Time series prediction based on echo state network tuned by divided adaptive multi-objective differential evolution algorithm

Ren, Weijie; Wang, Yiwen; Han, Min

doi:10.1007/s00500-020-05457-8

Cited by 6 publications

(1 citation statement)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ESN is a recurrent neural network with three layers: the input layer, the reservoir layer, and the output layer. The reservoir layer includes hundreds or thousands of sparsely and recursively connected neurons [31].…”

Section: Echo State Networkmentioning

confidence: 99%

Echo state network-based metamodel for the problem of stagnant petroleum cooling in pipe-in-pipe flowlines

Ribeiro da Silva Graciano,

da Silva Fortuna,

Oinhos Pedruzzi

et al. 2023

Sci. Plena

View full text Add to dashboard Cite

The low temperatures found in subsea water combined with the transport of hydrocarbons in flowlines that are immersed in this environment brings the problem of solid deposition inside the flowlines, mainly of paraffin and hydrates, which consists of considerable problems to the petroleum industry. To overcome this issue, computational simulation might be an ally. However, such simulations sometimes can be computationally demanded, making their use unpractical. In these cases, metamodels, which are computationally cheaper models, can be an alternative. The present paper proposes using an Echo State Network as a surrogate model for the problem of stagnant production fluid cooling in a subsea pipe-in-pipe system with active electric heating. The implemented metamodel shows to be twenty-four times faster than the Finite Volume Method, although it showed consistent correlations for all lags when a residual autocorrelation analysis was performed.

show abstract

Section: Echo State Networkmentioning

confidence: 99%

Echo state network-based metamodel for the problem of stagnant petroleum cooling in pipe-in-pipe flowlines

Ribeiro da Silva Graciano,

da Silva Fortuna,

Oinhos Pedruzzi

et al. 2023

Sci. Plena

View full text Add to dashboard Cite

show abstract

A Hidden Markov Model-based fuzzy modeling of multivariate time series

Pedrycz

Wang

et al. 2022

Soft Comput

View full text Add to dashboard Cite

A conformal predictive system for distribution regression with random features

Zhang

Wang

2023

Soft Comput

View full text Add to dashboard Cite

Distribution regression is the regression case where the input objects are distributions. Many machine learning problems can be analysed in this framework, such as multi-instance learning and learning from noisy data. This paper attempts to build a conformal predictive system(CPS) for distribution regression, where the prediction of the system for a test input is a cumulative distribution function(CDF) of the corresponding test label. The CDF output by a CPS provides useful information about the test label, as it can estimate the probability of any event related to the label and be transformed to prediction interval and prediction point with the help of the corresponding quantiles. Furthermore, a CPS has the property of validity as the prediction CDFs and the prediction intervals are statistically compatible with the realizations. This property is desired for many risk-sensitive applications, such as weather forecast. To the best of our knowledge, this is the first work to extend the learning framework of CPS to distribution regression problems. We first embed the input distributions to a reproducing kernel Hilbert space using kernel mean embedding approximated by random Fourier features, and then build a fast CPS on the top of the embeddings. While inheriting the property of validity from the learning framework of CPS, our algorithm is simple, easy to implement and fast. The proposed approach is tested on synthetic data sets and can be used to tackle the problem of statistical postprocessing of ensemble forecasts, which demonstrates the effectiveness of our algorithm for distribution regression problems.

show abstract

Time series prediction based on echo state network tuned by divided adaptive multi-objective differential evolution algorithm

Cited by 6 publications

References 36 publications

Echo state network-based metamodel for the problem of stagnant petroleum cooling in pipe-in-pipe flowlines

Echo state network-based metamodel for the problem of stagnant petroleum cooling in pipe-in-pipe flowlines

A Hidden Markov Model-based fuzzy modeling of multivariate time series

A conformal predictive system for distribution regression with random features

Contact Info

Product

Resources

About