Table 8 Parameters used in neural network-based models.

From: Learning model combined with data clustering and dimensionality reduction for short-term electricity load forecasting

Modes

Types

Parameters

ANN

Dense Layer

In = Input Size, Out = 256

Dense Layer

In = 256, Out = 24

Dense Layer

In = 4608, Out = 256

Dense Layer

In = 256, Out = 24

CNN

Conv 1D Layer

In = Input Size, Out = 512, Kernel Size = 3, Padding = 1

MaxPool 1D Layer

Kernel Size = 2

Conv 1D Layer

In=512, Out = 256, Kernel Size = 3, Padding = 1

MaxPool 1D Layer

Kernel Size = 2

Conv 1D Layer

In = 256, Out = 128, Kernel Size = 3, Padding = 1

MaxPool 1D Layer

Kernel Size = 2

Dense Layer

In=3,072, Out = 256

Dense Layer

In=256, Out = 24

LSTM

Dense Layer

In = Input Size, Out = 288

Bidirectional LSTM Layer

In = 288, Hidden = 64, Num of Layers = 5

Dense Layer

In = 3072, Out = 256

Dense Layer

In = 256, Out = 24

Back to article page