Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ensemble optimization”

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 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

STOCHASTIC OPTIMAL POWER FLOW FOR REAL-TIME MANAGEMENT OF DISTRIBUTED RENEWABLE GENERATION AND DEMAND RESPONSE (Final Report)

To meet the grand challenge of a sustainable energy future, there has been a surge of interest in renewable energy. Today, the uncertainty associated with renewable resources is handled by using operating reserves. The high penetration of renewable resources, however, introduces difficult-to-control dynamics and challenges for power system operation. Decision support tools are necessary at the bulk system operational level to recognize and efficiently utilize renewable resources and distributed demand response products in concert with traditional grid resources. It is envisaged that responsive load can potentially have very significant cost advantages over either spinning or non-spinning ramping reserve. Critical decisions are made during hour(s)-ahead and real-time power system operation regarding the commitment and dispatch of generators to ensure power delivery is both reliable and economic. These decisions are typically made by a security constrained optimal flow, which determines future generator commitments, dispatches, and ensures adequate reserves are available in the event of a contingency (unexpected outage) or if future system conditions deviate from forecasts. However, security has been always based on a pre-specified subset of contingency constraints whose enforcement does not guarantee security under all possible future possibilities while also giving little or no weight to the likelihood of each contingent event or the severity of its consequences. Existing tools, which are based exclusively on deterministic optimization models, do not yield optimal operational decisions to address these new challenges, in terms of both reliability and cost-effectiveness. This project has focused on developing a stochastic optimal power flow (SOPF) framework, which integrates renewable resource uncertainty, load uncertainty, distributed storage (DS), demand response (DR) products, in a holistic manner to address the uncertainty associated with ever-increasing renewable resources, along with the inclusion of distributed demand response products in future power systems. A proof-of-concept problem was created using the Pennsylvania-Jersey-Maryland (PJM) power system network. Synthetic wind generation was added to the system to simulate 50% wind penetration. A 1-hour test of SOPF operation indicated more than 6% operational cost savings. The project continued by adding the Midwestern Independent System Operator (MISO) as a partner, with focus shifting from SOPF to Stochastic Look-Ahead Unit Commitment (SLAC). Unlike PJM, MISO is faced with significant renewable energy resources within its footprint and is challenged with substantial uncertainty in its operations. The SLAC distinguishes itself from existing tools that operators use. At best, today’s tools solve two to three cases independently, where one or two system parameters, such as forecasted load level (e.g., a low, base, and high forecast), are varied and the resulting scenarios are analyzed independently. The stochastic-based optimization of SLAC leverages statistical information from an ensemble of potential operational scenarios and their respective likelihood. The SLAC output can be translated into valuable information to the operator such as suggested commitments, optimal scheduling and dispatch of resources, reserve requirements at both locational and zonal resolutions, ramping availability and requirements, availability of demand response including operational guidance concerning the near-term and real-time coordination between distributed energy resources, and utilization of distributed storage resources. The developed SOPF/SLAC tool, a stand-alone tool compatible with existing EMSs, will provide system operators with unprecedented visibility, flexibility and predictability to these resources and operational guidance concerning the real-time coordination between DERs and DR/DS products. The game changing and practical impact of this disruptive technology will be dramatic and will usher in a new era in the electric power industry, wherein green energy concepts are fully embraced, and electric power costs are lowered throughout the nation.

42 ENGINEERING↗
NASA NTRS1994

In Vitro Selection of Optimal DNA Substrates for Ligation by a Water-Soluble Carbodiimide

We have used in vitro selection to investigate the sequence requirements for efficient template-directed ligation of oligonucleotides at 0 deg C using a water-soluble carbodiimide as condensing agent. We find that only 2 bp at each side of the ligation junction are needed. We also studied chemical ligation of substrate ensembles that we have previously selected as optimal by RNA ligase or by DNA ligase. As anticipated, we find that substrates selected with DNA ligase ligate efficiently with a chemical ligating agent, and vice versa. Substrates selected using RNA ligase are not ligated by the chemical condensing agent and vice versa. The implications of these results for prebiotic chemistry are discussed.

