Article
Computer Science, Artificial Intelligence
M. Tanveer, A. Tiwari, R. Choudhary, M. A. Ganaie
Summary: This study proposes a novel large scale pinball twin support vector machine (LPTWSVM) to address the limitations of the twin support vector machines (TWSVMs), using a unique pinball loss function and improving model performance by eliminating matrix inversion calculation and minimizing structural risk.
Article
Computer Science, Artificial Intelligence
Chun-Na Li, Yuan-Hai Shao, Huajun Wang, Yu-Ting Zhao, Naihua Xiu, Nai-Yang Deng
Summary: This paper investigates the general forms and characteristics of nonparallel support vector machines (NSVMs) and categorizes them into two types. It reveals the advantages and defects of different types and points out the inconsistency problems. Based on this observation, a novel max-min distance-based NSVM is proposed with desired consistency. The proposed NSVM has the consistency of training and test and the consistency of metric, and it assigns each sample an ascertained loss.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Xiaoming Wang, Shitong Wang, Zengxi Huang, Yajun Du
Summary: This paper introduces a novel method called sparse support vector machine guided by radius-margin bound (RMB-SSVM) to efficiently condense the basis vectors in support vector machines. By selecting basis vectors and learning corresponding coefficients with a criterion related to SVM's generalization ability, the RMB-SSVM model can yield better performance.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Sambhav Jain, Reshma Rastogi
Summary: This paper proposes Parametric non-parallel support vector machines for binary pattern classification. The model brings noise resilience and sparsity by intelligently redesigning the Support vector machine optimization. The experimental results validate its scalability for large scale problems.
Article
Computer Science, Information Systems
Sebastian Maldonado, Julio Lopez, Carla Vairetti
Summary: The predictive performance of classification methods relies heavily on the nature of the environment and dataset shift issue. A novel Fuzzy Support Vector Machine strategy is proposed in this paper to improve performance by redefining the loss function and applying aggregation operators to deal with dataset shift. Our methods outperform traditional classifiers in terms of out-of-time prediction using simulated and real-world dataset for credit scoring.
INFORMATION SCIENCES
(2021)
Article
Energy & Fuels
Mathieu David, Joaquin Alonso-Montesinos, Josselin Le Gal La Salle, Philippe Lauret
Summary: The objective of this work is to propose a methodology to generate probabilistic solar forecasts as a binary event for very short-term horizons. By combining state-of-the-art models based on sky imagery and discrete choice models, as well as techniques like random forest, the quality of the forecasts can be significantly improved.
Article
Computer Science, Artificial Intelligence
Wangyong Lv, Tingting Li, Huali Ren, Shijing Zeng, Jiao Zhou
Summary: The IDH-MSVM algorithm adjusts the distance between hyperplanes and classical margins to handle multiclassification problems more flexibly. Experimental results on UCI standard data sets show that this method achieves better classification accuracy for multiclass data compared to other algorithms.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Energy & Fuels
Dragana Nikodinoska, Mathias Kaeso, Felix Muesgens
Summary: Accurate renewable energy feed-in forecasts are crucial for enhancing the efficiency of renewable energy integration, and combining forecasting models can improve accuracy. The dynamic elastic net estimation method can enhance the accuracy of photovoltaic and wind energy forecasts.
Article
Astronomy & Astrophysics
Miguel R. Alarcon, Miquel Serra-Ricart, Samuel Lemes-Perera, Manuel Mallorquin
Summary: In 2018, during Solar Cycle 24's solar minimum phase, 11 million zenithal night sky brightness (NSB) data were collected at different dark sites worldwide. By applying new filters, the natural NSB was calculated from 44 different photometers, revealing short-term variations related to airglow events forming above the mesosphere.
ASTRONOMICAL JOURNAL
(2021)
Article
Agronomy
Junliang Fan, Jing Zheng, Lifeng Wu, Fucang Zhang
Summary: Accurate estimation of plant transpiration (T) is crucial for agricultural production, and this study investigated the use of machine learning models to estimate daily T of summer maize. Incorporating soil water content and leaf area index variables improved model performance, with the deep neural network (DNN) model slightly outperforming others.
AGRICULTURAL WATER MANAGEMENT
(2021)
Article
Computer Science, Information Systems
Hunter Goddard, Lior Shamir
Summary: Deep convolutional neural networks (DCNNs) are superior in classifying image data with large labeled datasets, but SVMs provide higher accuracy when there are limited labeled images available.
