Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “decision tree”

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 559 records · Page 31

Pumped Storage Hydro Model Evaluation Report Task 1: Tool Assessment

Pumped storage hydro (PSH) projects support various aspects of power system operations. However, determining the value of PSH projects and the many services and contributions to the system they provide can be a challenge. The PSH techno-economic studies being carried out by a 5-lab consortium are defining extensive modeling approaches for evaluating the economic benefits of PSH projects. To enhance the impact of the valuation approach and follow-on techno-economic studies, Pacific Northwest National Laboratory (PNNL) will be developing a set of online tools that industry could adopt and use to advance PSH projects in the US. The problem is that the research team is not entirely clear on the features of the tools that would offer the most significant impact to industry. Thus, year one of this effort is designed to first define the needs of industry before developing the PSH valuation tool. The first step in defining industry needs is to evaluate existing tool packages, including those created by DOE and those by industry groups, to both measure usage levels and define tool attributes that maximize industry impact. The second step in this process will be to hold a workshop designed to elicit direct industry input. Given the breadth and nature of the task, which is to define ideal model attributes, the research team did not constrain the assessment to storage valuation tools. Several tools were assessed, and their characteristics compared against a list of “ideal” attributes that included ease of use, generalizability of results, and analytical flexibility. Information regarding usage and impact were then compared to specific model attributes to define which attributes appear to drive success. The following key lessons and implications can be drawn from the analysis: Models with high ease of use were often the most adopted by industry, though they did not necessarily have high publication reference records, meaning a published article made direct reference to the tool. For example, REScheck TM and COMcheck TM , two tools which share a site, have seen a high amount of web traffic and usage. Similarly, Storage Value Estimate Tool (StorageVET®), an energy storage valuation model, has a large number of organizations using it despite a low publication reference record. A common feature between these models is that they are web-hosted, theoretically making them more easily accessible with graphic user interfaces (GUIs) that are more navigable; Models with higher usage statistics were typically free and/or open source. While it may be unsurprising that this attribute could be considered a draw for users, there is the potential for overestimating the tool's value based on popularity due to the fact that some users may have used the model simply because it was free and not because it is the best model; Models that offer more in-depth analysis and a wide coverage of applications (e.g. VOLTTRON™, GridLAB-D™, and Integrated Valuation of Ecosystem Services and Tradeoffs (InVest)) have shown a high number of publication references as well as showing a high number of downloads in the case of GridLAB-D™. It should be noted that the figure below does not include the model with the highest number of publication references (EnergyPlus™) as it has been removed to show clearer distinction between the remaining tools; The most popular model within the evaluation list was EnergyPlus™. This tool has the highest number of publication references (2,034) of all models considered and contains both a high ease of use level as well as broad application flexibility. Its high usage may be correlated, however, with the presumed wider audience conducting energy consumption modeling compared to the audiences for some of the other tools in this list; Models that were low in both general usage and publication reference generally did not exhibit characteristics considered ideal by the research team. Some of these models offered a narrower range of analytical applications or lower generalizability of results. This would automatically restrict the pool of potential users to those who fall within a narrow range of specific technologies or applications. This report also explores two model structures for the PSH tool: decision tree-based, which would guide the user through the valuation methods defined in the PSH techno-economic studies currently being carried out by DOE, and an all-inclusive comprehensive model. The strengths of each were explored and a hybrid approach was defined that would embed a price taker model into the decision tree approach. Such an approach would provide a comprehensive approach for modeling small-scale PSH while recognizing the challenges associated with implementing a price-maker approach in an all-inclusive model. Insight gained in this analysis of other models can aid in narrowing down the list of effective and sought-after attributes, and in defining model structure. Next steps toward developing a model for storage valuation will be to gain direct industry input through workshops.