Harada, Kazuo↗
NASA NTRS2016

Application of Aviation Turbulence Information to Air-Traffic Management (ATM): Chapter - 24

Unexpected turbulence especially in the upper troposphere and lower stratosphere where cabin crews and passengers in cruising aircraft are likely to unbuckle causes in-flight injuries, structural damage, and flight delay. Therefore, turbulence information can be used to improve safety while pursuing efficiency in Air-Traffic Management (ATM). In this chapter, simple modeling of aircraft trajectories combined with wind and turbulence predictions can suggest the optimal solution of flight plans that minimizes both total flight time (e.g., fuel consumption) and potential encounters of turbulence from departure to arrival airports. Also, probabilistic ensemble turbulence forecasts are applied to suggest an optimal strategic and tactical ATM route planning in a given weather and turbulence condition in the United States which are evaluated against in situ Eddy Dissipation Rate observations from commercial aircraft. Finally, variations of long-haul trans-Oceanic flight routes and their turbulence potentials are investigated using a global reanalysis data to understand how the upper-level large-scale flow patterns can affect the long-term ATM planning through the changes of winds and turbulence conditions.

Wind Optimal↗
DOE OSTI2023

A machine learning approach targeting parameter estimation for plant functional type coexistence modeling using ELM-FATES (v2.0)

Tropical forest dynamics play a crucial role in the global carbon, water, and energy cycles. However, realistically simulating the dynamics of competition and coexistence between different plant functional types (PFTs) in tropical forests remains a significant challenge. This study aims to improve the modeling of PFT coexistence in the Functionally Assembled Terrestrial Ecosystem Simulator (FATES), a vegetation demography model implemented in the Energy Exascale Earth System Model (E3SM) land model (ELM), ELM-FATES. Specifically, we explore (1) whether plant trait relationships established from field measurements can constrain ELM-FATES simulations and (2) whether machine learning (ML)-based surrogate models can emulate the complex ELM-FATES model and optimize parameter selections to improve PFT coexistence modeling. We conducted three ensembles of ELM-FATES experiments at a tropical forest site near Manaus, Brazil. By comparing the ensemble experiments without (Exp-CTR) and with (Exp-OBS) consideration of observed trait relationships, we found that accounting for these relationships slightly improves the simulations of water, energy, and carbon variables when compared to observations but degrades the simulation of PFT coexistence. Using ML-based surrogate models trained on Exp-CTR, we optimized the trait parameters in ELM-FATES and conducted another ensemble of experiments (Exp-ML) with these optimized parameters. The proportion of PFT coexistence experiments significantly increased from 21 % in Exp-CTR to 73 % in Exp-ML. After filtering the experiments that allow for PFT coexistence to agree with observations (within 15 % tolerance), 33 % of the Exp-ML experiments were retained, which is a significant improvement compared to the 1.4 % in Exp-CTR. Exp-ML also accurately reproduces the annual means and seasonal variations in water, energy, and carbon fluxes and the field inventory of aboveground biomass. This study represents a reproducible method that utilizes machine learning to identify parameter values that improve model fidelity against observations and PFT coexistence in vegetation demography models for diverse ecosystems. Our study also suggests the need for new mechanisms to enhance the robust simulation of coexisting plants in ELM-FATES and has significant implications for modeling the response and feedbacks of ecosystem dynamics to climate change.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRSDate not supplied

DeepONet-Assisted Optimization of Surface Topography for Transition Delay in a Mach 4.5 Boundary Layer

We use deep learning, an ensemble variational technique (EnVar), and direct numerical simulations(DNS) to design an optimal topography for a two-dimensional roughness element that delays the on-set of laminar-turbulent transition in a Mach 4.5 flat-plate boundary layer. Deep operator networks (DeepONets), which have the known ability to learn complex nonlinear operators within dynamical systems, are used for machine learning. For the baseline configuration of a smooth flat plate, the second-mode waves at the DNS inflow cause a quick nonlinear breakdown of the high-speed boundary layer within the computational domain. Results reported in the present study validate the ability of DeepONets to model the transition delay via a given topography of the roughness element. The computing cost to optimize the rough-ness element for minimal skin-friction drag is substantially lowered by the DeepONets-based reduced-order model. In comparison to the baseline method of EnVar optimization based on DNS alone, the DeepONets-based EnVar optimizer is able to delay transition past the outflow boundary of the computational domain while utilizing almost 5–6 times fewer DNS.

Machine Learning↗
DOE OSTI2025

BEAST

The Bilinear Ensemble Actuation Synthesis Toolkit (BEAST) is a computational platform that optimizes time-varying control signals to achieve a specified transfer of states governed by bilinear ensemble systems in which model parameters are subject to uncertainty.