Review
Computer Science, Information Systems
Arijit Chakraborty, Sajal Mitra, Debashis De, Anindya Jyoti Pal, Ferial Ghaemi, Ali Ahmadian, Massimiliano Ferrara
Summary: Protein-Protein Interaction (PPI) is a crucial network in biology that requires fast, accurate, and critical analysis, with Support Vector Machine (SVM) being an effective tool for solving complex classification problems.
Article
Mathematics
Roberto Barcenas, Maria Gonzalez-Lima, Joaquin Ortega, Adolfo Quiroz
Summary: The effectiveness of subsampling methods in reducing the required instances in the training stage of using support vector machines (SVMs) for classification in big data scenarios is explored in this paper with theoretical results. The main theorem states that, under certain conditions, a feasible solution can be found for the SVM problem using a randomly chosen subsample, which can be as close as desired to the classifier trained with the complete dataset in terms of classification error. Additionally, a new subsampling method called importance sampling and bagging is proposed, which provides a faster solution to the SVM problem without significant loss in accuracy compared to existing techniques.
Article
Computer Science, Artificial Intelligence
Kai Qi, Hu Yang
Summary: The article introduces a new support vector machine model (ENNHSVM) that constructs two nonparallel classifying hyperplanes using elastic net penalty for slack variables, aiming to improve the consistency and prediction accuracy of classifiers. The theoretical properties of ENNHSVM are discussed and a violation tolerance upper bound is derived to better demonstrate the relative violations of training samples.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Huiru Wang, Jiayi Zhu, Feng Feng
Summary: In this paper, a new classifier called elastic net twin support vector machine (ETSVM) is introduced to enhance classification performance. ETSVM resolves two smaller-sized quadratic programming problems (QPPs) similar to twin support vector machine (TSVM), but with the use of elastic net penalty for slack variables. The key difference is that ETSVM does not involve matrix inversion, avoiding ill-conditioning cases. Theoretical properties are discussed and safe screening rules (SSR-ETSVM) are derived to increase computing efficiency. Comparison with other methods confirms the rationality and effectiveness of the proposed algorithms.
INFORMATION SCIENCES
(2023)
Article
Engineering, Electrical & Electronic
Tor Anderson, Manasa Muralidharan, Priyank Srivastava, Hamed Valizadeh Haghi, Jorge Cortes, Jan Kleissl, Sonia Martinez, Byron Washom
Summary: This paper demonstrates coordinated and distributed resource control for secondary frequency response in a power distribution grid, utilizing a distributed control setup and various algorithms to achieve accurate and fast real-time computation of optimization solutions and effective tracking of regulation signals. Economic benefit analysis confirms eligibility to participate in an ancillary services market and shows potential annual revenue up to $53,000 for the selected distributed energy resources population.
IEEE TRANSACTIONS ON SMART GRID
(2021)
Article
Energy & Fuels
Patrick Mathiesen, Michael Stadler, Jan Kleissl, Zachary Pecenak
Summary: The intra-hour intermittency of solar energy and demand presents challenges for microgrid design, with the need to address energy shortfalls and runtime issues. This research introduces a new method to optimize DER investments and dispatch planning using hourly data, incorporating variability to improve optimization times.
Article
Green & Sustainable Science & Technology
Shiyi Liu, Sushil Silwal, Jan Kleissl
Summary: Battery energy storage systems (BESSs) are commonly used for demand charge reduction. This study investigates the error in demand charge reduction caused by coarse-resolution modeling and compares the effects of different temporal resolutions on peak load reduction. The battery rating space is divided into oversized, power-constrained, and energy-constrained regions. The study finds that the sequence effect can cause energy-constrained batteries to underestimate peak shaving and demand charge reduction. Careful consideration of the battery power capacity is necessary when interpreting optimization results at low resolutions.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2022)
Article
Energy & Fuels
Fuying Chen, Qing Yang, Niting Zheng, Yuxuan Wang, Junling Huang, Lu Xing, Jianlan Li, Shuanglei Feng, Guoqian Chen, Jan Kleissl
Summary: The potential of concentrated solar power (CSP) technology in China was evaluated, taking into account the geographical, technical, and CO2 emission reduction aspects. The results show that China has a vast amount of suitable land for CSP development, and replacing current power supply with CSP could significantly reduce CO2 emissions. The study provides policy guidance and serves as a reference for future CSP development and site selection.