13 HYDRO ENERGY↗

A Continuous Action Space Tree search for INverse desiGn (CASTING) framework for materials discovery

Abstract Material properties share an intrinsic relationship with their structural attributes, making inverse design approaches crucial for discovering new materials with desired functionalities. Reinforcement Learning (RL) approaches are emerging as powerful inverse design tools, often functioning in discrete action spaces. This constrains their application in materials design problems, which involve continuous search spaces. Here, we introduce an RL-based framework CASTING (Continuous Action Space Tree Search for inverse design), that employs a decision tree-based Monte Carlo Tree Search (MCTS) algorithm with continuous space adaptation through modified policies and sampling. Using representative examples like Silver (Ag) for metals, Carbon (C) for covalent systems, and multicomponent systems such as graphane, boron nitride, and complex correlated oxides, we showcase its accuracy, convergence speed, and scalability in materials discovery and design. Furthermore, with the inverse design of super-hard Carbon phases, we demonstrate CASTING’s utility in discovering metastable phases tailored to user-defined target properties and preferences.

36 MATERIALS SCIENCE↗

Perovskite neural trees

Trees are used by animals, humans and machines to classify information and make decisions. Natural tree structures displayed by synapses of the brain involves potentiation and depression capable of branching and is essential for survival and learning. Demonstration of such features in synthetic matter is challenging due to the need to host a complex energy landscape capable of learning, memory and electrical interrogation. We report experimental realization of tree-like conductance states at room temperature in strongly correlated perovskite nickelates by modulating proton distribution under high speed electric pulses. This demonstration represents physical realization of ultrametric trees, a concept from number theory applied to the study of spin glasses in physics that inspired early neural network theory dating almost forty years ago. We apply the tree-like memory features in spiking neural networks to demonstrate high fidelity object recognition, and in future can open new directions for neuromorphic computing and artificial intelligence.

36 MATERIALS SCIENCE↗

Challenges for tree officers to enhance the provision of regulating ecosystem services from urban forests

Urbanisation and a changing climate are leading to more frequent and severe flood, heat and air pollution episodes in Britain's cities. Interest in nature-based solutions to these urban problems is growing, with urban forests potentially able to provide a range of regulating ecosystem services such as stormwater attenuation, heat amelioration and air purification. The extent to which these benefits are realized is largely dependent on urban forest management objectives, the availability of funding, and the understanding of ecosystem service concepts within local governments, the primary delivery agents of urban forests. This study aims to establish the extent to which British local authorities actively manage their urban forests for regulating ecosystem services, and identify which resources local authorities most need in order to enhance provision of ecosystem services by Britain's urban forests. Interviews were carried out with staff responsible for tree management decisions in fifteen major local authorities from across Britain, selected on the basis of their urban nature and high population density. Local authorities have a reactive approach to urban forest management, driven by human health and safety concerns and complaints about tree disservices. There is relatively little focus on ensuring provision of regulating ecosystem services, despite awareness by tree officers of the key role that urban forests can play in alleviating chronic air pollution, flood risk and urban heat anomalies. However, this is expected to become a greater focus in future provided that existing constraints – lack of understanding of ecosystem services amongst key stakeholders, limited political support, funding constraints – can be overcome. Our findings suggest that the adoption of a proactive urban forest strategy, underpinned by quantified and valued urban forest-based ecosystem services provision data, and innovative private sector funding mechanisms, can facilitate a change to a proactive, ecosystem services approach to urban forest management. - Highlights: • British urban forests are managed reactively, in response to risks and complaints. • Tree officers are aware of urban forests’ delivery of regulating ecosystem services. • Limited funding and lack of political support are obstacles to a proactive approach. • Awareness raising via ecosystem service valuation and strategic planning is needed. • Research into a public-private urban forest funding model would be beneficial.