Zlotnik, anatoly↗
DOE OSTI2020

GMRES with embedded ensemble propagation for the efficient solution of parametric linear systems in uncertainty quantification of computational models

In a previous work, embedded ensemble propagation was proposed to improve the efficiency of sampling-based uncertainty quantification methods of computational models on emerging computational architectures. It consists of simultaneously evaluating the model for a subset of samples together, instead of evaluating them individually. A first approach introduced to solve parametric linear systems with ensemble propagation is ensemble reduction. In Krylov methods for example, this reduction consists in coupling the samples together using an inner product that sums the sample contributions. Ensemble reduction has the advantages of being able to use optimized implementations of BLAS functions and having a stopping criterion which involves only one scalar. However, the reduction potentially decreases the rate of convergence due to the gathering of the spectra of the samples. In this paper, we investigate a second approach: ensemble propagation without ensemble reduction in the case of GMRES. This second approach solves each sample simultaneously but independently to improve the convergence compared to ensemble reduction. This raises two new issues which are solved in this paper: the fact that optimized implementations of BLAS functions cannot be used anymore and that ensemble divergence, whereby individual samples within an ensemble must follow different code execution paths, can occur. We tackle those issues by implementing a high-performing ensemble GEMV and by using masks. The proposed ensemble GEMV leads to a similar cost per GMRES iteration for both approaches, i.e. with and without reduction. For illustration, we study the performances of the new linear solver in the context of a mesh tying problem. Furthermore, this example demonstrates improved ensemble propagation speed-up without reduction.

BLAS↗
NASA NTRS2020

An Examination of the Impact of Grid Spacing on WRF Simulations of Wintertime Precipitation in the Mid-Atlantic United States

A large set of deterministic and ensemble forecasts was produced to identify the optimal spacing for forecasting U.S. East Coast snowstorms. WRF forecasts were produced on cloud-allowing (~1-km grid spacing) and convection-allowing (3–4 km) grids, and compared against forecasts with parameterized convection (>~10 km). Performance diagrams were used to evaluate 19 deterministic forecasts from the winter of 2013–14. Ensemble forecasts of five disruptive snowstorms spanning the years 2015–18 were evaluated using various methods to evaluate probabilistic forecasts. While deterministic forecasts using cloud-allowing grids were not better than convection-allowing forecasts, both had lower bias and higher success ratios than forecasts with parameterized convection. All forecasts were underdispersive. Nevertheless, forecasts on the higher-resolution grids were more reliable than those with parameterized convection. Forecasts on the cloud-allowing grid were best able to discriminate areas that received heavy snow and those that did not, while the forecasts with parameterized convection were least able to do so. It is recommended to use convection-resolving and (if computationally possible) to use cloud-allowing forecast grids when predicting East Coast winter storms.

Snowstorms↗
DOE OSTI2024

Optimal Twirling Depth for Classical Shadows in the Presence of Noise

The classical shadows protocol is an efficient strategy for estimating properties of an unknown state p using a small number of state copies and measurements. In its original form, it involves twirling the state with unitaries from some ensemble and measuring the twirled state in a fixed basis. It was recently shown that for computing local properties, optimal sample complexity (copies of the state required) is remarkably achieved for unitaries drawn from shallow depth circuits composed of local entangling gates, as opposed to purely local (zero depth) or global twirling (infinite depth) ensembles. Here, we consider the sample complexity as a function of the depth of the circuit, in the presence of noise. We find that this noise has important implications for determining the optimal twirling ensemble. Under fairly general conditions, we (i) show that any single-site noise can be accounted for using a depolarizing noise channel with an appropriate damping parameter f, (ii) compute thresholds f th at which optimal twirling reduces to local twirling for Pauli operators, (iii) nth order Renyi entropies (n ≥2), and (iv) provide a meaningful upper bound t max on the optimal circuit depth for any finite noise strength f, which applies to observables and entanglement entropy measurements. In conclusion, these thresholds strongly constrain the search for optimal strategies to implement shadow tomography and are easily tailored to the experimental system at hand.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

