Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Real-world time series with distinct periodic pattern.

More »

Fig 1 Expand

Fig 2.

Overall structure of woa-wtconv-kanformer.

More »

Fig 2 Expand

Fig 3.

Flowchart of whale optimization algorithm.

More »

Fig 3 Expand

Table 1.

The Algorithm of WOA.

More »

Table 1 Expand

Fig 4.

Diagram illustrating the grouped convolution within the wtconv module.

More »

Fig 4 Expand

Fig 5.

Illustration of wavelet deconvolution within the wtconv module.

More »

Fig 5 Expand

Table 2.

The Algorithm of WTConv model.

More »

Table 2 Expand

Fig 6.

Linear layer of MLP network.

More »

Fig 6 Expand

Fig 7.

Linear layers integrated with wav-kan.

More »

Fig 7 Expand

Table 3.

Multivariate prediction results, model prediction length is S ∈ {96,192,336,720}, and fixed lookback length T = 96.

More »

Table 3 Expand

Fig 8.

Time series prediction plot with a forecast horizon of 96 time steps.

More »

Fig 8 Expand

Fig 9.

Time series prediction plot with a forecast horizon of 336 time steps.

More »

Fig 9 Expand

Table 4.

Comparison of model efficiency.

More »

Table 4 Expand

Table 5.

Performance comparison of each module.

More »

Table 5 Expand

Table 6.

Algorithm analysis of each module.

More »

Table 6 Expand

Table 7.

Multivariate prediction results, model prediction length is S ∈ {96,336}, and fixed lookback length T = 96.

More »

Table 7 Expand

Fig 10.

Time series prediction plot with a forecast horizon of 96 time steps.

More »

Fig 10 Expand

Table 8.

Performance comparison of each model.

More »

Table 8 Expand

Table 9.

Multivariate prediction results, model prediction length is S ∈ {96,336}, and fixed lookback length T = 96.

More »

Table 9 Expand

Fig 11.

Time series prediction plot with a forecast horizon of 96 time steps.

More »

Fig 11 Expand

Fig 12.

Time series prediction plot with a forecast horizon of 336 time steps.

More »

Fig 12 Expand

Table 10.

WOA parameter optimization loss function value.

More »

Table 10 Expand

Table 11.

WOA parameter optimization time consumption.

More »

Table 11 Expand

Table 12.

The list of parameters with their respective definitions.

More »

Table 12 Expand

Table 13.

Parameter tuning iterative process.

More »

Table 13 Expand

Fig 13.

Iterative comparison of model loss function (mse) before and after woa optimization.

More »

Fig 13 Expand

Fig 14.

Prior and following the utilization of the WOA module.

More »

Fig 14 Expand

Table 14.

In a noisy environment multivariate prediction results, model prediction length is S ∈ {96,336}, and fixed lookback length T = 96.

More »

Table 14 Expand

Fig 15.

In a noisy environment multivariate prediction results, model prediction length is 96, and fixed lookback length T = 96.

More »

Fig 15 Expand

Fig 16.

In a noisy environment multivariate prediction results,The global prediction graph with a prediction step size of 96.

More »

Fig 16 Expand