54 ENVIRONMENTAL SCIENCES↗

Inexact fuzzy-stochastic constraint-softened programming - A case study for waste management

In this study, an inexact fuzzy-stochastic constraint-softened programming method is developed for municipal solid waste (MSW) management under uncertainty. The developed method can deal with multiple uncertainties presented in terms of fuzzy sets, interval values and random variables. Moreover, a number of violation levels for the system constraints are allowed. This is realized through introduction of violation variables to soften system constraints, such that the model's decision space can be expanded under demanding conditions. This can help generate a range of decision alternatives under various conditions, allowing in-depth analyses of tradeoffs among economic objective, satisfaction degree, and constraint-violation risk. The developed method is applied to a case study of planning a MSW management system. The uncertain and dynamic information can be incorporated within a multi-layer scenario tree; revised decisions are permitted in each time period based on the realized values of uncertain events. Solutions associated with different satisfaction degree levels have been generated, corresponding to different constraint-violation risks. They are useful for supporting decisions of waste flow allocation and system-capacity expansion within a multistage context.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Predictive understanding of the surface tension and velocity of sound in ionic liquids using machine learning

Knowledge of the physical properties of ionic liquids (ILs), such as the surface tension and speed of sound, is important for both industrial and research applications. Unfortunately, technical challenges and costs limit exhaustive experimental screening efforts of ILs for these critical properties. Previous work has demonstrated that the use of quantum-mechanics-based thermochemical property prediction tools, such as the conductor-like screening model for real solvents, when combined with machine learning (ML) approaches, may provide an alternative pathway to guide the rapid screening and design of ILs for desired physiochemical properties. However, the question of which machine-learning approaches are most appropriate remains. In the present study, we examine how different ML architectures, ranging from tree-based approaches to feed-forward artificial neural networks, perform in generating nonlinear multivariate quantitative structure–property relationship models for the prediction of the temperature- and pressure-dependent surface tension of and speed of sound in ILs over a wide range of surface tensions (16.9–76.2 mN/m) and speeds of sound (1009.7–1992 m/s). The ML models are further interrogated using the powerful interpretation method, shapley additive explanations. We find that several different ML models provide high accuracy, according to traditional statistical metrics. The decision tree-based approaches appear to be the most accurate and precise, with extreme gradient-boosting trees and gradient-boosting trees being the best performers. However, our results also indicate that the promise of using machine-learning to gain deep insights into the underlying physics driving structure–property relationships in ILs may still be somewhat premature.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Accelerating Random Forest Classification on GPU and FPGA

Random Forests (RFs) are a commonly used machine learning method for classification and regression tasks spanning a variety of application domains, including bioinformatics, business analytics, and software optimization. While prior work has focused primarily on improving performance of the training of RFs, many applications, such as malware identification, cancer prediction, and banking fraud detection, require fast RF classification. In this work, we accelerate RF classification on GPU and FPGA. In order to provide efficient support for large datasets, we propose a hierarchical memory layout suitable to the GPU/FPGA memory hierarchy. We design three RF classification code variants based on that layout, and we investigate GPU- and FPGA-specific considerations for these kernels. Our experimental evaluation, performed on an Nvidia Xp GPU and on a Xilinx Alveo U250 FPGA accelerator card using publicly available datasets on the scale of millions of samples and tens of features, covers various aspects. First, we evaluate the performance benefits of our hierarchical data structure over the standard compressed sparse row (CSR) format. Second, we compare our GPU implementation with cuML, a machine learning library targeting Nvidia GPUs. Third, we explore the performance/accuracy tradeoff resulting from the use of different tree depths in the RF. Finally, we perform a comparative performance analysis of our GPU and FPGA implementations. Our evaluation shows that for high accuracy targets, our GPU implementation yields 5-9x speedup over CSR, and up to a 2x speedup over cuML.

