Jose Enrique Bernardo

Development of a Rapid Fleet-Level Noise Computation Model

Future air transportation demand forecasts suggest that environmental concerns such as noise will be exacerbated beyond their current level. Although detailed airport noise modeling with tools such as the integrated noise model and the Federal Aviation Administration’s Aviation Environmental Design Tool are available, these software require relatively long setup and run times due to the number of inputs available to the user and the general fidelity level of the models. A rapid, flexible, and more simplified method that reduces the input variables to a critical few and can provide results in minutes is desired to evaluate fleet-level metrics with respect to new technologies or forecasted changes in demand. Current lower-fidelity methods only calculate a change in contour area due to changes within the overall fleet composition. These methods cannot account for the shape of the contour. This paper presents a rapid airport noise computation model that leverages the fidelity of detailed models. By performing...

DNL Contour Area Sensitivity to Fleet-Level Operational Characteristics

Several aviation demand forecasts project increases in operations in the coming decades. As a result, the environmental impacts of aviation operations are likely to increase if proper technology mitigation strategies are not pursued. Several technology programs are supporting the search for technologies such that fuel burn, NOx emissions, and noise do not become serious constraints on aviation growth. Vehicle-level environmental technologies, however, must ultimately be judged at the fleet-level to provide a complete picture of their system-level impact. This evaluation is particularly difficult with respect to noise, as it has spatial and temporal components and is measured in exposure to noise levels above DNL 65 dB. This exposure is evaluated by overlaying constant level noise contours with population information. At the fleet-level the summation of airport-level contour areas can be tracked as a primary metric and serve as a reasonable analog for the population exposed. Noise is typically expensive to compute, and comprehensive sensitivity analyses have not previously been possible. Because the development of rapid automated noise models make use of simplifying assumptions to provide rapid estimates of airport-level noise, these analyses are now possible through the execution of Design of Experiments (DOE). The results of these DOEs can be used to build surrogate models and examine the relative impacts of the important operational variables with respect to noise. Utilizing Generic Vehicle and Generic Airport models to represent the aircraft and airport diversity present in the system, this research will perform a contour sensitivity analysis at different airport types to examine the importance of aircraft types, operations, trip length distribution, and other factors to contour measures. The surrogate models developed were applied to a future scenario analysis simulating technology and market performance factors to identify potential situations that would result in system-level reduction.

Development of a Generic Fleet-Level Noise Methodology

Future air transportation demand forecasts suggest that environmental concerns such as noise will be exacerbated beyond their current level. Although detailed airport noise modeling is available with the current fleet, a rapid, flexible, and more generic method is desired to evaluate fleet-level metrics with respect to new technologies or forecasted changes in demand. Although some work has been done to this extent, only the area of the 65 dB contour has been considered as the relevant metric. These methods cannot account for the shape of the contour, which has far-reaching implications for the ultimate metric: population affected. Moreover, the 65 dB contour as the lone contour of interest is a paradigm that, recently, has begun shifting to include other contour levels. This paper presents a generic fleet-level noise methodology that leverages the fidelity of detailed modeling software. By performing generic aircraft operations up front, these events can be rapidly recombined later to perform trades of various noise mitigation strategies. By moving the detailed noise modeling ‘off-line’ and making the appropriate assumptions, some of the fidelity of these models can be propagated earlier into the decision-making process. The methodology was used to demonstrate two simple proofs of concept that evaluate the method by accuracy and process criteria. Finally, discussions for validation of assumptions and future work to improve accuracy are included.

Average Generic Vehicle Method for Fleet-Level Analysis of Noise and Emission Tradeoffs

A method is proposed for modeling average generic vehicles for fleet-level analysis of technology impacts on environmental metrics. Vehicles are grouped into classes distinguished by aircraft-level metrics, which include fuel burn (a surrogate of carbon dioxide emissions), nitrogen oxide species (NOx) emissions, and sound exposure level noise contours. Target metrics are established for a subset of 94 airports by designing a series of tests of sequentially increasing complexity, with ideal average generic vehicle models minimizing the error distributions across these airports when aircraft-level performance is aggregated to fleet levels. A Latin hypercube design of experiments is employed to explore the aircraft-level design space and construct surrogate models for each metric. Monte Carlo samplings of these surrogate models are paired with desirability functions to rapidly identify locations in the design space that simultaneously match aggregate fuel burn, NOx emissions, and day/night-level contours. In...

Development of Generic Airport Categories for Rapid Fleet-level Noise Modeling

Operations forecast are projecting significant growth in total operations in the United States and internationally. As a result there has been a concerted effort to identify technology options to reduce the environmental impacts of aviation including fuel burn, NOx , and noise emissions. To rapidly evaluate the impact of large sets of diverse and interacting technologies, a screening fidelity generic fleet-level approach to measuring environmental impacts is required. Fuel burn and NOx emissions are easily scaled to the fleet-level by generalizing specific flights by aircraft type and route distance. Noise requires airport-level analysis because it is a local and spatial effect. Unique modeling of specific airports does not suit the rapid simplified models of a generic framework. This research discusses a process to create a set of generic airports by decoupling a subset of U.S. airports into their operational and geometric characteristic to perform grouping. The resulting models are demonstrated to provide an accurate system-level estimation of fleet-level contour area through a series of verification and validation tests. The resulting set of Generic Airports can be used to model the baseline set of airports, as well as to categorize other airports not considered for grouping.

