PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid

Huaiyuan Liu; Hongzhi Wang; Hekai Huang; Donghua Yang; Yong Tang; Yanhao Huang

doi:10.1007/978-3-031-30678-5_50

PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid

Huaiyuan Liu
, Hongzhi Wang^*
, Hekai Huang
, Donghua Yang
, Yong Tang
, Yanhao Huang

^*Corresponding author for this work

Faculty of Computing

Harbin Institute of Technology
ZTE Corporation
State Grid Corporation of China

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Various stability analyses of the power system are based on the results of power flow calculation, which is not always convergent. In practice, large manual efforts are required to be repeated many times by electrical engineers to ensure the convergence of power flow calculation, which consumes much manpower and time cost. Motivated by this, we develop a novel knowledge-driven system that can automatically improve power flow convergence based on knowledge from experiences, called PFKMaster. Based on the features of power flow calculation, a SAS triplet mechanism is designed to better represent the experience knowledge of humans. In our system, a knowledge model is proposed to guide the power flow calculation automatically and universally. To achieve the goal, we build a specific knowledge base. In this system, new knowledge is discovered based on the method of representation learning and affair logic. To ensure knowledge quality, we also propose knowledge cleaning techniques in our system. We test PFKMaster on a real large-scale power grid, and the experimental results demonstrate that 94.8% of the non-convergent samples can be improved by 1.92 s per sample.

Original language	English
Title of host publication	Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings
Editors	Xin Wang, Maria Luisa Sapino, Wook-Shin Han, Amr El Abbadi, Gill Dobbie, Zhiyong Feng, Yingxiao Shao, Hongzhi Yin
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	650-657
Number of pages	8
ISBN (Print)	9783031306778
DOIs	https://doi.org/10.1007/978-3-031-30678-5_50
State	Published - 2023
Event	28th International Conference on Database Systems for Advanced Applications, DASFAA 2023 - Tianjin, China Duration: 17 Apr 2023 → 20 Apr 2023

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13946 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	28th International Conference on Database Systems for Advanced Applications, DASFAA 2023
Country/Territory	China
City	Tianjin
Period	17/04/23 → 20/04/23

Keywords

Knowledge graph
Knowledge-driven
Power flow calculation

Access to Document

10.1007/978-3-031-30678-5_50

Cite this

Liu, H., Wang, H., Huang, H., Yang, D., Tang, Y., & Huang, Y. (2023). PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid. In X. Wang, M. L. Sapino, W.-S. Han, A. El Abbadi, G. Dobbie, Z. Feng, Y. Shao, & H. Yin (Eds.), Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings (pp. 650-657). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13946 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-30678-5_50

Liu, Huaiyuan ; Wang, Hongzhi ; Huang, Hekai et al. / PFKMaster : A Knowledge-Driven Flow Control System for Large-Scale Power Grid. Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings. editor / Xin Wang ; Maria Luisa Sapino ; Wook-Shin Han ; Amr El Abbadi ; Gill Dobbie ; Zhiyong Feng ; Yingxiao Shao ; Hongzhi Yin. Springer Science and Business Media Deutschland GmbH, 2023. pp. 650-657 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{1e814d195b9d46f1ad309bbdb188ded5,

title = "PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid",

abstract = "Various stability analyses of the power system are based on the results of power flow calculation, which is not always convergent. In practice, large manual efforts are required to be repeated many times by electrical engineers to ensure the convergence of power flow calculation, which consumes much manpower and time cost. Motivated by this, we develop a novel knowledge-driven system that can automatically improve power flow convergence based on knowledge from experiences, called PFKMaster. Based on the features of power flow calculation, a SAS triplet mechanism is designed to better represent the experience knowledge of humans. In our system, a knowledge model is proposed to guide the power flow calculation automatically and universally. To achieve the goal, we build a specific knowledge base. In this system, new knowledge is discovered based on the method of representation learning and affair logic. To ensure knowledge quality, we also propose knowledge cleaning techniques in our system. We test PFKMaster on a real large-scale power grid, and the experimental results demonstrate that 94.8\% of the non-convergent samples can be improved by 1.92 s per sample.",

keywords = "Knowledge graph, Knowledge-driven, Power flow calculation",

author = "Huaiyuan Liu and Hongzhi Wang and Hekai Huang and Donghua Yang and Yong Tang and Yanhao Huang",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 28th International Conference on Database Systems for Advanced Applications, DASFAA 2023 ; Conference date: 17-04-2023 Through 20-04-2023",