FPGA, Xilinx FPGA, GPU, Random Forest classificati↗

Mesoscopic Modeling and Rapid Simulation of Incremental Changes in Epidemic Scenarios on GPUs

In simulation-based studies and analyses of epidemics, a major challenge lies in resolving the conflict between fidelity of models and the speed of their simulation. Another related challenge arises in dealing with the large number of what–if scenarios that need to be explored. Here, we describe new computational methods that together provide an approach to dealing with both challenges. A mesoscopic modeling approach is described that strikes a middle ground between macroscopic models based on coupled differential equations and microscopic models built on fine-grained behaviors at the individual entity level. The mesoscopic approach offers the ability to incorporate complex compositions of multiple layers of dynamics even while retaining the potential for aggregate behaviors at varying levels. It also is an excellent match to the accelerator-based architectures of modern computing platforms in which graphical processing units (GPUs) can be exploited for fast simulation via the parallel execution mode of single instruction multiple thread (SIMT). The challenge of simulating a large number of scenarios is addressed via a method of sharing model state and computation across a tree of what–if scenarios that are localized, incremental changes to a large base simulation. A combination of the mesoscopic modeling approach and the incremental what–if scenario tree evaluation has been implemented in the software on modern GPUs. Synthetic simulation scenarios are presented to demonstrate the computational characteristics of our approach. Results from the experiments with large population data, including USA, UK, and India, illustrate the modeling methodology and computational performance on thousands of synthetically generated what–if scenarios. Execution of our implementation scaled to 8192 GPUs of supercomputing platforms demonstrates the ability to rapidly evaluate what–if scenarios several orders of magnitude faster than the conventional methods.

97 MATHEMATICS AND COMPUTING↗

Early Fault Detection in Particle Accelerator Power Electronics Using Ensemble Learning

Early fault detection and fault prognosis are crucial to ensure efficient and safe operations of complex engineering systems such as the Spallation Neutron Source (SNS) and its power electronics (high voltage converter modulators). Following an advanced experimental facility setup that mimics SNS operating conditions, the authors successfully conducted 21 early fault detection experiments, where fault precursors are introduced in the system to a degree enough to cause degradation in the waveform signals, but not enough to reach a real fault. Nine different machine learning techniques based on ensemble trees, convolutional neural networks, support vector machines, and hierarchical voting ensembles are proposed to detect the fault precursors. Although all 9 models have shown a perfect and identical performance during the training and testing phase, the performance of most models has decreased in the next test phase once they got exposed to realworld data from the 21 experiments. The hierarchical voting ensemble, which features multiple layers of diverse models, maintains a distinguished performance in early detection of the fault precursors with 95% success rate (20/21 tests), followed by adaboost and extremely randomized trees with 52% and 48% success rates, respectively. The support vector machine models were the worst with only 24% success rate (5/21 tests). The study concluded that a successful implementation of machine learning in the SNS or particle accelerator power systems would require a major upgrade in the controller and the data acquisition system to facilitate streaming and handling big data for the machine learning models. In addition, this study shows that the best performing models were diverse and based on the ensemble concept to reduce the bias and hyperparameter sensitivity of individual models.

43 PARTICLE ACCELERATORS↗

Scalable Cloning on Large-Scale GPU Platforms with Application to Time-Stepped Simulations on Grids

Cloning is a technique to efficiently simulate a tree of multiple what-if scenarios that are unraveled during the course of a base simulation. However, cloned execution is highly challenging to realize on large, distributed memory computing platforms, due to the dynamic nature of the computational load across clones, and due to the complex dependencies spanning the clone tree. In this paper, we present the conceptual simulation framework, algorithmic foundations, and runtime interface of CloneX, a new system we designed for scalable simulation cloning. It efficiently and dynamically creates whole logical copies of a dynamic tree of simulations across a large parallel system without full physical duplication of computation and memory. The performance of a prototype implementation executed on up to 1,024 graphical processing units of a supercomputing system has been evaluated with three benchmarks—heat diffusion, forest fire, and disease propagation models—delivering a speed up of over two orders of magnitude compared to replicated runs. Finally, the results demonstrate a significantly faster and scalable way to execute many what-if scenario ensembles of large simulations via cloning using the CloneX interface.