Methodology for Calibration of ANGIM Subjected to Atmospheric Uncertainties

The Airport Noise Grid Integration Method (ANGIM) was developed to enable rapid airport and fleet-level noise contour analysis, but the savings in computational speed over higher-fidelity noise models such as the Aviation Environmental Design Tool (AEDT) are accomplished by requiring a few simplifying assumptions. These assumptions limit the robustness of ANGIM under varying atmospheric and operational conditions. The underlying concept of ANGIM is rooted in the fact that Day-Night Average Level (DNL) noise grids calculated by a detailed noise tool are simply logarithmic additions of all the Sound Exposure Level (SEL) events occurring during a given flight schedule. ANGIM enables rapid calculation of airport-level DNL grids by pre-calculating SEL grids for individual aircraft at standard day sea-level atmospheric conditions and maintaining a database of these grids. To enable calculation of DNL grids in ANGIM under atmospheric uncertainty would require generating and storing grids for every aircraft under every possible atmospheric condition, which would be both time and cost prohibitive. This study outlines a methodology for calibration of standard day sea-level SEL grids for varying atmospheric conditions using neural nets. This calibration method would improve the robustness of ANGIM under varying atmospheric conditions with a small increase in computation time, but this increase in computation time is still much faster than a detailed model and expands the applicability of ANGIM as a screening tool.

Application of Mixture Design of Experiments for Dynamic Fleet-Level Evaluation of Multi-Objective Environmental Technology Trade-offs

Recent and projected increases in aviation demand have caused concerns that the environmental impact of aviation may also increase if not properly mitigated. Such issues have engendered technology identification programs to investigate and invest in technologies to reach a level where industry can implement these technologies. Design methodologies have similarly been under development to provide sophisticated frameworks for analyzing large libraries of technologies and technology combinations. While vehicle-level evaluation is critical in any aircraft design, from an environmental metrics standpoint, fleet-level analyses are much more instructive. Evaluating technology impacts at the fleet-level is typically a matter of scaling vehicle-level metrics by the number of flights. For noise, however, because the metric is the contour area of a fixed Day-Night Average Level (DNL), this vehicle-to-fleet scaling cannot be done. Noise contours must be calculated from a noise grid of noise levels at various observer points. At proper grid refinements, this would be akin to scaling hundreds of thousands of vehicle-level metrics to the fleet-level. This time consuming process hinders the ability to analyze fleet-level noise dynamically in concert with fuel burn and NOx emissions. Of particular interest is the ability to dynamically change technology variants at the vehicle level, to enter the fleet and measure the environmental impact dynamically. Because of the spatial and temporal nature of noise, fleet-level surrogate modeling requires sampling an airport-level model directly and creating surrogates at the airport-level that can then be scaled to the fleet level. The fleet problem is cast as a mixture-of-mixtures-plusprocess-variables problem where the aircraft classes and their variants are the mixture-ofmixture components, and the total flights etc. are the process variables. To analyze this space without reducing model terms, however, requires a caseload in the hundreds of billions. Even a screening test of the space requires too many cases. The terms that can be removed, however, are unknown a priori. The research examines a reduced fleet-level problem at sequentially increasing layers of complexity, characterizing the response space along the way and identifying which factor interactions can be eliminated from the surrogate model. The resulting models are combined with fleet-level fuel burn and NOx emissions models to examine notional technology scenarios. Recommendations regarding the future use of the traditional mixture-modeling framework as applied to the fleet-level noise prediction problem are provided.

Analysis of Vehicle Class Contributions to Total DNL Response

Aviation operations are projected to increase, potentially resulting in increased environmental impacts with respect to fuel burn, NOx emissions, and community noise. A number of programs are involved in identifying technological advances required to mitigate these environmental impacts. These technologies must be analyzed at the vehicle-level, but also at the fleet-level to predict the expected impact in the face of increasing operations. Airport community noise is particularly difficult to model due to the spatial and temporal nature of noise, resulting in a reduced understanding of the fleet-level impacts that are possible as a result of certain aircraft-level changes. The objective of this research is to analyze the contribution to the total DNL response at several airport types. By utilizing a generic framework to intelligently reduce aircraft and airport diversity, contributions of aircraft types at different airport types can be reported. Results included spatial analysis of noise contributions, concluding that the largest contributors affect the lateral regions of a noise contour, while a greater number of vehicle classes impact the noise near the ground track. Results demonstrate that there is some variation in the greatest contributors by airport type. The Large Twin-Aisle and the Very Large Aircraft classes contributions are generally not significant to the airport noise response. Two notional technology infusion scenarios were examined for an approximate time-frame of 2030. The results show that Large Single-Aisle aircraft dominate the total noise response in this forecast year.

A Multi-Stage Surrogate Modeling Approach to Examine Vehicle-Level Technology Impacts at the Airport-Level

Fleet-level analysis of technology scenarios are necessary to examine the system impact of potential technology packages applied at the aircraft level. Typically, fleet analyses require significant amounts of information to perform detailed model runs. This approach makes it difficult to analyze a broad set of technology scenarios because each one requires time consuming modeling. This paper presents a multi-stage surrogate modeling approach capable of examining vehicle-level technology impacts to noise at the airport-level. This process was developed to provide a dynamic dashboard evaluation of vehicle-level technology rolled up to the system-level (a representative airport in this case). The approach defines a subset grid of data that are each treated like an individual noise metric at the aircraft level. Surrogate models are developed to represent aircraft-level noise at these locations as a function of aircraft technology and design characteristics. The subset grid is small enough that aircraft-level noise can be aggregated to the airport level almost instantaneously. Once at the airport-level a separate set of surrogate models utilizes the airport-level noise values to determine the location and size of the Day-Night Average Level (DNL) 65 dB contour. The approach is described in detail, including generation of the surrogate models. Finally several notional scenarios are examined to evaluate the ability of the approach to represent the size and location of the DNL 65 dB contour as a function of aircraft-level settings.