year = "2023",

doi = "10.1007/978-3-031-30678-5\_50",

language = "英语",

isbn = "9783031306778",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "650--657",

editor = "Xin Wang and Sapino, \{Maria Luisa\} and Wook-Shin Han and \{El Abbadi\}, Amr and Gill Dobbie and Zhiyong Feng and Yingxiao Shao and Hongzhi Yin",

booktitle = "Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings",

address = "德国",

}

Liu, H, Wang, H, Huang, H, Yang, D, Tang, Y & Huang, Y 2023, PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid. in X Wang, ML Sapino, W-S Han, A El Abbadi, G Dobbie, Z Feng, Y Shao & H Yin (eds), Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13946 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 650-657, 28th International Conference on Database Systems for Advanced Applications, DASFAA 2023, Tianjin, China, 17/04/23. https://doi.org/10.1007/978-3-031-30678-5_50

PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid. / Liu, Huaiyuan; Wang, Hongzhi; Huang, Hekai et al.
Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings. ed. / Xin Wang; Maria Luisa Sapino; Wook-Shin Han; Amr El Abbadi; Gill Dobbie; Zhiyong Feng; Yingxiao Shao; Hongzhi Yin. Springer Science and Business Media Deutschland GmbH, 2023. p. 650-657 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13946 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - PFKMaster

T2 - 28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

AU - Liu, Huaiyuan

AU - Wang, Hongzhi

AU - Huang, Hekai

AU - Yang, Donghua

AU - Tang, Yong

AU - Huang, Yanhao

PY - 2023

Y1 - 2023

N2 - Various stability analyses of the power system are based on the results of power flow calculation, which is not always convergent. In practice, large manual efforts are required to be repeated many times by electrical engineers to ensure the convergence of power flow calculation, which consumes much manpower and time cost. Motivated by this, we develop a novel knowledge-driven system that can automatically improve power flow convergence based on knowledge from experiences, called PFKMaster. Based on the features of power flow calculation, a SAS triplet mechanism is designed to better represent the experience knowledge of humans. In our system, a knowledge model is proposed to guide the power flow calculation automatically and universally. To achieve the goal, we build a specific knowledge base. In this system, new knowledge is discovered based on the method of representation learning and affair logic. To ensure knowledge quality, we also propose knowledge cleaning techniques in our system. We test PFKMaster on a real large-scale power grid, and the experimental results demonstrate that 94.8% of the non-convergent samples can be improved by 1.92 s per sample.

AB - Various stability analyses of the power system are based on the results of power flow calculation, which is not always convergent. In practice, large manual efforts are required to be repeated many times by electrical engineers to ensure the convergence of power flow calculation, which consumes much manpower and time cost. Motivated by this, we develop a novel knowledge-driven system that can automatically improve power flow convergence based on knowledge from experiences, called PFKMaster. Based on the features of power flow calculation, a SAS triplet mechanism is designed to better represent the experience knowledge of humans. In our system, a knowledge model is proposed to guide the power flow calculation automatically and universally. To achieve the goal, we build a specific knowledge base. In this system, new knowledge is discovered based on the method of representation learning and affair logic. To ensure knowledge quality, we also propose knowledge cleaning techniques in our system. We test PFKMaster on a real large-scale power grid, and the experimental results demonstrate that 94.8% of the non-convergent samples can be improved by 1.92 s per sample.

KW - Knowledge graph

KW - Knowledge-driven

KW - Power flow calculation

UR - https://www.scopus.com/pages/publications/85161677231

U2 - 10.1007/978-3-031-30678-5_50

DO - 10.1007/978-3-031-30678-5_50

M3 - 会议稿件

AN - SCOPUS:85161677231

SN - 9783031306778

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 650

EP - 657

BT - Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings

A2 - Wang, Xin

A2 - Sapino, Maria Luisa

A2 - Han, Wook-Shin

A2 - El Abbadi, Amr

A2 - Dobbie, Gill

A2 - Feng, Zhiyong

A2 - Shao, Yingxiao

A2 - Yin, Hongzhi

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 17 April 2023 through 20 April 2023

ER -

Liu H, Wang H, Huang H, Yang D, Tang Y, Huang Y. PFKMaster: A Knowledge-Driven Flow Control System for Large-Scale Power Grid. In Wang X, Sapino ML, Han WS, El Abbadi A, Dobbie G, Feng Z, Shao Y, Yin H, editors, Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 650-657. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-30678-5_50