97 MATHEMATICS AND COMPUTING↗

Characterization of genetic variability of Venezuelan equine encephalitis viruses

Venezuelan equine encephalitis virus (VEEV) is a mosquito-borne alphavirus that has caused large outbreaks of severe illness in both horses and humans. New approaches are needed to rapidly infer the origin of a newly discovered VEEV strain, estimate its equine amplification and resultant epidemic potential, and predict human virulence phenotype. We performed whole genome single nucleotide polymorphism (SNP) analysis of all available VEE antigenic complex genomes, verified that a SNP-based phylogeny accurately captured the features of a phylogenetic tree based on multiple sequence alignment, and developed a high resolution genome-wide SNP microarray. We used the microarray to analyze a broad panel of VEEV isolates, found excellent concordance between array- and sequence-based SNP calls, genotyped unsequenced isolates, and placed them on a phylogeny with sequenced genomes. The microarray successfully genotyped VEEV directly from tissue samples of an infected mouse, bypassing the need for viral isolation, culture and genomic sequencing. Lastly, we identified genomic variants associated with serotypes and host species, revealing a complex relationship between genotype and phenotype.

59 BASIC BIOLOGICAL SCIENCES↗

Learning epistatic polygenic phenotypes with Boolean interactions

Detecting epistatic drivers of human phenotypes is a considerable challenge. Traditional approaches use regression to sequentially test multiplicative interaction terms involving pairs of genetic variants. For higher-order interactions and genome-wide large-scale data, this strategy is computationally intractable. Moreover, multiplicative terms used in regression modeling may not capture the form of biological interactions. Building on the Predictability, Computability, Stability (PCS) framework, we introduce the epiTree pipeline to extract higher-order interactions from genomic data using tree-based models. The epiTree pipeline first selects a set of variants derived from tissue-specific estimates of gene expression. Next, it uses iterative random forests (iRF) to search training data for candidate Boolean interactions (pairwise and higher-order). We derive significance tests for interactions, based on a stabilized likelihood ratio test, by simulating Boolean tree-structured null (no epistasis) and alternative (epistasis) distributions on hold-out test data. Finally, our pipeline computes PCS epistasis p-values that probabilisticly quantify improvement in prediction accuracy via bootstrap sampling on the test set. We validate the epiTree pipeline in two case studies using data from the UK Biobank: predicting red hair and multiple sclerosis (MS). In the case of predicting red hair, epiTree recovers known epistatic interactions surrounding MC1R and novel interactions, representing non-linearities not captured by logistic regression models. In the case of predicting MS, a more complex phenotype than red hair, epiTree rankings prioritize novel interactions surrounding HLA-DRB1 , a variant previously associated with MS in several populations. Taken together, these results highlight the potential for epiTree rankings to help reduce the design space for follow up experiments.

59 BASIC BIOLOGICAL SCIENCES↗

Optimal Control of Markov Processes with Age-Dependent Transition Rates

We study optimal control of Markov processes with age-dependent transition rates. The control policy is chosen continuously over time based on the state of the process and its age. We study infinite horizon discounted cost and infinite horizon average cost problems. Our approach is via the construction of an equivalent semi-Markov decision process. We characterise the value function and optimal controls for both discounted and average cost cases.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Data-enabled structure–property mappings for lanthanide-activated inorganic scintillators

