Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Training Time”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,225 records · Page 68
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

A Quantum-Classical Collaborative Training Architecture Based on Quantum State Fidelity

Recent advancements have highlighted the limitations of current quantum systems, particularly the restricted number of qubits available on near-term quantum devices. This constraint greatly inhibits the range of applications that can leverage quantum computers. Moreover, as the available qubits increase, the computational complexity grows exponentially, posing additional challenges. Consequently, there is an urgent need to use qubits efficiently and mitigate both present limitations and future complexities. To address this, existing quantum applications attempt to integrate classical and quantum systems in a hybrid framework. In this study, we concentrate on quantum deep learning and introduce a collaborative classical-quantum architecture called co-TenQu. The classical component employs a tensor network for compression and feature extraction, enabling higher-dimensional data to be encoded onto logical quantum circuits with limited qubits. On the quantum side, we propose a quantum-state-fidelity-based evaluation function to iteratively train the network through a feedback loop between the two sides. co-TenQu has been implemented and evaluated with both simulators and the IBM-Q platform. Compared to state-of-the-art approaches, co-TenQu enhances a classical deep neural network by up to 41.72% in a fair setting. Additionally, it outperforms other quantum-based methods by up to 1.9 times and achieves similar accuracy while utilizing 70.59% fewer qubits.

42 ENGINEERING↗
DOE OSTI2024

Hamiltonian learning using machine-learning models trained with continuous measurements

Here, we build upon recent work on the use of machine-learning models to estimate Hamiltonian parameters using continuous weak measurement of qubits as input. We consider two settings for the training of our model: (1) supervised learning, where the weak-measurement training record can be labeled with known Hamiltonian parameters, and (2) unsupervised learning, where no labels are available. The first has the advantage of not requiring an explicit representation of the quantum state, thus potentially scaling very favorably to a larger number of qubits. The second requires the implementation of a physical model to map the Hamiltonian parameters to a measurement record, which we implement using an integrator of the physical model with a recurrent neural network to provide a model-free correction at every time step to account for small effects not captured by the physical model. We test our construction on a system of two qubits and demonstrate accurate prediction of multiple physical parameters in both the supervised context and the unsupervised context. We demonstrate that the model benefits from larger training sets, establishing that it is “learning,” and we show robustness regarding errors in the assumed physical model by achieving accurate parameter estimation in the presence of unanticipated single-particle relaxation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Effects of random forest modeling decisions on biogeochemical time series predictions

Abstract Random forests (RF) are an increasingly popular machine learning approach used to model biogeochemical processes in the Earth system. While RF models are robust to many assumptions that complicate deterministic models, there are several important parameterization decisions for appropriate use and optimal model fit. We explored the role that parameter decisions, including training/testing data splitting strategies, variable selection, and hyperparameters play on RF goodness‐of‐fit by constructing models using 1296 unique parameter combinations to predict concentrations of nitrate, a key nutrient for biogeochemical cycling in aquatic ecosystems. Models were built on long‐term, publicly available water quality and meteorology time series collected by the National Estuarine Research Reserve monitoring network for two contrasting ecosystems representing freshwater and brackish estuaries. We found that accounting for temporal dependence when splitting data into training and testing subsets was key for avoiding over‐estimation of model predictive power. In addition, variable selection, the ratio of training to testing data, and to a lesser degree, variables per split and number of trees, were significant parameters for optimizing RF goodness‐of‐fit. We also explored how model parameter decisions influenced interpretation of the relative importance of predictors to the model, and model predictor‐dependent variable relationships, with results suggesting that both data structure and model parameterization influence these factors. Because much of the current RF literature is written for the computational and statistical science communities, the primary goal of this study is to provide guidelines for aquatic scientists new to machine learning to apply RF techniques appropriately to aquatic biogeochemical datasets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Data-driven selection of stiff chemistry ODE solver in operator-splitting schemes

