Forecasting Method of LV Distribution’s Load CurveBy Means of Machine Learning Utilizing Smart Meter Data

Thumbnail Image
Files
CIRED 2019 - 806.pdf (1.27 MB)
Paper number
806
Working Group Number
Conference name
CIRED 2019
Conference date
3-6 June 2019
Conference location
Madrid, Spain
Peer-reviewed
Yes
Short title
Convener
Authors
Kanazawa, Yuki, Chubu Electric Power Company, Japan
Ishikawa, Hiroyuki, Chubu Electric Power Company, Japan
Uenishi, Hirokazu, Chubu Electric Power Company, Japan
Ichinomiya, Hiroki, Mitsubishi Research Institute, Japan
Abstract
Maximum current is necessary to be estimated when designing a LV distribution grid, i.e. MV/LV pole-mounted transformers and LV feeders, which supplies customers of several in minimum and dozens in maximum. This is currently estimated from the total contract capacities of the customers supplied by the grid, multiplying diversity and demand factor corresponding to the number of customers. This method is rather efficient and simple because the calculation can be done simply by using the contract capacities and the numbers of contracts.However, there are some constraints in this method. Diversity and demand factors may not be correct as they were derived statistically from sampled current measurements of customers. Additionally, this research was carried out years ago and have never been updated as it takes vast time and effort to re-measure them. As a result, these factors may be out-of-date as recent changes in electricity usage are not taken into consideration, such as the introduction of highly-efficient appliances or the diversities in lifestyles of customers. Thus, there is a possibility that the current estimated from the conventional method which is used for designing the grid diverge much from the actual state. This leads to over investment.This paper describes our new methodology for estimating the load curves of newly-connected customers based on the off-line information, such as contract capacity, contract types, location of the customer, presence of PV, etc.
Table of content
Keywords
Publisher
AIM
Date
2019-06-03
Published in
CIRED 2019 Conference
Permanent link to this record
https://cired-repository.org/handle/20.500.12455/155
http://dx.doi.org/10.34890/310
ISSN
2032-9644
ISBN
978-2-9602415-0-1