Engineering topics

Negri, Cristina

Publications and source records attributed to Negri, Cristina.

6 records

DOE OSTI2024

RESIN: Responsible Innovation for Highly Recyclable Plastics - TASK 4: Risk Assessment Framework

This report covers the entirety of Task 4, but its main purpose is to deliver milestones ML4.4 and ML4.5, the last two SOPO milestones under Task 4. ML4.4 reports on the compatibility of polymer properties that affect both environmental performance and functional performance and the tradeoffs involved in turning these properties to the benefit of each. ML4.5 presents a complete set of information on the critical properties of benign target products, where benign products are defined as those with the shortest environmental lifetime which meet performance requirements. To support and provide context to the discussions of ML4.4 and ML4.5 and to provide a complete picture of Task 4, milestones 4.1-4.3 are briefly summarized at the beginning of the report. The discussion of ML4.4 introduces the notion of polymer persistence as a proxy for environmental risk. It then discusses the development and comparison of two machine leaning models explored for estimating polymer degradation rates, a random forest (RF) classifier and an RF regressor. Given the advantage of continuous outputs rather than simple classes, the RF regressor was incorporated in the Excel risk calculator, which to this point could implement the objectives of subtasks 4.1-4.3, estimating polymer release and redistribution. The ML4.4 discussion then addresses the effect of each of the three polymer features used by the RF regressor on polymer functional performance. These features are number molecular weight (Mn), glass transition temperature (T g ) and heat of fusion (H fus ). The discussion of ML4.5 reviews the conceptual framework of the risk model, which served as the foundation for developing the Excel risk calculator and describes the use of, and assumptions within, the calculator. Appendix A is further provided as a user’s guide for the calculator. To demonstrate how the calculator is intended to be used by developers in the design of low-risk polymers, an analysis of 27 hypothetical polymers defined by varying values for the three polymer features used by the RF regressor is presented. The range of parameter values selected produces estimates of polymer degradation rates and lifetimes that may be typical of various consumer products made from polyurethane polymers and shows how changes in polymer features affect lifetimes. The demonstration also predicts the final distribution of released polymer materials in environmental compartments as a function of consumer product mix and assumed leakage rates of end-of-life processes. Lastly, this report summarizes the achievement of Task 4 goals from original conception to final delivery and discusses how and to what degree to which each subtask goal was achieved.

36 MATERIALS SCIENCE↗

DOE OSTI2023

Predicting Biomass Yields of Advanced Switchgrass Cultivars for Bioenergy and Ecosystem Services Using Machine Learning

The production of advanced perennial bioenergy crops within marginal areas of the agricultural landscape is gaining interest due to its potential to sustainably produce feedstocks for biofuels and bioproducts while also improving the sustainability and resilience of commodity crop production. However, predicting the biomass yields of this production system is challenging because marginal areas are often relatively small and spread around agricultural fields and are typically associated with various abiotic conditions that limit crop production. Machine learning (ML) offers a viable solution as a biomass yield prediction tool because it is suited to predicting relationships with complex functional associations. The objectives of this study were to (1) evaluate the accuracy of commonly applied ML algorithms in agricultural applications for predicting the biomass yields of advanced switchgrass cultivars for bioenergy and ecosystem services and (2) determine the most important biomass yield predictors. Datasets on biomass yield, weather, land marginality, soil properties, and agronomic management were generated from three field study sites in two U.S. Midwest states (Illinois and Iowa) over three growing seasons. The ML algorithms evaluated in the study included random forests (RFs), gradient boosting machines (GBMs), artificial neural networks (ANNs), K-neighbors regressor (KNR), AdaBoost regressor (ABR), and partial least squares regression (PLSR). Coefficient of determination (R 2 ) and mean absolute error (MAE) were used to evaluate the predictive accuracy of the tested algorithms. Results showed that the ensemble methods, RF (R 2 = 0.86, MAE = 0.62 Mg/ha), GBM (R 2 = 0.88, MAE = 0.57 Mg/ha), and GBM (R 2 = 0.78, MAE = 0.66 Mg/ha), were the most accurate in predicting biomass yields of the Independence, Liberty, and Shawnee switchgrass cultivars, respectively. This is in agreement with similar studies that apply ML to multi-feature problems where traditional statistical methods are less applicable and datasets used were considered to be relatively small for ANNs. Consistent with previous studies on switchgrass, the most important predictors of biomass yield included average annual temperature, average growing season temperature, sum of the growing season precipitation, field slope, and elevation. This study helps pave the way for applying ML as a management tool for alternative bioenergy landscapes where understanding agronomic and environmental performance of a multifunctional cropping system seasonally and interannually at the sub-field scale is critical.

