Advancing solar energy forecasting with modified ANN and light GBM learning algorithms

AIMS energy, Год журнала: 2024, Номер 12(2), С. 350 - 386

Опубликована: Янв. 1, 2024

<abstract> <p>In the evolving field of solar energy, precise forecasting Solar Irradiance (SI) stands as a pivotal challenge for optimization photovoltaic (PV) systems. Addressing inadequacies in current techniques, we introduced advanced machine learning models, namely Rectified Linear Unit Activation with Adaptive Moment Estimation Neural Network (RELAD-ANN) and Support Vector Machine Individual Parameter Features (LSIPF). These models broke new ground by striking an unprecedented balance between computational efficiency predictive accuracy, specifically engineered to overcome common pitfalls such overfitting data inconsistency. The RELAD-ANN model, its multi-layer architecture, sets standard detecting nuanced dynamics SI meteorological variables. By integrating sophisticated regression methods like Regression (SVR) Lightweight Gradient Boosting Machines (Light GBM), our results illuminated intricate relationship influencing factors, marking novel contribution domain energy forecasting. With R² 0.935, MAE 8.20, MAPE 3.48%, model outshone other signifying potential accurate reliable forecasting, when compared existing Multi-Layer Perceptron, Long Short-Term Memory (LSTM), Multilayer-LSTM, Gated Recurrent Unit, 1-dimensional Convolutional Network, while LSIPF showed limitations ability. Light GBM emerged robust approach evaluating environmental influences on SI, outperforming SVR model. Our findings contributed significantly systems could be applied globally, offering promising direction renewable management real-time forecasting.</p> </abstract>

Язык: Английский

---

Combined deep-learning optimization predictive models for determining carbon dioxide solubility in ionic liquids

Journal of Industrial Information Integration, Год журнала: 2024, Номер 41, С. 100662 - 100662

Опубликована: Июль 11, 2024

Язык: Английский

---

An integrated network topology and deep learning model for prediction of Alzheimer disease candidate genes

Soft Computing, Год журнала: 2023, Номер 27(19), С. 14189 - 14203

Опубликована: Май 15, 2023

Язык: Английский

---

An energy-efficient unrelated parallel machine scheduling problem with learning effect of operators and deterioration of jobs

International Journal of Environmental Science and Technology, Год журнала: 2024, Номер 21(15), С. 9651 - 9676

Опубликована: Апрель 15, 2024

Язык: Английский

---

A damping grey multivariable model and its application in online public opinion prediction

Engineering Applications of Artificial Intelligence, Год журнала: 2022, Номер 118, С. 105661 - 105661

Опубликована: Дек. 8, 2022

Язык: Английский

---

GTSNet: Flexible architecture under budget constraint for real-time human activity recognition from wearable sensor

Engineering Applications of Artificial Intelligence, Год журнала: 2023, Номер 124, С. 106543 - 106543

Опубликована: Июнь 7, 2023

Human activity recognition is an essential task for human-centered intelligent systems such as healthcare and smart vehicles, which can be accomplished by analyzing time-series signals collected from sensors in wearable devices. In these applications, real-time response vital because prompt action necessary urgent events elderly person falling or driving while drowsy. Although recurrent neural networks have been widely used owing to their temporal modeling capabilities, recent studies focused on convolutional (CNNs) that are suitable responses they incur lower computational costs. However, CNNs with a manual design may fail achieve optimal accuracy due varying budgets applications this paper, we propose novel framework uses mathematical approach derive CNN architecture given budget. As result, introduce grouped shift network (GTSNet) the flexibly modified predefining theoretical computation cost. We demonstrate effectiveness of our experiments, achieving best performance well-known public benchmark datasets under limited budgets. The source codes GTSNet publicly available at https://github.com/jgpark92/GTSNet.

Язык: Английский

---

Evaluation of Prediction Models of the Microwire EDM Process of Inconel 718 Using ANN and RSM Methods

Materials, Год журнала: 2022, Номер 15(23), С. 8317 - 8317

Опубликована: Ноя. 23, 2022

Precise machining of micro parts from difficult-to-cut materials requires using advanced technology such as wire electrical discharge (WEDM). In order to enhance the productivity WEDM, key role is understanding influence process parameters on surface topography and material’s removal rate (MRR). Furthermore, effective models which allow us predict micro-WEDM qualitative effects are required. This paper influences energy, time interval, speed topography’s properties, namely Sa, Sk, Spk, Svk, MRR, after Inconel 718 were described. Developed RSM ANN model process, showing that energy had main (over 70%) parameters. However, for interval was also significant. a reduction in can lead decrease cost have positive environment sustainability process. Evaluation developed prediction indicates lower value relative error compared with did not exceed 4%.

Язык: Английский

---

Metaheuristic Algorithms and Their Applications in Different Fields

Опубликована: Март 29, 2024

A potent method for resolving challenging optimization issues is provided by metaheuristic algorithms, which are heuristic approaches. They provide an effective technique to explore huge solution spaces and identify close ideal or optimal solutions. iterative often inspired natural social processes. This study provides comprehensive information on algorithms the many areas in they used. Heuristic well-known their success handling issues. a tool problem-solving. Twenty such as tabu search, particle swarm optimization, ant colony genetic simulated annealing, harmony included article. The article extensively explores applications of these diverse domains engineering, finance, logistics, computer science. It underscores particular instances where have found utility, optimizing structural design, controlling dynamic systems, enhancing manufacturing processes, managing supply chains, addressing problems artificial intelligence, data mining, software engineering. paper thorough insight into versatile deployment across different sectors, highlighting capacity tackle complex wide range real-world scenarios.

Язык: Английский

---

Short-term Gini coefficient estimation using nonlinear autoregressive multilayer perceptron model

Heliyon, Год журнала: 2024, Номер 10(4), С. e26438 - e26438

Опубликована: Фев. 1, 2024

Poverty, an intricate global challenge influenced by economic, political, and social elements, is characterized a deficiency in crucial resources, necessitating collective efforts towards its mitigation as embodied the United Nations' Sustainable Development Goals. The Gini coefficient statistical instrument used nations to measure income inequality, economic status, disparity, escalated inequality often parallels high poverty rates. Despite standard annual computation, impeded logistical hurdles gradual transformation of we suggest that short-term forecasting could offer instantaneous comprehension shifts during swift transitions, such variances due seasonal employment patterns expanding gig economy. System Identification (SI), methodology utilized domains like engineering mathematical modeling construct or refine dynamic system models from captured data, relies significantly on Nonlinear Auto-Regressive (NAR) model reliability capability integrating nonlinear functions, complemented contemporary machine learning strategies computational algorithms approximate complex dynamics address these limitations. In this study, introduce NAR Multi-Layer Perceptron (MLP) approach for brief term estimation coefficient. Several parameters were tested discover optimal Malaysia's within 1987-2015, namely output lag space, hidden units, initial random seeds. One-Step-Ahead (OSA), residual correlation, histograms test validity model. results demonstrate model's efficacy over 28-year period with superior fit (MSE: 1.14 × 10

Язык: Английский

---

An integrated weighted multi-criteria decision making method using Z-number and its application in failure modes and effect analysis

Journal of Industrial Information Integration, Год журнала: 2025, Номер unknown, С. 100805 - 100805

Опубликована: Март 1, 2025

Язык: Английский

---