The use of machine learning (ML) models toward development of validated structure–property relationships for two fundamental properties of activated inorganic scintillators for high energy radiation detection, namely the light yield (LY) and the decay time constant, is explored. The ML models are built on easily accessible proxies of materials—interchangeably referred to as features, descriptors or fingerprints—that are carefully selected on the basis of a physical understanding of the scintillation mechanism. Our study indicates that the developed physics-based ML models employing kernel ridge regression (KRR) and AdaBoost algorithm applied on top of a decision tree-based regression are able to “learn” the underlying design rules in a multi-dimensional feature space and thereby enable reasonably accurate predictions of the two target properties on unseen compounds (i.e., on a held-out test set). For instance, within a set of twenty-five cerium- or europium-doped scintillator materials, our analysis reveals a strong correlation between the average ionic part of the dielectric constant and the LY, irrespective of the specific chemistry of the compounds, indicating that the average ionic part of the dielectric constant is a particularly relevant descriptor toward prediction of the LY. Our results also demonstrate that, despite the use of small training datasets, the developed models are able to quickly distinguish high performing chemistries from those with relatively poor performance and therefore can play a crucial role in screening of new compounds with an attractive combination of targeted properties. Finally, the present study provides necessary motivation for future efforts involving ML models with relatively large training datasets, vast feature space explorations, and experimental design in search of promising novel scintillator chemistries.

36 MATERIALS SCIENCE↗

A Machine Learning Model for Predicting Composition of Catalytic Coprocessing Products from Molecular Beam Mass Spectra

Demand for the development of an automated and integrated refining process for biofuels has increased in recent years due to the lack of generalized process inspection tools. In bio-oil upgrading processes, all process variables are maintained based on the offline specification of intermediates and products. A lack of real-time product specifications in batch-wise monitoring can cause process failure and wasted resources. Therefore, there is a need for a fast and accurate intermediates/product specification tool that can be used for real-time specification to reduce waste and mitigate the risk of process failure. Here, to address this gap, we developed a machine learning (ML) model for predicting speciated bio-oil composition, including paraffin, iso-paraffins, olefins, naphthene, and aromatics. The model is trained using the mass spectra from upgraded products collected in the vapor phase before condensation and predicts the composition of the condensed product. Training ML models using raw mass spectra is challenging due to numerous overlapped peaks originating from different parent compounds. With this in mind, we propose a protocol that (i) transforms raw mass spectra to chemistry-inspired predefined features and (ii) trains decision tree-based models using these features. Our results show that the random forest model was robust against overfitting and had the highest accuracy compared to other models. Moreover, a stochastic ablation method determined the eight most significant features while maximizing the accuracy. Our protocol facilitates real-time compositional analysis of upgraded bio-oils and thus real-time process monitoring. Additionally, this protocol enables the rational design of efficient catalysts and the determination of optimal process conditions.

09 BIOMASS FUELS↗

Weakly supervised anomaly detection with event-level variables

We introduce a new topology for weakly supervised anomaly detection searches, diobject plus X. In this topology, one looks for a resonance decaying to two standard model particles produced in association with other anomalous event activity (X). This additional activity is used for classification. We demonstrate how anomaly detection techniques which have been developed for dijet searches focusing on jet substructure anomalies can be applied to event-level anomaly detection in this topology. To robustly capture event-level features of multiparticle kinematics, we employ new physically motivated variables derived from the geometric structure of a collision’s phase space manifold. As a proof of concept, we explore the application of this approach to several benchmark signals in the di-𝜏 and di-𝜇 plus X final states. We demonstrate that our anomaly detection approach can reach discovery-level significances for signals that would be missed in a conventional bump-hunt approach.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Machine learning based job status prediction in scientific clusters

Large high-performance computing systems are built with increasing number of components with more CPU cores, more memory, and more storage space. At the same time, scientific applications have been growing in complexity. Together, they are leading to more frequent unsuccessful job statuses on HPC systems. From measured job statuses, 23.4% of CPU time was spent to the unsuccessful jobs. Here, we set out to study whether these unsuccessful job statuses could be anticipated from known job characteristics. To explore this possibility, we have developed a job status prediction method for the execution of jobs on scientific clusters. The Random Forests algorithm was applied to extract and characterize the patterns of unsuccessful job statuses. Experimental results show that our method can predict the unsuccessful job statuses from the monitored ongoing job executions in 99.8% the cases with 83.6% recall and 94.8% precision. Lastly, this prediction accuracy can be sufficiently high that it can be used to mitigation procedures of predicted failures.

97 MATHEMATICS AND COMPUTING↗