Modeling and Predicting Heavy-Duty Vehicle Engine-Out and Tailpipe Nitrogen Oxide (NO x ) Emissions Using Deep Learning.

As emissions regulations for transportation become stricter, it is increasingly important to develop accurate nitrogen oxide (NO x ) emissions models for heavy-duty vehicles. However, estimation of transient NO x emissions using physics-based models is challenging due to its highly dynamic nature, which arises from the complex interactions between power demand, engine operation, and exhaust aftertreatment efficiency. As an alternative to physics-based models, a multi-dimensional data-driven approach is proposed as a framework to estimate NO x emissions across an extensive set of representative engine and exhaust aftertreatment system operating conditions. This paper employs Deep Neural Networks (DNN) to develop two models, an engine-out NO x and a tailpipe NO x model, to predict heavy-duty vehicle NO x emissions. The DNN models were developed using variables that are available from On-board Diagnostics from two datasets, an engine dynamometer and a chassis dynamometer dataset. Results from trained DNN models using the engine dynamometer dataset showed that the proposed approach can predict NO x emissions with high accuracy, where R 2 scores are higher than 0.99 for both engine-out and tailpipe NO x models on cold/hot Federal Test Procedure (FTP) and Ramped Mode Cycle (RMC) data. Similarly, the engine-out and tailpipe NO x models using the chassis dynamometer dataset achieved R 2 scores of 0.97 and 0.93, respectively. All models developed in this study have a mean absolute error percentage of approximately 1% relative to maximum NO x in the datasets, which is comparable to that of physical NO x emissions measurement analyzers. The input feature importance studies conducted in this work indicate that high accuracy DNN models (R 2 = 0.92-0.95) could be developed by utilizing minimal significant engine and aftertreatment inputs. This study also demonstrates that DNN NO x emissions models can be very effective tools for fault detection in Selective Catalytic Reduction (SCR) systems.

Experimental evaluation of aerosol mitigation strategies in large open-plan dental clinics.

BACKGROUND: Aerosols are generated routinely during patient care in dentistry. Managing exposure risk requires understanding characteristics of aerosols created during procedures such as those performed using high-speed drills that operate at 200,000 revolutions per minute. METHODS: A trained dentist performed drilling procedures on a manikin's incisors (teeth nos. 8 and 9) using a high-speed drill and high-volume evacuator. The authors used high-speed imaging to visualize the formation and transport of aerosol clouds and particle sampling to measure aerosol concentration and size distribution at several locations. The authors studied several aerosol mitigation strategies. RESULTS: Aerosols produced during high-speed drilling were erratic and yielded high concentrations that were at least an order of magnitude above baseline. High-speed imaging showed aerosols initially travelled at 1 m per second. Owing to erratic behavior of aerosols, supplemental suction was not effective at collecting all aerosols; however, barriers were effective. CONCLUSIONS: Barriers are the most effective mitigation strategy. Other methods studied have limitations and risks. To the authors' knowledge, this article presents the first characterization of aerosols generated during high-speed drilling by a dentist. PRACTICAL IMPLICATIONS: With thorough preoperative planning and the use of this investigation's findings about effectiveness of mitigation strategies as a guide, dental offices may be able to return to prepandemic productivity.

Disease transmission through expiratory aerosols on an urban bus.

Airborne respiratory diseases such as COVID-19 pose significant challenges to public transportation. Several recent outbreaks of SARS-CoV-2 indicate the high risk of transmission among passengers on public buses if special precautions are not taken. This study presents a combined experimental and numerical analysis to identify transmission mechanisms on an urban bus and assess strategies to reduce risk. The effects of the ventilation and air-conditioning systems, opening windows and doors, and wearing masks are analyzed. Specific attention is paid to the transport of submicron- and micron-sized particles relevant to typical respiratory droplets. High-resolution instrumentation was used to measure size distribution and aerosol response time on a campus bus of the University of Michigan under these different conditions. Computational fluid dynamics was employed to measure the airflow within the bus and evaluate risk. A risk metric was adopted based on the number of particles exposed to susceptible passengers. The flow that carries these aerosols is predominantly controlled by the ventilation system, which acts to uniformly distribute the aerosol concentration throughout the bus while simultaneously diluting it with fresh air. The opening of doors and windows was found to reduce the concentration by approximately one half, albeit its benefit does not uniformly impact all passengers on the bus due to the recirculation of airflow caused by entrainment through windows. Finally, it was found that well fitted surgical masks, when worn by both infected and susceptible passengers, can nearly eliminate the transmission of the disease.

Aerosol generation during chest compression and defibrillation in a swine cardiac arrest model.

AIM: It remains unclear whether cardiac arrest (CA) resuscitation generates aerosols that can transmit respiratory pathogens. We hypothesize that chest compression and defibrillation generate aerosols that could contain the SARS-CoV-2 virus in a swine CA model. METHODS: To simulate witnessed CA with bystander-initiated cardiopulmonary resuscitation, 3 female non-intubated swine underwent 4 min of ventricular fibrillation without chest compression or defibrillation (no-flow) followed by ten 2-min cycles of mechanical chest compression and defibrillation without ventilation. The diameter (0.3-10 µm) and quantity of aerosols generated during 45-s intervals of no-flow and chest compression before and after defibrillation were analyzed by a particle analyzer. Aerosols generated from the coughs of 4 healthy human subjects were also compared to aerosols generated by swine. RESULTS: There was no significant difference between the total aerosols generated during chest compression before defibrillation compared to no-flow. In contrast, chest compression after defibrillation generated significantly more aerosols than chest compression before defibrillation or no-flow (72.4 ±â€¯41.6 × 104 vs 12.3 ±â€¯8.3 × 104 vs 10.5 ±â€¯11.2 × 104; p < 0.05), with a shift in particle size toward larger aerosols. Two consecutive human coughs generated 54.7 ±â€¯33.9 × 104 aerosols with a size distribution smaller than post-defibrillation chest compression. CONCLUSIONS: Chest compressions alone did not cause significant aerosol generation in this swine model. However, increased aerosol generation was detected during chest compression immediately following defibrillation. Additional research is needed to elucidate the clinical significance and mechanisms by which aerosol generation during chest compression is modified by defibrillation.

Influence of Ash-Soot Interactions on the Reactivity of Soot from a Gasoline Direct Injection Engine.

Gasoline particulate filters (GPF) are being utilized in certain markets on gasoline direct injection (GDI) vehicles to reduce tailpipe particulate emissions as required by particle number regulations. GPF filtration efficiency is dependent on soot build-up within the filter. Since soot oxidizes within the GPF during normal vehicle operation, an understanding of soot reactivity is important for optimizing aftertreatment architecture and engine calibration. Past work has indicated that gasoline soot reactivity may depend on levels of metallic ash species. In this work, carbonaceous particulate matter from a GDI engine are evaluated from engine operation at a consistent speed and load but with different levels of fuel injection pressures and timings to vary the relative ash to soot ratio. Soot reactivity is found to vary significantly with the ratio of ash to soot present. Interestingly, the more reactive soots possess a unique oxidation profile by which a conventional Arrhenius type expression cannot be used to quantify reactivity. To understand the mechanisms driving such distinct oxidation differences, soot samples are analyzed after being partially oxidized. Particulate characteristics are evaluated by x-ray photoelectron spectroscopy (XPS), Raman spectroscopy, high-resolution transmission electron microscopy (HR-TEM), and scanning transmission electron microscopy with energy dispersive spectroscopy (STEM + EDS). A mechanism is proposed that may explain further why ash affects gasoline soot reactivity to the extent seen in this and other work.