Machine learning-aided modeling for predicting freshwater production of a membrane desalination system: A long-short-term memory coupled with election-based optimizer

Alexandria Engineering Journal, Journal Year: 2023, Volume and Issue: 86, P. 690 - 703

Published: Dec. 28, 2023

Membrane desalination (MD) is an efficient process for desalinating saltwater, combining the uniqueness of both thermal and separation distillation configurations. In this context, optimization strategies sizing methodologies are developed from balance system's energy demand. Therefore, robust prediction modeling thermodynamic behavior freshwater production crucial optimal design MD systems. This study presents a new advanced machine-learning model to obtain permeate flux tubular direct contact membrane unit. The was established by optimizing long-short-term memory (LSTM) election-based algorithm (EBOA). inputs were temperatures feed flow, rate salinity flow. optimized compared with other LSTM models sine–cosine (SCA), artificial ecosystem optimizer (AEO), grey wolf (GWO). All trained, tested, evaluated using different accuracy measures. LSTM-EBOA outperformed in predicting based on had highest coefficient determination 0.998 0.988 lowest root mean square error 1.272 4.180 training test, respectively. It can be recommended that paper provide useful pathway parameters selection performance systems makes optimally designed rates without costly experiments.

Language: Английский

---

Hourly load prediction based feature selection scheme and hybrid CNN‐LSTM method for building's smart solar microgrid

Expert Systems, Journal Year: 2024, Volume and Issue: 41(7)

Published: Jan. 4, 2024

Abstract The short‐term load prediction is the critical operation in peak demand administration and power generation scheduling of buildings that integrated smart solar microgrid (SSM). Many research studies have proved hybrid deep learning strategies achieve more accuracy feasibility practical applications than individual algorithms. Moreover, many SSM on rooftop with battery management system (BMS) to enhance energy efficiency management. However, traditional methodologies only processed weather parameters information for prediction, ignoring collected data from BMS by advanced metering infrastructures (AMI), which probably improved accuracy. In this research, accumulated building are before methodology implementation. Considering diversities BMS, an adaptive convolution neural network long memory (CNN‐LSTM) proposed hourly electrical prediction. CNN could extract large‐scale input feature, while LSTM better accurate forecasts. Pearson correlation matrix calculated feature selection scheme different units. hyperparameter tuning utilized obtaining optimized CNN‐LSTM algorithm. K‐fold cross‐validation employed algorithm verification, includes LSTM, GRU, CNN, Bi‐LSTM methodologies. results prove achieved outperformed improvements, 20.57%, 29.63%, 19.06% MSE, MAE, MAPE, 21.24%, 22.02%, 3.82% validating respectively. combined superior predicting accuracies, proving adaptability ability integrating into (EMS) building's SSM.

Language: Английский

---

Application of artificial neural networks in predicting the performance of ice thermal energy storage systems

Journal of Energy Storage, Journal Year: 2024, Volume and Issue: 95, P. 112547 - 112547

Published: June 14, 2024

Language: Английский

---

Prediction model of the large commercial building cooling loads based on rough set and deep extreme learning machine

Journal of Building Engineering, Journal Year: 2023, Volume and Issue: 80, P. 107958 - 107958

Published: Oct. 17, 2023

Language: Английский

---

Sparrow search mechanism-based effective feature mining algorithm for the broken wire signal detection of prestressed concrete cylinder pipe

Mechanical Systems and Signal Processing, Journal Year: 2024, Volume and Issue: 212, P. 111270 - 111270

Published: Feb. 24, 2024

Language: Английский

---

Comparison of energy consumption prediction models for air conditioning at different time scales for large public buildings

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 96, P. 110423 - 110423

Published: Aug. 14, 2024

Language: Английский

---

A Review of Research on Building Energy Consumption Prediction Models Based on Artificial Neural Networks

Sustainability, Journal Year: 2024, Volume and Issue: 16(17), P. 7805 - 7805

Published: Sept. 7, 2024

Building energy consumption prediction models are powerful tools for optimizing management. Among various methods, artificial neural networks (ANNs) have become increasingly popular. This paper reviews studies since 2015 on using ANNs to predict building use and demand, focusing the characteristics of different ANN structures their applications across phases—design, operation, retrofitting. It also provides guidance selecting most appropriate each phase. Finally, this explores future developments in ANN-based predictions, including improving data processing techniques greater accuracy, refining parameterization better capture features, algorithms faster computation, integrating with other machine learning such as ensemble hybrid models, enhance predictive performance.

Language: Английский

---

Time series analysis in compressor-based machines: a survey

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: March 5, 2025

Language: Английский

---

AIoT-based Indoor Air Quality Prediction for Building Using Enhanced Metaheuristic Algorithm and Hybrid Deep Learning

Journal of Building Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 112448 - 112448

Published: March 1, 2025

Language: Английский

---

Energy Consumption Prediction for Water-Based Thermal Energy Storage Systems Using an Attention-Based TCN-LSTM Model

Sustainable Cities and Society, Journal Year: 2025, Volume and Issue: unknown, P. 106383 - 106383

Published: April 1, 2025

Language: Английский

---