A Framework for Evaluating Climate Model Performance Metrics

The CMIP5 archive contains future climate projections from over 50 models provided by dozens of modeling centers from around the world. Individual model projections, however, are subject to biases created by structural model uncertainties. As a result, ensemble averaging of multiple models is often used to add value to model projections: consensus projections have been shown to consistently outperform individual models. Previous reports for the IPCC establish climate change projections based on an equal-weighted average of all model projections. However, certain models reproduce climate processes better than other models. Should models be weighted based on performance? Unequal ensemble averages have previously been constructed using a variety of mean state metrics. What metrics are most relevant for constraining future climate projections? This project develops a framework for systematically testing metrics in models to identify optimal metrics for unequal weighting multi-model ensembles. A unique aspect of this project is the construction and testing of climate process-based model evaluation metrics. A climate process-based metric is defined as a metric based on the relationship between two physically related climate variables?e.g., outgoing longwave radiation and surface temperature. Metrics are constructed using high-quality Earth radiation budget data from NASA's Clouds and Earth's Radiant Energy System (CERES) instrument and surface temperature data sets. It is found that regional values of tested quantities can vary significantly when comparing weighted and unweighted model ensembles. For example, one tested metric weights the ensemble by how well models reproduce the time-series probability distribution of the cloud forcing component of reflected shortwave radiation. The weighted ensemble for this metric indicates lower simulated precipitation (up to .7 mm/day) in tropical regions than the unweighted ensemble: since CMIP5 models have been shown to overproduce precipitation, this result could indicate that the metric is effective in identifying models which simulate more realistic precipitation. Ultimately, the goal of the framework is to identify performance metrics for advising better methods for ensemble averaging models and create better climate predictions.

Noel C Baker↗
DOE OSTI2025

Minimizing CGYRO HPC Communication Costs in Ensembles with XGYRO by Sharing the Collisional Constant Tensor Structure

First-principles fusion plasma simulations are both compute and memory intensive, and CGYRO is no exception. The use of many HPC nodes to fit the problem in the available memory thus results in significant communication overhead, which is hard to avoid for any single simulation. That said, most fusion studies are composed of ensembles of simulations, so we developed a new tool, named XGYRO, that executes a whole ensemble of CGYRO simulations as a single HPC job. By treating the ensemble as a unit, XGYRO can alter the global buffer distribution logic and apply optimizations that are not feasible on any single simulation, but only on the ensemble as a whole. The main saving comes from the sharing of the collisional constant tensor structure, since its values are typically identical between parameter-sweep simulations. This data structure dominates the memory consumption of CGYRO simulations, so distributing it among the whole ensemble results in drastic memory savings for each simulation, which in turn results in overall lower communication overhead.

CGYRO↗
DOE OSTI2021

Ensemble Learning Based Convex Approximation of Three-Phase Power Flow

Though the convex optimization has been widely used in power systems, it still cannot guarantee to yield a tight (accurate) solution to some problems. To mitigate this issue, this paper proposes an ensemble learning based convex approximation for alternating current (AC) power flow equations that differs from the existing convex relaxations. The proposed approach is based on three-phase quadratic power flow equations in rectangular coordinates. To develop this data-driven convex approximation of power flows, the polynomial regression (PR) is first deployed as a basic learner to fit convex relationships between the independent and dependent variables. Then, ensemble learning algorithms such as gradient boosting (GB) and bagging are introduced to combine learners to boost model performance. Based on the learned convex approximation of power flow, optimal power flow (OPF) is formulated as a convex quadratic programming problem. The simulation results on IEEE standard cases of both balanced and unbalanced systems show that, in the context of solving OPF, the proposed data-driven convex approximation outperforms the conventional semi-definite programming (SDP) relaxation in both accuracy and computational efficiency, especially in the cases that the conventional SDP relaxation fails

Convex approximation↗
DOE OSTI2022

Application-specific optimal model weighting of global climate models: A red tide example

Global climate models (GCMs) and Earth system models (ESMs) provide many climate services with environmental relevance. The High Resolution Model Inter-comparison Project (HighResMIP) of the Coupled Model Intercomparison Project Phase 6 (CMIP6) provides model runs of GCMs and ESMs to address regional phenomena. Developing a parsimonious ensemble of CMIP6 requires multiple ensemble methods such as independent-model subset selection, prescreening-based subset selection, and model weighting. The work presented here focuses on application-specific optimal model weighting, with prescreening-based subset selection. As such, independent ensemble members are categorized, selected, and weighted based on their ability to reproduce physically-interpretable features of interest that are problem-specific. We discuss the strengths and caveats of optimal model weighting using a case study of red tide prediction in the Gulf of Mexico along the West Florida Shelf. Red tide is a common name of specific harmful algal blooms that occur worldwide, causing adverse socioeconomic and environmental impacts. Our results indicate the importance of prescreening-based subset selection as optimal model weighting can underplay robust ensemble members by optimizing error cancellation. Prescreening-based subset selection also provides insights about the validity of the model weights. By illustrating the caveats of using non-representative models when optimal model weighting is used, the findings and discussion of this study are pertinent to many other climate services.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Evolving Ensembles of Spiking Neural Networks for Neuromorphic Systems

