Optimizing a continuous action learning automata (CALA) optimizer for training artificial neural networks

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

Published: Feb. 6, 2025

Language: Английский

---

A Multilayer Perceptron Feedforward Neural Network and Particle Swarm Optimization Algorithm for Optimizing Biogas Production

Energies, Journal Year: 2025, Volume and Issue: 18(4), P. 1002 - 1002

Published: Feb. 19, 2025

Efficient biogas production significantly impacts greenhouse gas (GHG) emissions and carbon sequestration by reducing enhancing storage. Nonetheless, the consistency optimization of are hindered fluctuations in key input variables, namely, pH, moisture content, organic loading rate (OLR), temperature, which impact quality agricultural waste biomass production. Any these variables can affect productivity. This study aims to provide valuable parameters for maximum using rice straw cow dung as materials. Therefore, machine learning techniques such multilayer perceptron feedforward neural networks with a particle swarm (PSO) combination generate optimal values each variable uses three variants training function networks, namely gradient descent momentum adaptive rate, momentum, rate. The findings reveal that, under an optimum pH value 6.0000, humidity 62.3176%, OLR 67.6823 kg.m3/day, temperature 37.0482 °C, has potential increase 2.91 m³/day high accuracy testing R2 = 0.90. These methods use accurately predict parameters, deviation 8.48% from experimentally derived mean square error (MSE) 0.0051243. emphasizes benefits optimize operational

Language: Английский

---

Maximum energy entropy: A novel signal preprocessing approach for data-driven monthly streamflow forecasting

Ecological Informatics, Journal Year: 2023, Volume and Issue: 79, P. 102452 - 102452

Published: Dec. 28, 2023

In recent years, the application of Data-Driven Models (DDMs) in ecological studies has garnered significant attention due to their capacity accurately simulate complex hydrological processes. These models have proven invaluable comprehending and predicting natural phenomena. However, achieve improved outcomes, certain additive components such as signal analysis (SAM) input variable selections (IVS) are necessary. SAMs unveil hidden characteristics within time series data, while IVS prevents utilization inappropriate data. realm research, understanding these patterns is pivotal for grasping implications streamflow dynamics guiding effective management decisions. Addressing need more precise forecasting, this study proposes a novel SAM called "Maximum Energy Entropy (MEE)" forecast monthly Ajichai basin, located northwestern Iran. A comparative was conducted, pitting MEE against well-known methods Discreet Wavelet (DW) Wavelet-Entropy (DWE), ultimately demonstrating superiority MEE. The results showcased superior performance our proposed method, with an NSE value 0.72, compared DW (NSE 0.68) DWE 0.68). Furthermore, exhibited greater reliability, boasting lower Standard Deviation 0.13 (0.26) (0.19). equips researchers decision-makers accurate predictions, facilitating well-informed water resource planning. To further evaluate MEE's accuracy using various DDMs, we integrated Artificial Neural Network (ANN) Genetic Programming (GP). Additionally, GP served method selecting appropriate variables. Ultimately, combination ANN forecasting model (MEE-GP-ANN) yielded most favorable results.

Language: Английский

---

Differential Mutation Incorporated Quantum Honey Badger Algorithm with Dynamic Opposite Learning and Laplace Crossover for Fuzzy Front-End Product Design

Biomimetics, Journal Year: 2024, Volume and Issue: 9(1), P. 21 - 21

Published: Jan. 2, 2024

In this paper, a multi-strategy fusion enhanced Honey Badger algorithm (EHBA) is proposed to address the problem of easy convergence local optima and difficulty in achieving fast (HBA). The adoption dynamic opposite learning strategy broadens search area population, enhances global ability, improves population diversity. honey harvesting stage badger (development), differential mutation strategies are combined, selectively introducing quantum that enhance capabilities improve optimization accuracy, or Laplacian crossover operators can speed, while reducing odds HBA sinking into optima. Through comparative experiments with other algorithms on CEC2017, CEC2020, CEC2022 test sets, three engineering examples, EHBA has been verified have good solving performance. From analysis graphs, box plots, performance tests, it be seen compared eight algorithms, better results, significantly improving its ability application prospects field problems.