Most computational fluid dynamics simulations of practical combustion applications employ operator-splitting schemes, where chemistry and transport are separated and integrated with distinct numerical methods. The changes in composition due to chemistry are evaluated by solving ordinary differential equations (ODE) in each cell of the computational domain, which typically dominates the computational cost when detailed chemistry is considered. In this work, a data-driven approach for the selection of chemistry ODE solvers in operator-splitting schemes is presented. Neural networks are used to predict the ODE solvers CPU times and errors for a given thermochemical state. This allows the selection of an optimal ODE solver on a cell-by-cell, timestep-by-timestep basis. The models are trained using a wide set of thermochemical states generated through partially-stirred reactors and flames simulations. The methodology is validated by quantifying the prediction errors, the classification accuracy, and the computational speedup. The model predicts the optimal ODE solver for 70 to 95% of the validation cases and decreases the computional cost by a factor of 3 or more. The generalizability of the methodology to different chemical mechanisms and different fuels is assessed and it is shown that the model’s performance is only slightly degraded and its applicability is significantly enhanced if the inputs to the neural networks are restricted to a small set of thermochemical state variables present in most chemical mechanisms. In conclusion, the models are used in an homogeneous reactor case and a multi-dimensional CFD simulation of a diesel spray at high pressure where a speedup of more than 3 is achieved.

42 ENGINEERING↗
DOE OSTI2023

Angularly resolved photoionization dynamics in atoms and molecules combining temporally and spectrally resolved experiments at ATTOLab and Synchrotron SOLEIL

We report results for XUV-IR two-photon ionization of Ar, Ne, NO, and O2, where an XUV attosecond pulse train is superimposed with a synchronized IR pulse, obtained at the ATTOLab laser facility using electron–ion coincidence 3D momentum spectroscopy. Temporally resolved photoelectron angular distributions providing angle-resolved time-delays for np ionization of Ar and Ne, achieved by reconstruction of attosecond beating by interference of two-photon transitions through a unified formalism (Joseph et al. in J Phys B At Mol Opt Phys 53:184007, 2020), are summarized. For inner valence XUV-IR dissociative photoionization of NO and O2 molecules, we report electron–ion kinetic energy correlation diagrams and disentangle the dissociative photoionization processes relying on parallel XUV experiments at Synchrotron SOLEIL. For ionization into the NO+(c3Π) ionic state, extending the formalism developed for single-photon ionization, we focus on photoelectron angular distributions averaged on the delay between the XUV and the IR field in the field frame, molecular frame, and electron frame of reference.

Joseph, J↗
DOE OSTI2022

Continual Learning for Pattern Recognizers using Neurogenesis Deep Learning

Deep neural networks have emerged as a leading set of algorithms to infer information from a variety of data sources such as images and time series data. In their most basic form, neural networks lack the ability to adapt to new classes of information. Continual learning is a field of study attempting to give previously trained deep learning models the ability to adapt to a changing environment. Previous work developed a CL method called Neurogenesis for Deep Learning (NDL). Here, we combine NDL with a specific neural network architecture (the Ladder Network) to produce a system capable of automatically adapting a classification neural network to new classes of data. The NDL Ladder Network was evaluated against other leading CL methods. While the NDL and Ladder Network system did not match the cutting edge performance achieved by other CL methods, in most cases it performed comparably and is the only system evaluated that can learn new classes of information with no human intervention.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Advanced Data Science Model for Detecting Intelligent Malware

