AMBWO: An Augmented Multi-Strategy Beluga Whale Optimization for Numerical Optimization Problems DOI Creative Commons

Guoping You,

Zhong Lu, Zhipeng Qiu

et al.

Biomimetics, Journal Year: 2024, Volume and Issue: 9(12), P. 727 - 727

Published: Nov. 28, 2024

Beluga whale optimization (BWO) is a swarm-based metaheuristic algorithm inspired by the group behavior of beluga whales. BWO suffers from drawbacks such as an insufficient exploration capability and tendency to fall into local optima. To address these shortcomings, this paper proposes augmented multi-strategy (AMBWO). The adaptive population learning strategy proposed improve global BWO. introduction roulette equilibrium selection allows have more reference points choose among during exploitation phase, which enhances flexibility algorithm. In addition, avoidance improves algorithm’s ability escape optima enriches quality. order validate performance AMBWO, extensive evaluation comparisons with other state-of-the-art improved algorithms were conducted on CEC2017 CEC2022 test sets. Statistical tests, convergence analysis, stability analysis show that AMBWO exhibits superior overall performance. Finally, applicability superiority was further verified several engineering problems.

Language: Английский

A survey of Beluga whale optimization and its variants: Statistical analysis, advances, and structural reviewing DOI
Sang-Woong Lee, Amir Haider, Amir Masoud Rahmani

et al.

Computer Science Review, Journal Year: 2025, Volume and Issue: 57, P. 100740 - 100740

Published: March 3, 2025

Language: Английский

Citations

0
Multi-Scale Graph Attention Network Based on Encoding Decomposition for Electricity Consumption Prediction DOI Creative Commons
Sheng Huang,

Huakun Que,

Lukun Zeng

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(23), P. 5813 - 5813

Published: Nov. 21, 2024

Accurate electricity consumption forecasting is essential for power scheduling. In short-term forecasting, data exhibit periodic patterns, as well fluctuations associated with production events. Traditional methods typically focus on sequential features of the data, which may lead to an over-smoothing issue fluctuations. practice, these events tend follow recognizable patterns. By emphasizing impact experiential current prediction process, we can capture volatility variations alleviate problem. To this end, propose encoding decomposition-based multi-scale graph neural network (CMNN). The CMNN starts by decomposing into various components. For high-order components that approximate behavior, designs a Multi-scale Bi-directional Long Short-Term Memory (MBLSTM) fitting and prediction. low-order fluctuations, transforms from one-dimensional time series two-dimensional component model components, proposes Gaussian Graph Auto-Encoder forecast Finally, combines predicted produce final Experiments demonstrate enhances accuracy predictions.

Language: Английский

Citations

0
AMBWO: An Augmented Multi-Strategy Beluga Whale Optimization for Numerical Optimization Problems DOI Creative Commons

Guoping You,

Zhong Lu, Zhipeng Qiu

et al.

Biomimetics, Journal Year: 2024, Volume and Issue: 9(12), P. 727 - 727

Published: Nov. 28, 2024

Beluga whale optimization (BWO) is a swarm-based metaheuristic algorithm inspired by the group behavior of beluga whales. BWO suffers from drawbacks such as an insufficient exploration capability and tendency to fall into local optima. To address these shortcomings, this paper proposes augmented multi-strategy (AMBWO). The adaptive population learning strategy proposed improve global BWO. introduction roulette equilibrium selection allows have more reference points choose among during exploitation phase, which enhances flexibility algorithm. In addition, avoidance improves algorithm’s ability escape optima enriches quality. order validate performance AMBWO, extensive evaluation comparisons with other state-of-the-art improved algorithms were conducted on CEC2017 CEC2022 test sets. Statistical tests, convergence analysis, stability analysis show that AMBWO exhibits superior overall performance. Finally, applicability superiority was further verified several engineering problems.

Language: Английский

Citations

0