Article
Energy & Fuels
Laetitia Uwineza, Hyun-Goo Kim, Jan Kleissl, Chang Ki Kim
Summary: This study introduces a dispatch algorithm specifically designed to minimize the NPC by maximizing the usage of FCs in HESs. The algorithm resolves the deficiencies of existing algorithms and achieves cost savings and higher electricity production.
Article
Green & Sustainable Science & Technology
Avik Ghosh, Monica Zamora Zapata, Sushil Silwal, Adil Khurram, Jan Kleissl
Summary: This paper addresses the gap in the literature on the effects of electric vehicles on total electricity costs in commercial buildings by incorporating different charging methods. The study found that V2B charging costs are lower than V1G, and with longer layover times, more V2B charging stations can be installed without exceeding original electricity costs. Sensitivity analysis showed that total electricity costs are influenced by the final state of charge of the EVs.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2022)
Article
Green & Sustainable Science & Technology
Melvin Lugo-Alvarez, Jan Kleissl, Adil Khurram, Matthew Lave, C. Birk Jones
Summary: This paper develops a systematic procedure to address the engineering challenge of reducing the duration and frequency of blackouts in remote communities by creating microgrids and prioritizing high value assets within vulnerable communities. The study uses nighttime satellite imagery to identify vulnerable communities, and employs an asset classification and rating system to prioritize multi-asset clusters within these communities. Infrastructure data, geographic information systems, satellite imagery, and spectral clustering are utilized to form and rank microgrid candidates, with a microgrid sizing algorithm included to guide the design process.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2022)
Article
Green & Sustainable Science & Technology
Monica Zamora Zapata, Jan Kleissl
Summary: Solar variability statistics are compared at different time intervals and classified by cloud categories. The study finds that longer time intervals overestimate the mean clear sky index of low and mid-clouds.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2022)
Editorial Material
Green & Sustainable Science & Technology
Jan Kleissl
Summary: This article discusses the issues of soiling and abrasion in solar energy systems. It covers climatological analyses, metrology, and best installation practices to reduce soiling and abrasion, as well as improvements to equipment and materials to mitigate these issues.
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY
(2022)
Review
Green & Sustainable Science & Technology
Dazhi Yang, Wenting Wang, Christian A. Gueymard, Tao Hong, Jan Kleissl, Jing Huang, Marc J. Perez, Richard Perez, Jamie M. Bright, Xiang'ao Xia, Dennis van der Meer, Ian Marius Peters
Summary: The ability to forecast solar irradiance is crucial for planning and operating power systems under high solar power generation. The collaboration between atmospheric scientists and power engineers is necessary to increase solar penetration while maintaining grid stability. This review discusses the current state and technical aspects of solar forecasting, as well as potential research topics for atmospheric scientists. A pathway towards high PV penetration and long-term carbon neutrality is also presented.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2022)
Article
Green & Sustainable Science & Technology
Yinghao Chu, Mengying Li, Hugo T. C. Pedro, Carlos F. M. Coimbra
Summary: A low-cost network of cameras has been installed in the Los Angeles basin for spatial solar irradiance assessments. An algorithm called Image to Irradiance (I2I) is proposed to derive high-resolution diffuse, direct, and global solar irradiance from sky images. The network of cameras provides more accurate spatially resolved Global Horizontal Irradiance (GHI) compared to satellite images when the distance to the nearest site is less than 40 km.
Article
Computer Science, Information Systems
Yinghao Chu, Daquan Feng, Zuozhu Liu, Zizhou Zhao, Zhenzhong Wang, Xiang-Gen Xia, Tony Q. S. Quek
Summary: This article presents an edge-computing-enabled IoT system based on a hybrid learning method for visual surface quality inspection. The method uses a deep neural network to synthesize global representations of industrial images and applies an unsupervised clustering algorithm for anomaly detection. With a small amount of labeled data and minimum iterative optimization efforts, the method achieves high accuracies and recalls in real-world factories.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Changfu Li, Yi-An Chen, Chenrui Jin, Ratnesh Sharma, Jan Kleissl
Summary: This paper proposes a method based on deep reinforcement learning algorithm to coordinate PV smart inverters. Through offline simulations and adaptation to load and solar variations, the method can intelligently coordinate multiple inverters to maintain grid voltage within the limits.