This study focused on developing a robust artificial intelligence (AI) model capable of detecting and characterizing advanced malware in Internet of Things (IoT) devices using network data. By analyzing network traffic with various machine learning (ML) models, our AI model can identify and characterize malicious activities to significantly improve malware detection accuracy and reliability as compared to traditional methods. The developed AI/ML model was trained using network data from IoT devices, leveraging classifiers such as Random Forest, Gradient Boosting, AdaBoost, and others to optimize detection performance. This project demonstrates a scalable framework for real-time malware detection and characterization in IoT networks, capable of identifying infected devices and facilitating the necessary steps to remove or isolate them, thereby preventing further infections. Although digital twin (DT) integration is not yet implemented in the current model, it represents a promising future enhancement. By creating a virtual replica of physical IoT devices, DT technology would allow for real-time monitoring and analysis without directly accessing operational technology, thus reducing the risk of compromising or reducing the performance of actual devices. This integration would further enhance the security of IoT ecosystems, combining AI technology to better flag and detect indications of malware-infected devices within a nuclear system environment.

42 ENGINEERING↗
DOE OSTI2023

An extended focused assessment with sonography in trauma ultrasound tissue-mimicking phantom for developing automated diagnostic technologies

Medical imaging-based triage is critical for ensuring medical treatment is timely and prioritized. However, without proper image collection and interpretation, triage decisions can be hard to make. While automation approaches can enhance these triage applications, tissue phantoms must be developed to train and mature these novel technologies. Here, we have developed a tissue phantom modeling the ultrasound views imaged during the enhanced focused assessment with sonography in trauma exam (eFAST). The tissue phantom utilized synthetic clear ballistic gel with carveouts in the abdomen and rib cage corresponding to the various eFAST scan points. Various approaches were taken to simulate proper physiology without injuries present or to mimic pneumothorax, hemothorax, or abdominal hemorrhage at multiple locations in the torso. Multiple ultrasound imaging systems were used to acquire ultrasound scans with or without injury present and were used to train deep learning image classification predictive models. Performance of the artificial intelligent (AI) models trained in this study achieved over 97% accuracy for each eFAST scan site. We used a previously trained AI model for pneumothorax which achieved 74% accuracy in blind predictions for images collected with the novel eFAST tissue phantom. Grad-CAM heat map overlays for the predictions identified that the AI models were tracking the area of interest for each scan point in the tissue phantom. Overall, the eFAST tissue phantom ultrasound scans resembled human images and were successful in training AI models. Tissue phantoms are critical first steps in troubleshooting and developing medical imaging automation technologies for this application that can accelerate the widespread use of ultrasound imaging for emergency triage.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

KGML-ag: a modeling framework of knowledge-guided machine learning to simulate agroecosystems: a case study of estimating N<sub>2</sub>O emission using data from mesocosm experiments

Abstract. Agricultural nitrous oxide (N2O) emission accounts for a non-trivial fraction of global greenhouse gas (GHG) budget. To date, estimating N2O fluxes from cropland remains a challenging task because the related microbial processes (e.g., nitrification and denitrification) are controlled by complex interactions among climate, soil, plant and human activities. Existing approaches such as process-based (PB) models have well-known limitations due to insufficient representations of the processes or uncertainties of model parameters, and due to leverage recent advances in machine learning (ML) a new method is needed to unlock the “black box” to overcome its limitations such as low interpretability, out-of-sample failure and massive data demand. In this study, we developed a first-of-its-kind knowledge-guided machine learning model for agroecosystems (KGML-ag) by incorporating biogeophysical and chemical domain knowledge from an advanced PB model, ecosys, and tested it by comparing simulating daily N2O fluxes with real observed data from mesocosm experiments. The gated recurrent unit (GRU) was used as the basis to build the model structure. To optimize the model performance, we have investigated a range of ideas, including (1) using initial values of intermediate variables (IMVs) instead of time series as model input to reduce data demand; (2) building hierarchical structures to explicitly estimate IMVs for further N2O prediction; (3) using multi-task learning to balance the simultaneous training on multiple variables; and (4) pre-training with millions of synthetic data generated from ecosys and fine-tuning with mesocosm observations. Six other pure ML models were developed using the same mesocosm data to serve as the benchmark for the KGML-ag model. Results show that KGML-ag did an excellent job in reproducing the mesocosm N2O fluxes (overall r2=0.81, and RMSE=3.6 mgNm-2d-1 from cross validation). Importantly, KGML-ag always outperforms the PB model and ML models in predicting N2O fluxes, especially for complex temporal dynamics and emission peaks. Besides, KGML-ag goes beyond the pure ML models by providing more interpretable predictions as well as pinpointing desired new knowledge and data to further empower the current KGML-ag. We believe the KGML-ag development in this study will stimulate a new body of research on interpretable ML for biogeochemistry and other related geoscience processes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

