Multi-Strategy Improved Aquila Optimizer Algorithm and Its Application in Railway Freight Volume Prediction DOI Open Access
Lei Bai, Z. Y. Pei, Jia-Sheng Wang

и другие.

Electronics, Год журнала: 2025, Номер 14(8), С. 1621 - 1621

Опубликована: Апрель 17, 2025

This study proposes a multi-strategy improved Aquila optimizer (MIAO) to address the key limitations of original (AO). First, phasor operator is introduced eliminate excessive control parameters in X2 phase, transforming it into an adaptive parameter-free process. Second, flow direction enhances X3 phase by improving population diversity and local exploitation. The MIAO algorithm applied optimize Long Short-Term Memory (LSTM) hyperparameters, forming MIAO_LSTM model for monthly railway freight forecasting. Comprehensive evaluations on 15 benchmark functions show MIAO’s superior performance over SOA, PSO, SSA, AO. Using data (2005–2021), achieves lower MAE, MSE, RMSE compared traditional LSTM hybrid models (SSA_LSTM, PSO_LSTM, etc.). Further, Grey Relational Analysis selects high-correlation features (≥0.8) boost accuracy. results validate MIAO_LSTM’s effectiveness practical predictions.

Язык: Английский

Multi-Strategy Improved Aquila Optimizer Algorithm and Its Application in Railway Freight Volume Prediction DOI Open Access
Lei Bai, Z. Y. Pei, Jia-Sheng Wang

и другие.

Electronics, Год журнала: 2025, Номер 14(8), С. 1621 - 1621

Опубликована: Апрель 17, 2025

This study proposes a multi-strategy improved Aquila optimizer (MIAO) to address the key limitations of original (AO). First, phasor operator is introduced eliminate excessive control parameters in X2 phase, transforming it into an adaptive parameter-free process. Second, flow direction enhances X3 phase by improving population diversity and local exploitation. The MIAO algorithm applied optimize Long Short-Term Memory (LSTM) hyperparameters, forming MIAO_LSTM model for monthly railway freight forecasting. Comprehensive evaluations on 15 benchmark functions show MIAO’s superior performance over SOA, PSO, SSA, AO. Using data (2005–2021), achieves lower MAE, MSE, RMSE compared traditional LSTM hybrid models (SSA_LSTM, PSO_LSTM, etc.). Further, Grey Relational Analysis selects high-correlation features (≥0.8) boost accuracy. results validate MIAO_LSTM’s effectiveness practical predictions.

Язык: Английский

Процитировано

0