ELECTRIC POWER SYSTEMS RESEARCH
(2022)
Article
Computer Science, Information Systems
Yinghao Chu, Daquan Feng, Zuozhu Liu, Lei Zhang, Zizhou Zhao, Zhenzhong Wang, Zhiyong Feng, Xiang-Gen Xia
Summary: In this study, a low-cost edge computing-based IoT system is developed, which uses an innovative fine-grained attention model (FGAM) to enhance the accuracy of robot guidance and localization in manufacturing. The FGAM-based system shows superior performance compared to benchmark models and has been successfully deployed in a real-world factory for mass production. The deployed system achieves an average process and transmission time of 200 ms and an overall localization accuracy of up to 99.998%.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Daquan Feng, Junjie Peng, Yuan Zhuang, Chongtao Guo, Tingting Zhang, Yinghao Chu, Xiaoan Zhou, Xiang-Gen Xia
Summary: Indoor positioning system (IPS) is important in IoT applications, and Ultrawideband (UWB)-based IPS has shown superior performance. However, non-line-of-sight (NLOS) situations degrade accuracy. To address this, a SVM-based channel detection method is proposed to distinguish LOS and NLOS conditions, and a DAPA-EKF algorithm is proposed for NLOS environment. For LOS environment, LS-AEKF and LS-VEKF algorithms are developed. TDOA and KF are combined to further improve performance. Simulation results show improved positioning accuracy. LS-AEKF achieves 73.8%-74.1% higher accuracy than LS-VEKF.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Energy & Fuels
Siddharth Sradhasagar, Omkar Subhasish Khuntia, Srikanta Biswal, Sougat Purohit, Amritendu Roy
Summary: In this study, machine learning models were developed to predict the bandgap and its character of double perovskite materials, with LGBMRegressor and XGBClassifier models identified as the best predictors. These models were further employed to predict the bandgap of novel bismuth-based transition metal oxide double perovskites, showing high accuracy, especially in the range of 1.2-1.8 eV.
Article
Energy & Fuels
Wei Shuai, Haoran Xu, Baoyang Luo, Yihui Huang, Dong Chen, Peiwang Zhu, Gang Xiao
Summary: In this study, a hybrid model based on numerical simulation and deep learning is proposed for the optimization and operation of solar receivers. By applying the model to different application scenarios and considering multiple performance objectives, small errors are achieved and optimal structure parameters and heliostat scales are identified. This approach is not only applicable to gas turbines but also heating systems.
Article
Energy & Fuels
Mubashar Ali, Zunaira Bibi, M. W. Younis, Muhammad Mubashir, Muqaddas Iqbal, Muhammad Usman Ali, Muhammad Asif Iqbal
Summary: This study investigates the structural, mechanical, and optoelectronic properties of the BaCuF3 fluoroperovskite using the first-principles modelling approach. The stability and characteristics of different cubic structures of BaCuF3 are evaluated, and the alpha-BaCuF3 and beta-BaCuF3 compounds are found to be mechanically stable with favorable optical properties for solar cells and high-frequency UV applications.
Article
Energy & Fuels
Dong Le Khac, Shahariar Chowdhury, Asmaa Soheil Najm, Montri Luengchavanon, Araa mebdir Holi, Mohammad Shah Jamal, Chin Hua Chia, Kuaanan Techato, Vidhya Selvanathan
Summary: A novel recycling system is proposed in this study to decompose and reclaim the constituent materials of organic-inorganic perovskite solar cells (PSCs). By utilizing a one-step solution process extraction approach, the chemical composition of each layer is successfully preserved, enabling their potential reuse. The proposed recycling technique helps mitigate pollution risks, minimize waste generation, and reduce recycling costs.
Article
Energy & Fuels
Peijie Lin, Feng Guo, Xiaoyang Lu, Qianying Zheng, Shuying Cheng, Yaohai Lin, Zhicong Chen, Lijun Wu, Zhuang Qian
Summary: This paper proposes an open-set fault diagnosis model for PV arrays based on 1D VoVNet-SVDD. The model accurately diagnoses various types of faults and is capable of identifying unknown fault types.