Language: Английский

---

Optimization algorithm analysis of EV waste battery recycling logistics based on neural network

Electrical Engineering, Journal Year: 2024, Volume and Issue: 106(2), P. 1403 - 1424

Published: March 1, 2024

Language: Английский

---

MTLBORKS-CNN: An Innovative Approach for Automated Convolutional Neural Network Design for Image Classification

Mathematics, Journal Year: 2023, Volume and Issue: 11(19), P. 4115 - 4115

Published: Sept. 28, 2023

Convolutional neural networks (CNNs) have excelled in artificial intelligence, particularly image-related tasks such as classification and object recognition. However, manually designing CNN architectures demands significant domain expertise involves time-consuming trial-and-error processes, along with substantial computational resources. To overcome this challenge, an automated network design method known Modified Teaching-Learning-Based Optimization Refined Knowledge Sharing (MTLBORKS-CNN) is introduced. It autonomously searches for optimal architectures, achieving high performance on specific datasets without human intervention. MTLBORKS-CNN incorporates four key features. employs effective encoding scheme various hyperparameters, facilitating the search innovative valid architectures. During modified teacher phase, it leverages a social learning concept to calculate unique exemplars that effectively guide learners while preserving diversity. In learner self-learning adaptive peer are incorporated enhance knowledge acquisition of during architecture optimization. Finally, dual-criterion selection scheme, considering both fitness diversity, determine survival subsequent generations. rigorously evaluated across nine image compared state-of-the-art methods. The results consistently demonstrate MTLBORKS-CNN’s superiority terms accuracy complexity, suggesting its potential infrastructural development smart devices.

Language: Английский

---

Emerging trends in computational swarm intelligence: A comprehensive overview

Elsevier eBooks, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 40

Published: Jan. 1, 2024

Language: Английский

---

Special Issue: Neural Networks, Fuzzy Systems and Other Computational Intelligence Techniques for Advanced Process Control

Processes, Journal Year: 2023, Volume and Issue: 11(8), P. 2278 - 2278

Published: July 28, 2023

Computational intelligence (CI) techniques have developed very fast over the past two decades, with many new methods emerging [...]

Language: Английский

---

Leveraging artificial intelligence for simplified adiabatic compression heating prediction: Comparing the use of artificial neural networks with conventional numerical approach

Innovative Food Science & Emerging Technologies, Journal Year: 2023, Volume and Issue: 91, P. 103546 - 103546

Published: Dec. 9, 2023

This study presents a comprehensive evaluation of artificial neural networks (ANNs) for predicting adiabatic compression heating in high-pressure processing (HPP) and thermal (HPTP). The ANNs were thoroughly compared with experimental data, as well data predicted by previously developed approach. previous approach used numerical methods to extract relevant material properties from which was then predictions solving an ordinary differential equation. new work showcases the efficacy accurately effects. results indicate that offer robust, flexible, efficient modelling heating, significant implications design optimisation HPTP treatments. underscores transformative potential AI advancing food preservation technologies.

Language: Английский

---

Optimization of land subsidence prediction features based on machine learning and SHAP value with Sentinel-1 InSAR Data

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 23, 2024

Abstract Land subsidence has always been a concern of geoscience, and exploring the factors affecting land to predict future is essential research. However, current research rarely scientific unified feature screening process for features. This study applies neural networks SHAP values prediction. We used instead traditional random forest (RF) quantify features areas where likely occur in cities Chongqing Chengdu, encompassing majority possible scenarios future. The results show that prediction using improves model accuracy by 16% compared method. After input optimization, performance nearly 22%. found optimization method based on proposed this more helpful prediction, derived from data analysis with complex terrain are also consistent previous studies. can contribute variable selection model, improve accuracy, provide solid theoretical support preventing urban subsidence.

Language: Английский

---