BETO 2021 Peer Review - Electrocatalytic CO2 Utilization

The goal of the ChemCatBio DataHub project is to accelerate the catalyst and process development cycle by developing transformational tools for prediction and collaboration in catalyst R&D. The project is currently focused on the development of the Catalyst Property Database (CPD), a free and public resource released in September 2020. The CPD was designed to advance the state of the art for application of computational data. When computational data, such as computed reaction energetics, is used in catalyst design, it is almost always generated by the researchers seeking to use it, even if similar data has been published previously. One barrier to data reuse that results in this duplication of effort is the difficult process of finding and applying published data, which can be slow, error-prone, and manual. The CPD seeks to overcome these challenges by creating a centralized, searchable database of quality catalyst property data. At present, the CPD contains computed adsorption energies for intermediates along catalytic pathways. During FY21 and FY22, development of the CPD continues with a focus on external users and meeting their requirements. A batch upload capability, training and curation procedures, user interviews, and a demonstration of the CPD's utility in accelerating catalyst research are planned. Overall, the Data Hub project and CPD aim to reduce the time and cost of catalyst research by harnessing the power of data in catalyst discovery.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2021

BETO 2021 Peer Review - ChemCatBio Data Hub 2.6.2.500

The goal of the ChemCatBio DataHub project is to accelerate the catalyst and process development cycle by developing transformational tools for prediction and collaboration in catalyst R&D. The project is currently focused on the development of the Catalyst Property Database (CPD), a free and public resource released in September 2020. The CPD was designed in response to the observation that when data, such as computed reaction energetics, is used in catalyst design, it is almost always generated by the researchers seeking to use it, even if similar data has been published previously. One barrier to data reuse that results in this duplication of effort is the difficult process of finding and applying published data, which can be slow, error-prone, and manual. The CPD seeks to overcome these challenges by creating a centralized, searchable database of quality catalyst property data. At present, the CPD contains computed adsorption energies for intermediates along catalytic pathways. During FY21 and FY22, development of the CPD continues with a focus on external users and meeting their requirements. A batch upload capability, training and curation procedures, user interviews, and a demonstration of the CPD's utility in accelerating catalyst research are planned. Overall, the Data Hub project and CPD aim to reduce the time and cost of catalyst research by harnessing the power of data in catalyst discovery.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2026

Enhanced accuracy through ensembling of randomly initialized auto-regressive models for dynamical systems

Computational mechanics simulations using traditional finite element methods (FEM) require prohibitively expensive computational resources for real-time engineering applications, design optimization, and digital twin implementations. While machine learning (ML) surrogate models offer significant computational speedups, autoregressive ML models for time-dependent mechanical systems suffer from error accumulation that compromises long-term prediction reliability - a critical concern for engineering applications where accuracy over extended time horizons is essential for safety and performance assessments. Here, we propose a deep ensemble framework specifically designed to address this challenge in computational mechanics applications, where multiple ML surrogate models with random weight initializations are trained in parallel and their predictions aggregated during inference. This approach leverages statistical diversity to maximize information gain from a fixed set of training data and to mitigate error propagation, while maintaining the computational efficiency that makes ML surrogates attractive for engineering practice. We validate the framework on three representative problems spanning critical areas of computational mechanics: stress field evolution in heterogeneous microstructures under complex loading (relevant to advanced materials design and composite analysis), planetary-scale shallow water dynamics (applicable to environmental and geotechnical engineering), and Gray-Scott reaction-diffusion systems (relevant to mass transport and chemical process engineering). Across all test cases, the ensemble approach demonstrates consistent error reduction of 15-33% compared to individual models. The codes for this work are available on GitHub (https://github.com/Graham-Brady-Research-Group/AutoregressiveEnsemble_SpatioTemporal_Evolution).

