Method for electricity-related security awareness of distributed power supply systems considering spatio-temporal distribution of rainstorms

Patent metadata

Field	Value
Publication number	US-11821961-B2
Application number	US-202318185429-A
Country	US
Kind code	B2
Filing date	Mar 17, 2023
Priority date	May 5, 2022
Publication date	Nov 21, 2023
Grant date	Nov 21, 2023

Abstract

A method for electricity-related security awareness of distributed power supply systems considering spatio-temporal distribution of rainstorms, including: establishing a multi-dimensional parallel parasitic capacitance calculation model of the distributed photovoltaic-energy storage power supply system considering accumulated water depth and micro-terrain environment; performing multi-source spatio-temporal hierarchical correlation analysis between rainstorm spatio-temporal distribution characteristics (including rainfall peak position, cloud movement, rainfall intensity and rainfall duration) and an operating state of the distributed power supply system; constructing a leakage current probability prediction model considering unevenness and randomness of the rainstorm spatio-temporal distribution; and establishing an electricity-related security awareness model based on deep meta-learning.

First claim

What is claimed is:

1. A method for electricity-related security awareness of distributed power supply systems considering spatio-temporal distribution of rainstorms, comprising:
(a) deriving a multi-dimensional parallel parasitic capacitance analysis model of a distributed photovoltaic-energy storage power supply system (DPSPSS) considering accumulated water depth and micro-terrain environment, to establish a leakage current calculation model of the DPSPSS under rainstorm conditions;
(b) collecting and preprocessing a dataset; dividing the dataset into a support set, a query set, a training set and a test set; wherein the dataset includes an input parameter and an output parameter; the input parameter comprises a characteristic parameter of spatio-temporal distribution of rainstorm and a micro-terrain characteristic parameter of a location of the photovoltaic-energy storage power supply system; the output parameter comprises model fitting parameters and micro-terrain fitting parameters; the characteristic parameter of spatio-temporal distribution of rainstorm comprises rainfall peak position, cloud movement, rainfall intensity, and rainfall duration; and the micro-terrain characteristic parameter includes a roof length, a roof width, a roof inclination angle, a roof drainage rate, photovoltaic installation per unit area, a photovoltaic installation inclination angle, and energy storage battery installation per unit area;
(c) performing spatio-temporal correlation analysis between a characteristic parameter of spatio-temporal distribution of rainstorm of a to-be-tested area and characteristic parameters of spatio-temporal distribution of rainstorms of surrounding areas; and selecting a characteristic parameter of spatio-temporal distribution of rainstorm of a surrounding area with high correlation followed by adding to a sample of the to-be-tested area to train the leakage current calculation model; and
(d) establishing a leakage current probability prediction model of the DPSPSS considering nonuniformity and randomness of spatio-temporal distribution of rainstorm; and performing leakage current risk perception in the to-be-tested area through the leakage current calculation model and the leakage current probability prediction model;
wherein the step (d) comprises:
establishing an electricity-related security awareness model of the DPSPSS based on deep meta-learning;
with the dataset as a training sample, dividing the training sample into different subtask samples according to spatio-temporal distribution; and pre-training the electricity-related security awareness model by using the subtask samples;
updating weight parameters and bias parameters of a training model according to spatio-temporal correlation of weight parameters and bias parameters obtained by pre-training;
training the training model based on the weight parameters and bias parameters obtained by pre-training to obtain the leakage current probability prediction model; and
according to the leakage current calculation model, obtaining a critical water depth corresponding to individual leakage current levels of the DPSPSS in the to-be-tested area; and inputting the critical water depth into the leakage current probability prediction model to identify a spatio-temporal distribution of electricity-related security risks of the DPSPSS in the to-be-tested area;
the step (a) comprises:
(a1) deriving the multi-dimensional parallel parasitic capacitance analysis model; and
(a2) based on the multi-dimensional parallel parasitic capacitance analysis model, establishing the leakage current calculation model;
wherein the step (a1) comprises:
(a11) obtaining a first parasitic capacitance model considering a depth of accumulated water on a surface of a photovoltaic cell and a depth of accumulated water between the photovoltaic cell and a first frame, wherein calculation formulas of the first parasitic capacitance model are listed as follows:

C
f
r

=

(

2
⁢

L
P

+

2
⁢

W
P

-

4
⁢

L
e

)

·

C
f

·

S
P

-

S
w

S
p

+

C
b
′

·

C
E
′

C
b
′

+

C
E
′

;

C
f

=

C
1

+

C
2

+

C
3

+

C
4

+

C
5

;

C
1

=

Assignees

• Univ Hunan

Inventors

• Zhou Bin
• Da Ziqi
• Shuai Zhikang
• Zhu Lipeng
• Zhang Cong
• Li Jiayong
• Jin Yifang

Classifications

• H02J2103/30

  Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks · CPC title

• H02J2101/24

  Photovoltaics · CPC title

• G01R31/52 Primary

  Testing for short-circuits, leakage current or ground faults · CPC title

• G01R27/2605

  Measuring capacitance (capacitive sensors G01D5/24) · CPC title

• G06Q10/04

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title