Smart Monitoring and Diagnostics for Fault Detection in Power Plant Equipment DOI

Shawetangi Kala,

Pradeep Kumar S, G. Meerimatha

et al.

Published: April 26, 2024

In recent years, Fault Diagnosis and Identification (FDI) has grown significantly. Generally used methodology in these types of frameworks include model-based approaches, fault pattern recognition techniques. However, it was essential to society's energy supply, therefore maintaining its safe effective functioning is utmost importance. This research sophisticated Detection (FDD) methods diagnose the problems contemporary power plants order decrease maintenance shutdowns expenses. Therefore, this applied deep learning image processing using efficient Long Short-Term Memory (LSTM)-based for object categorization. Using observed frequency as input, trained LSTM utilized detect variations real time. Control devices, such synchronous generators Energy Storage Systems (ESSs), maintain a steady by detected variations. According results, suggested approach demonstrated superior performance terms LG faults, with 0.427%, LLG faults 0.800%, LGG 0.186 %, LL 0.741%. Compared existing namely Sparrow Search Algorithm (SSA) Deep (LSTM) shows that are achieved better performances less respectively.

Language: Английский

Integration of Electric Vehicle Charging Stations and DSTATCOM in Practical Indian Distribution Systems Using Bald Eagle Search Algorithm DOI Creative Commons

T. Yuvaraj,

K. R. Devabalaji,

Thanikanti Sudhakar Babu

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 55149 - 55168

Published: Jan. 1, 2023

Because of the increasing growth Electric Vehicle (EV) in India, more electricity is required to power such vehicles. It also gaining popularity because its low maintenance, improved performance, and zero carbon impact. As usage electric vehicles grows, distribution system's performance impacted. an outcome, reliability system (DS) dependent on position vehicle charging station (EVCS). The fundamental difficulty deterioration DS due incorrect EVCS location. linked works with static compensator (DSTATCOM) minimize impact EVCS. A new nature-inspired Bald Eagle Search Algorithm (BESA) based optimization technique was utilized find optimal allocation DSTATCOM DS. proposed strategy for mitigating real loss has been tested practical Indian 28-bus 108-bus networks. Power reduction optimizes net savings, voltage stability, bus voltage. test case findings show that BESA-based accurate regarding mitigation, enhancement, annual saving improvement than BA-based

Language: Английский

Citations

41
Bald eagle search algorithm: a comprehensive review with its variants and applications DOI Creative Commons
M.A. El‐Shorbagy, Anas Bouaouda, Hossam A. Nabwey

et al.

Systems Science & Control Engineering, Journal Year: 2024, Volume and Issue: 12(1)

Published: Aug. 1, 2024

Bald Eagle Search (BES) is a recent and highly successful swarm-based metaheuristic algorithm inspired by the hunting strategy of bald eagles in capturing prey. With its remarkable ability to balance global local searches during optimization, BES effectively addresses various optimization challenges across diverse domains, yielding nearly optimal results. This paper offers comprehensive review research on BES. Beginning with an introduction BES's natural inspiration conceptual framework, it explores modifications, hybridizations, applications domains. Then, critical evaluation performance provided, offering update effectiveness compared recently published algorithms. Furthermore, presents meta-analysis developments outlines potential future directions. As swarm-inspired algorithms become increasingly important tackling complex problems, this study valuable resource for researchers aiming understand algorithms, mainly focusing comprehensively. It investigates evolution, exploring solving intricate fields.

Language: Английский

Citations

12
Modified bald eagle search algorithm for lithium-ion battery model parameters extraction DOI
Seydali Ferahtia, Hegazy Rezk,

Ali Djerioui

et al.

ISA Transactions, Journal Year: 2022, Volume and Issue: 134, P. 357 - 379

Published: Aug. 30, 2022

Language: Английский

Citations

29
Self‐paced learning long short‐term memory based on intelligent optimization for robust wind power prediction DOI Creative Commons

Shun Yang,

Xiaofei Deng, Dongran Song

et al.

IET Control Theory and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: April 3, 2024

Abstract Given the unpredictable and intermittent nature of wind energy, precise forecasting power is crucial for ensuring safe stable operation systems. To reduce influence noise data on robustness prediction, a prediction method proposed that leverages an enhanced multi‐objective sand cat swarm algorithm (MO‐SCSO) self‐paced long short‐term memory network (spLSTM). First, actual processed into time series as input output. Then, progressive advantage learning used to effectively solve instability caused by noisy during (LSTM) training. Following this, improved MO‐SCSO employed iteratively optimize hyperparameters spLSTM. Ultimately, combined MO‐SCSO‐spLSTM model constructed prediction. This validated with onshore farms in Austria offshore Denmark. The experimental results show compared traditional LSTM method, has better accuracy robustness. Specifically, experiments, reduces minimum MAE 5.44% 4.96%, respectively, range 4.45% 17.21%, which could be conducive system.

Language: Английский

Citations

5
Time-Series Power Forecasting for Wind and Solar Energy Based on the SL-Transformer DOI Creative Commons

Jian Zhu,

Zhiyuan Zhao, Xiaoran Zheng

et al.