autoregressive prediction↗
DOE OSTI2020

FPDeep: Scalable Acceleration of CNN Training on Deeply-Pipelined FPGA Clusters

In this paper, we propose a framework called FPDeep, which uses a hybrid of model and layer paral- lelism to configure distributed reconfigurable clusters to train DNNs. This approach has numerous benefits. First, the design does not suffer from batch size growth. Second, novel workload and weight partitioning leads to balanced loads of both among nodes. And third, the entire system is fine-grained pipeline. This leads to high parallelism and utilization and also minimizes the time features need to be cached while waiting for back-propagation.

Wang, Tianqi↗
DOE OSTI2026

Microsecond-latency feedback at a particle accelerator by online reinforcement learning on hardware

The commissioning and operation of future large-scale scientific experiments will challenge current tuning and control methods. Reinforcement learning (RL) algorithms are a promising solution due to their ability to dynamically adapt to changing environments and consider delayed consequences. In many real-world applications, RL policies must produce actions in real time, often within microseconds to milliseconds, imposing significant constraints on system latency and computational overhead that conventional machine learning libraries are not designed to handle. To control phenomena in real time at these timescales, RL needs to be deployed on-the-edge, namely on dedicated hardware located near the system it controls, without relying on a host CPU or cloud-based inference. In this work we present the design and deployment of an experience accumulator system in a particle accelerator. In this system, deep-RL algorithms run using hardware acceleration and act within a few microseconds, enabling the use of RL for control of phenomena like beam instabilities. The training uses the collected data offline to reduce the number of operations carried out on the acceleration hardware. The proposed architecture was tested in real experimental conditions at the Karlsruhe research accelerator, a synchrotron light source, where the system was used to control artificially induced horizontal betatron oscillations in real-time, with a control loop period of just 2.7 μs. The results showed a performance comparable to the commercial feedback system available at the accelerator, demonstrating the viability and potential of this approach. Due to the self-learning and reconfiguration capability of this implementation, a seamless application to other control problems is possible. Applications range from particle accelerators to large-scale research and industrial facilities.

FPGA↗
DOE OSTI2020

Photometric Redshifts with the LSST. II. The Impact of Near-infrared and Near-ultraviolet Photometry

Accurate photometric redshift (photo-z) estimates are essential to the cosmological science goals of the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST). In this work, we use simulated photometry for mock galaxy catalogs to explore how LSST photo-z estimates can be improved by the addition of near-infrared (NIR) and/or ultraviolet (UV) photometry from the Euclid, Wide-Field InfrarRed Survey Telescope (WFIRST), and/or Cosmological Advanced Survey Telescope for Optical and ultraviolet Research (CASTOR) space telescopes. Generally, we find that deeper optical photometry can reduce the standard deviation of the photo-z estimates more than adding NIR or UV filters but that additional filters are the only way to significantly lower the fraction of galaxies with catastrophically under- or overestimated photo-z. For Euclid, we find that the addition of JH 5σ photometric detections can reduce the standard deviation for galaxies with z > 1 (z > 0.3) by ∼20% (∼10%), and the fraction of outliers by ∼40% (∼25%). For WFIRST, we show how the addition of deep YJHK photometry could reduce the standard deviation by ≳50% at z > 1.5 and drastically reduce the fraction of outliers to just ∼2% overall. For CASTOR, we find that the addition of its UV- and u-band photometry could reduce the standard deviation by ∼30% and the fraction of outliers by ∼50% for galaxies with z < 0.5. We also evaluate the photo-z results within sky areas that overlap with both the NIR and UV surveys and when spectroscopic training sets built from the surveys’ small-area deep fields are used.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48% ($H_T$) and 28% (AD), with a cumulative gain of up to 2% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56% ($H_T$) and 28% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO_LHC (see repo for details).