Evolutionary algorithms have been proposed as a solution to overcome many of the challenges associated with training spiking neural networks. While evolutionary optimization for spiking neural networks is very flexible, its performance has difficulty scaling to complex tasks and correspondingly complex network structures. Here we propose a method for evolving ensembles of spiking neural networks. By using ensemble learning, the flexibility of evolutionary optimization is fully preserved while scaling to more challenging tasks. We test the performance of the proposed method using handwritten digit classification. We investigate multiple strategies for constructing ensembles of spiking neural networks, and demonstrate that evolving ensembles of SNNs offers significant performance advantages over evolutionary optimization.

Elbrecht, Daniel↗
DOE OSTI2021

Resilience-based Recovery Scheduling of Transportation Network in Mixed Traffic Environment: A Deep-Ensemble-Assisted Active Learning Approach

Devising effective post-hazard recovery strategies is critical in enhancing the resilience of transportation networks (TNs). However, existing work does not consider the multiclass users’ travel behavior in network functionality quantification and the metaheuristic solution procedures often suffer from extensive computational burden due to the exploration need in large solution space and the expensive functionality quantification. Here, this study develops a bilevel decision-making framework for the resilience-based recovery scheduling of the TN in a mixed traffic environment with connected and autonomous vehicles (CAVs) and human-driven vehicles (HDVs). The lower level quantifies the TN's functionality over time considering different travel behavior of CAV and HDV users arisen from their different levels of information perception. The upper level presents a novel deep-ensemble-assisted active learning approach to balance optimization performance and computational cost. This framework can help decision makers better quantify the TN's functionality to support effective recovery scheduling of TN with different mixed traffic scenarios ranging from HDV-only to future CAV-dominant traffic. The optimization approach bears the potential to be extended to solving general large-scale network recovery scheduling problems effectively and efficiently. The proposed methodology is demonstrated using a real-world traffic network in Southern California under earthquake considering deterministic and stochastic repair durations.

42 ENGINEERING↗
DOE OSTI2020

Strain-Modulated Platinum–Palladium Nanowires for Oxygen Reduction Reaction

Electrocatalytic activity of alloy nanocatalytsts can be manipulated effectively by tuning their physical properties (ensemble, geometric, and ligand effects) to afford optimal surface structure and compositions for proton exchange membrane fuel cell (PEMFC) application. Herein, highly catalytic platinum-palladium nanowires (Pt n Pd 100-n NWs) with a subtle lattice strain and Boerdijk-Coxeter helix type morphology are synthesized through a surfactant-free, thermal single phase solvent method. X-ray diffraction results show that Pt n Pd 100-n NWs are exposed through the (111) facets and their shrinking or expanding lattice parameters can be modulated by the alloy compositions. Electrochemical results reveal that their high catalytic activity correlates with the lattice shrinking, facets, and bimetallic compositions, showing higher activity when the ratio of Pt and Pd is similar to ~78:22, which is further supported by DFT results. Compared to the nanoparticle type platinum-palladium alloyed catalysts with similar metal compositions (Pt n Pd 100-n NPs), the Pt n Pd 100-n NWs exhibit significantly improved electrocatalytic activity and stability for the oxygen reduction reaction. Our findings open new strategies to design the highly active and stable alloy nanocatalysts with controllable compositions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Models, data, and scripts associated with “Prediction of Distributed River Sediment Respiration Rates using Community-Generated Data and Machine Learning”