Energies, Journal Year: 2023, Volume and Issue: 16(22), P. 7610 - 7610

Published: Nov. 16, 2023

As the urgency to adopt renewable energy sources escalates, so does need for accurate forecasting of power output, particularly wind and solar power. Existing models often struggle with noise temporal intricacies, necessitating more robust solutions. In response, our study presents SL-Transformer, a novel method rooted in deep learning paradigm tailored green forecasting. To ensure reliable basis further analysis modeling, free from outliers, we employed SG filter LOF algorithm data cleansing. Moreover, incorporated self-attention mechanism, enhancing model’s ability discern dynamically fine-tune input weights. When benchmarked against other premier models, SL-Transformer distinctly outperforms them. Notably, it achieves near-perfect R2 value 0.9989 significantly low SMAPE 5.8507% predictions. For forecasting, has achieved 4.2156%, signifying commendable improvement 15% over competing models. The experimental results demonstrate efficacy

Language: Английский

Citations

12
Development of a hybrid LSTM with chimp optimization algorithm for the pressure ventilator prediction DOI Creative Commons
Fatma Refaat Ahmed, Samira Ahmed Alsenany, Sally Mohammed Farghaly Abdelaliem

et al.

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Nov. 27, 2023

The utilization of mechanical ventilation is utmost importance in the management individuals afflicted with severe pulmonary conditions. During periods a pandemic, it becomes imperative to build ventilators that possess capability autonomously adapt parameters over course treatment. In order fulfil this requirement, research investigation was undertaken aim forecasting magnitude pressure applied on patient by ventilator. aforementioned forecast derived from comprehensive analysis many variables, including ventilator's characteristics and patient's medical state. This conducted utilizing sophisticated computational model referred as Long Short-Term Memory (LSTM). To enhance predictive accuracy LSTM model, researchers utilized Chimp Optimization method (ChoA) method. integration ChoA led development LSTM-ChoA which successfully tackled issue hyperparameter selection for model. experimental results revealed exhibited superior performance compared alternative optimization algorithms, namely whale grey wolf optimizer (GWO), algorithm (WOA), particle swarm (PSO). Additionally, outperformed regression models, K-nearest neighbor (KNN) Regressor, Random Forest (RF) Support Vector Machine (SVM) accurately predicting ventilator pressure. findings indicate suggested LSTM-ChoA, demonstrates reduced mean square error (MSE) value. Specifically, when comparing GWO, MSE fell around 14.8%. Furthermore, PSO WOA, decreased approximately 60%. variance (ANOVA) p-value 0.000, less than predetermined significance level 0.05. indicates are statistically significant.

Language: Английский

Citations

10
Ultra-short-term Wind power prediction algorithm based on bidirectional neural controlled differential equations DOI

Chu Li,

Bingjia Xiao, Qiping Yuan

et al.

Electric Power Systems Research, Journal Year: 2025, Volume and Issue: 243, P. 111479 - 111479

Published: Feb. 12, 2025

Language: Английский

Citations

0
Optimizing cavitation fault forecasting in pump-turbine systems through gated recurrent units DOI

Yi Zhao,

Shengnan Yan, Wangxu Li

et al.

Physics of Fluids, Journal Year: 2025, Volume and Issue: 37(3)

Published: March 1, 2025

The pressure fluctuation data in the pump-turbine runner region exhibit significant nonlinearity. method of utilizing neural networks is employed to analyze fluctuations order determine occurrence cavitation phenomena. This paper presents a model that utilizes VMD (variational mode decomposition)-optimized algorithm combined with GRU (gated recurrent unit)–attention for prediction fluctuations, aiming facilitate forecasting cavitation-induced failures. Using collected from real machine over course one day, predictions were made using three different models: standalone GRU, combination and Attention mechanisms, optimization algorithms. evaluation performance indicates VMD–dung beetle optimization–GRU–attention not only captures nonlinear characteristics actual values but also aligns more closely trend data. error assessment results demonstrate this exhibits superior predictive performance. Analyze pulsation at locations between guide vane, top bottom cover. enables effective conditions up 50 minutes advance, showcasing its potential practical engineering applications.

Language: Английский

Citations

0
AQIPred: A Hybrid Model for High Precision Time Specific Forecasting of Air Quality Index with Cluster Analysis DOI Creative Commons
Farhana Yasmin, Md. Mehedi Hassan, Mahade Hasan

et al.

Human-Centric Intelligent Systems, Journal Year: 2023, Volume and Issue: 3(3), P. 275 - 295

Published: Aug. 7, 2023

Abstract The discipline of forecasting and prediction is witnessing a surge in the application these techniques as direct result strong empirical performance that approaches based on machine learning (ML) have shown over past few years. Especially to predict wind direction, air water quality, flooding. In context doing this research, an MLP-LSTM Hybrid Model was developed be able generate predictions nature. An investigation into Beijing Multi-Site Air-Quality Data Set carried out experiment. particular scenario, model generated MSE values came at 0.00016, MAE 0.00746, RMSE 13.45, MAPE 0.42, R 2 0.95. This indication functioning effectively. conventional modeling for forecasting, do not give level required. On other hand, results study will useful any type time-specific requires high accuracy.

Language: Английский

Citations

9
Multi-threshold segmentation of grayscale and color images based on Kapur entropy by bald eagle search optimization algorithm with horizontal crossover and vertical crossover DOI
Guoyuan Ma, Xiaofeng Yue, Juan Zhu

et al.

Soft Computing, Journal Year: 2023, Volume and Issue: 27(20), P. 14759 - 14790

Published: June 24, 2023

Language: Английский

Citations

8