Ding, Zixin [Chicago U.]↗
DOE OSTI2026

Monitoring of Liquid Metal Reactor Heater Zones with Recurrent Neural Network Learning of Temperature Time Series

Advanced high-temperature fluid reactors (ARs), such as sodium fast reactors (SFRs) and molten salt cooled reactors (MSCRs) utilize high-temperature fluids at ambient pressure. To melt the fluid during reactor startup and prevent fluid freezing during cooldown, the thermal–hydraulic systems of such ARs include heater zones consisting of specific heaters with controllers, temperature sensors, and thermal insulation. The failure of heater zones due to insulation material degradation or improper installation, resulting in parasitic heat losses, can lead to fluid freezing. The detection of faults using a heat-transfer model is difficult because of a lack of knowledge of the experimental details. Data-driven machine learning of heater zone temperature time series offers a viable alternative. In this study, we benchmarked the performance of recurrent neural networks (RNNs) in an analysis of heat-up transient temperature time series of heater zones installed on a liquid sodium vessel. The RNN models include long short-term memory (LSTM) and gated recurrent unit (GRU) networks, as well as their bi-directional variants, BiLSTM and BiGRU. Anomalous temperature points were designated using a percentile-based threshold applied to residual fluctuations in the detrended temperature time series. Additionally, the impact of the exponentially weighted moving average (EWMA) method on detection accuracy was examined. The RNN models’ performance was assessed using precision, recall, and F 1 score metrics. Results demonstrated that RNN models effectively detect anomalies in temperature time series with the best models for each heater zone achieving F 1 scores of over 93%. To explain the variations in RNN model performance across different heater zones, we used Kullback–Leibler (KL) divergence to quantify the relative entropy between training and testing data, and the Detrended Fluctuation Analysis (DFA) to assess long-range temporal correlations. For datasets with strong long-range correlations and minimal relative entropy between training and testing data, GRU is the best-performing model. When the data exhibits weaker long-term correlations and a significant relative entropy between training and testing distributions, BiGRU shows the best performance. For the data sets with intermediate values of both KL divergence and DFA, the best performance is obtained with LSTM and BiLSTM, respectively.

gated recurrent unit↗
DOE OSTI2023

Privacy-preserving federated learning: Application to behind-the-meter solar photovoltaic generation forecasting

Here, the growing usage of decentralized renewable energy sources has made accurate estimation of their aggregated generation crucial for maintaining grid flexibility and reliability. However, the majority of distributed photovoltaic (PV) systems are behind-the-meter (BTM) and invisible to utilities, leading to three challenges in obtaining an accurate forecast of their aggregated output. Firstly, traditional centralized prediction algorithms used in previous studies may not be appropriate due to privacy concerns. There is therefore a need for decentralized forecasting methods, such as federated learning (FL), to protect privacy. Secondly, there has been no comparison between localized, centralized, and decentralized forecasting methods for BTM PV production, and the trade-off between prediction accuracy and privacy has not been explored. Lastly, the computational time of data-driven prediction algorithms has not been examined. This article presents a FL power forecasting method for PVs, which uses federated learning as a decentralized collaborative modeling approach to train a single model on data from multiple BTM sites. The machine learning network used to design this FL-based BTM PV forecasting model is a multi-layered perceptron, which ensures privacy and security of the data. Comparing the suggested FL forecasting model to non-private centralized and entirely private localized models revealed that it has a high level of accuracy, with an RMSE that is 18.17% lower than localized models and 9.9% higher than centralized models.

14 SOLAR ENERGY↗