09 BIOMASS FUELS↗

DOE OSTI2023

Microbiome response in an urban river system is dominated by seasonality over wastewater treatment upgrades

Background: Microorganisms such as coliform-forming bacteria are commonly used to assess freshwater quality for drinking and recreational use. However, such organisms do not exist in isolation; they exist within the context of dynamic, interactive microbial communities which vary through space and time. Elucidating spatiotemporal microbial dynamics is imperative for discriminating robust community changes from ephemeral ecological trends, and for improving our overall understanding of the relationship between microbial communities and ecosystem health. We conducted a seven-year (2013–2019) microbial time-series investigation in the Chicago Area Waterways (CAWS): an urban river system which, in 2016, experienced substantial upgrades to disinfection processes at two wastewater reclamation plants (WRPs) that discharge into the CAWS and improved stormwater capture, to improve river water quality and reduce flooding. Using culture-independent and culture-dependent approaches, we compared CAWS microbial ecology before and after the intervention. Results: Examinations of time-resolved beta distances between WRP-adjacent sites showed that community similarity measures were often consistent with the spatial orientation of site locations to one another and to the WRP outfalls. Fecal coliform results suggested that upgrades reduced coliform-associated bacteria in the effluent and the downstream river community. However, examinations of whole community changes through time suggest that the upgrades did little to affect overall riverine community dynamics, which instead were overwhelmingly driven by yearly patterns consistent with seasonality. Conclusions: This study presents a systematic effort to combine 16S rRNA gene amplicon sequencing with traditional culture-based methods to evaluate the influence of treatment innovations and systems upgrades on the microbiome of the Chicago Area Waterway System, representing the longest and most comprehensive characterization of the microbiome of an urban waterway yet attempted. We found that the systems upgrades were successful in improving specific water quality measures immediately downstream of wastewater outflows. Additionally, we found that the implementation of the water quality improvement measures to the river system did not disrupt the overall dynamics of the downstream microbial community, which remained heavily influenced by seasonal trends. Such results emphasize the dynamic nature of microbiomes in open environmental systems such as the CAWS, but also suggest that the seasonal oscillations remain consistent even when perturbed.

16S rRNA gene sequencing↗

DOE OSTI2021

America Resilient Climate Conference

On April 14, 2021, scientists, policymakers, and other interested parties from research institutes, academia, and other organizations gathered together virtually at the America Resilient Climate Conference to discuss one of the most pressing challenges of the 21st century: building resilience to climate change. Climate change affects the security and health of all Americans. Coastal areas are enduring more frequent and severe flooding due to sea level rise and storm surge; western states and Alaska have experienced increasingly devastating wildfires, driven in part by hotter, drier, and longer fire seasons; and communities across the nation have suffered through extreme precipitation events and heat waves. Even if emissions are reduced aggressively in the near future, the world—and the United States—will continue to feel the impacts of climate change for decades to come, due to the continued accumulation of greenhouse gasses in the atmosphere. Consequently, it is essential to act now to protect natural and human assets from the gradual—as well as extreme—impacts of a changing climate. To build resilient communities, leaders and community members need science-based information about the potential impacts climate change will have decades into the future and for specific regions. Therefore, it is essential to develop high-resolution climate models that can project both various climate impacts and the interactions between earth system variables and humans down to regional and local scales. Collecting and curating data for such models and their computational requirements poses large challenges. In the future, artificial intelligence will be needed to increase their accuracy and reduce associated uncertainties. The investments in Earth system science and artificial intelligence made by the U.S. Department of Energy and other federal entities will be essential in addressing these challenges.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Emergent Concepts from a Community Ideation on AI4ESP

This AI4ESP white paper is the outcome of an open ideation session led by PNNL that numerous institutions participated in. The leadership team included Amy Goldman, Huiying Ren, Tim Scheibe, and James Stegen. The white paper is broken into two sections: (1) A summary of the ideation session, and (2) a short summary of a ‘meta idea’ that emerged from combining multiple ideas spanning the three primary focal area themes articulated in BER’s AI4ESP guidance: A - Data Acquisition and Assimilation, B - Predictive Modeling, and C - Insight from Complex Data. The white paper, therefore, has two goals. The first section is meant to be a resource for innovating around how to pursue open, community-based ideation in virtual environments. The second section is meant as an example of the kinds of ‘meta ideas’ that can emerge from bringing diverse ideas together in a single shared space. We consider the second section to be a submission to BER’s request for white papers around AI4ESP.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Assessing Climate Risks to Midwest Infrastructure Workshop

Stakeholders in industry, agriculture, and city governments are struggling to understand how to manage climate change risks. Climate scientists, meanwhile, want to know what specific information their fellow citizens need from them to take action.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