided by some related variables

and the high complexity of the model. Current research concerning causality in power systems focus mainly on causality between two variables

which is hardly applicable for multiple variables. To this end

an improved fast casual inference algorithm based on multiple causality inference method is proposed in this paper and applied to infer the causality between load and multiple meteorological data series. Case studies based on real data from 2019 to 2022 are conducted to compare with a method based on correlation analysis

which validates the effectiveness of the proposed algorithm.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Short-circuit Ratio Indicator for Grid-forming DC-supported Renewable Energy Radiative Integration System

Design and Optimization of Integrated Energy System for Novel Solar Driven Power Generation and Hydrogen Production

Microstructure Evolution and Mechanical Performance During Sodiation of Hard Carbon Anode in Sodium-ion Battery by ReaxFF-MD Method

Construction and Thermodynamic Analysis of a Coal-fired Peak Shaving System Integrated With Compressed Air Energy Storage and Gas Turbine

Series-capacitor-transformer Hybrid High Step-down Ratio Buck Converter

Related Author

No data

Related Institution

No data

AI问答

⁰