Skip to main content

Table 1 Neural networks in the literature predicting (sub-)hourly local loads. Description is provided in “Neural network applications” section

From: Adequacy of neural networks for wide-scale day-ahead load forecasts on buildings and distribution systems using smart meter data

Reference

Network

Load

Dataset

Horizon

Inputs

Setup

Hidden

Hidden

Training

   

(months)

   

layers

neurons

algorithm

Bagnasco et al. (2015)

MLP

hospital

12

DALF

load, calendar, weather

manual

1

20

BP

Marinescu et al. (2013)

MLP

aggregation

17

DALF

load, weather

manual

1

15

BP

Hernández et al. (2014)

MLP

microgrid

54

DALF

load, calendar, weather

auto

1

21

var

Amjady et al. (2010)

MLP

microgrid

12

DALF

load, calendar, weather

auto

1

n/s

LM

Llanos et al. (2012)

MLP

microgrid

5

48 h

load

manual

1

8

LM

Chitsaz et al. (2015b)

RNN

edu. building

12

DALF

load, calendar, weather

manual

1

n/s

LM

Mena et al. (2014)

RNN

lab. building

18

1 h

load, calendar, weather

manual

1

10

LM

Pîrjan et al. (2017)

RNN

hypermarket

12

1 h

load, calendar, weather

manual

1

var

var

Hayes and Prodanovic (2016)

RNN

46 MV/LV

24

24 h

load, calendar, weather

manual

1

10

LM

Mocanu et al. (2016)

RBM

home

48

DALF

load

manual

1

10

LM

Marino et al. (2016)

LSTM

home

48

60 h

load, calendar

manual

2

10

BP

Amarasinghe et al. (2017)

CNN

home

48

DALF

load, calendar

manual

2

20

BP

Shi et al. (2016)

ESN

office building

48

1 h

load

auto

1

50

n/s

Kong et al. (2017)

LSTM

69 homes

48

6 h

load, calendar

manual

2

20

BP

Ryu et al. (2016)

RBM

40 enterprises

36

DALF

load, calendar, weather

manual

4

150

LM