This data package is associated with the publication “Prediction of Distributed River Sediment Respiration Rates using Community-Generated Data and Machine Learning’’ submitted to the Journal of Geophysical Research: Machine Learning and Computation (Scheibe et al. 2024). River sediment respiration observations are expensive and labor intensive to obtain and there is no physical model for predicting this quantity. The Worldwide Hydrobiogeochemisty Observation Network for Dynamic River Systems (WHONDRS) observational data set (Goldman et al.; 2020) is used to train machine learning (ML) models to predict respiration rates at unsampled sites. This repository archives training data, ML models, predictions, and model evaluation results for the purposes of reproducibility of the results in the associated manuscript and community reuse of the ML models trained in this project. One of the key challenges in this work was to find an optimum configuration for machine learning models to work with this feature-rich (i.e. 100+ possible input variables) data set. Here, we used a two-tiered approach to managing the analysis of this complex data set: 1) a stacked ensemble of ML models that can automatically optimize hyperparameters to accelerate the process of model selection and tuning and 2) feature permutation importance to iteratively select the most important features (i.e. inputs) to the ML models. The major elements of this ML workflow are modular, portable, open, and cloud-based, thus making this implementation a potential template for other applications. This data package is associated with the GitHub repository found at Please see the file level metadata (flmd; “sl-archive-whondrs_flmd.csv”) for a list of all files contained in this data package and descriptions for each. Please see the data dictionary (dd; “sl-archive-whondrs_dd.csv”) for a list of all column headers contained within comma separated value (csv) files in this data package and descriptions for each. The GitHub repository is organized into five top-level directories: (1) “input_data” holds the training data for the ML models; (2) “ml_models” holds machine learning models trained on the data in “input_data”; (3) “scripts” contains data preprocessing and postprocessing scripts and intermediate results specific to this data set that bookend the ML workflow; (4) “examples” contains the visualization of the results in this repository including plotting scripts for the manuscript (e.g., model evaluation, FPI results) and scripts for running predictions with the ML models (i.e., reusing the trained ML models); (5) “output_data” holds the overall results of the ML model on that branch. Each trained ML model resides on its own branch in the repository; this means that inputs and outputs can be different branch-to-branch. Furthermore, depending on the number of features used to train the ML models, the preprocessing and postprocessing scripts, and their intermediate results, can also be different branch-to-branch. The “main-*” branches are meant to be starting points (i.e. trunks) for each model branch (i.e. sprouts). Please see the Branch Navigation section in the top-level README.md in the GitHub repository for more details. There is also one hidden directory “.github/workflows”. This hidden directory contains information for how to run the ML workflow as an end-to-end automated GitHub Action but it is not needed for reusing the ML models archived here. Please the top-level README.md in the GitHub repository for more details on the automation.

13C↗
DOE OSTI2021

HAIMOS Ensemble Forecasts for Intra-day and Day- Ahead GHI, DNI and Ramps

The objective of this research is to develop a hybrid physics-based/data-driven forecast model to improve direct normal and global horizontal irradiance (DNI and GHI) prediction for horizons ranging from 1 to 72 hours. Project objectives also address key gaps in state-of-the-art solar forecasting: accurate probabilistic solar forecasts and the forecasting of large irradiance ramps (ramp onset and magnitude). The proposed model ensembles Numerical Weather Prediction (NWP) forecasts, determinist physics-based algorithms, and new-generation cloud cover products (high-resolution rapid refresh satellite images and Large Eddy Simulations). The result is the Hybrid Adaptive Input Model Objective Selection (HAIMOS) ensemble model. HAIMOS blends state of the art machine learning methodologies with physics-based models for cloud cover and cloud optical depth forecasts. The technical activities followed a two-pronged strategy. First, the preprocessing of data, the selection of inputs to the nonlinear approximators, the type of approximator and objective functions, and post-processing ensembling techniques included in HAIMOS were all optimized adaptively to find the best model for a specific goal (reduce DNI/GHI forecast error, improve the prediction of ramp onset, etc.). Second, a large effort was put in improving cloud identification and the forecast of cloud cover and cloud optical depth. To this end, new-generation cloud parametrization products were developed in this work. These include improved algorithms to assist in cloud identification, cloud classification and cloud parametrization from satellite images – three key factors in the accuracy of 1 to 6-hours irradiance forecasts and prediction of ramp onset. Furthermore, we also included cloud information extracted from high resolution rapid refresh satellite images (GOES-16) and Large Eddy Simulations (LES). LES was used to model the atmosphere in detail over locations of interest and produce cloud optical depth forecasts. Once these data streams were validated, they were used as input data to the HAIMOS forecast. The model was developed using data from several climatologically distinct locations with potential for high solar penetration. In the last year of the project, we conducted a validation campaign according to the guidelines stipulated by the Topic Area 1 project as described in the FOA. This effort brings, for the first time, proven machine-learning methodologies for generating state-of-the-art solar forecasts interweaved with detailed physics-based models for cloud detection, and cloud optical depth forecasts. HAIMOS will generate accurate irradiance probabilistic forecast to assist in reducing solar generation prediction error. Globally optimized solar forecast models are more likely to impact solar energy stakeholders. The goal of this project was to increase the state-of-the-art forecast skill from their present values of 10 to 35%. At the end of the project, we achieved between 30% and 50% forecast skill across a wide range of horizons for both GHI and DNI.

14 SOLAR ENERGY↗