National Academies Press: OpenBook

Integrated Noise Model Accuracy for General Aviation Aircraft (2014)

Chapter: 7 Detailed Analysis Causes of Error: Test Examples

« Previous: 6 Detailed Analysis Select Analysis Aircraft
Page 35
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 35
Page 36
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 36
Page 37
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 37
Page 38
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 38
Page 39
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 39
Page 40
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 40
Page 41
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 41
Page 42
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 42
Page 43
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 43
Page 44
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 44
Page 45
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 45
Page 46
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 46
Page 47
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 47
Page 48
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 48
Page 49
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 49
Page 50
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 50
Page 51
Suggested Citation:"7 Detailed Analysis Causes of Error: Test Examples." National Academies of Sciences, Engineering, and Medicine. 2014. Integrated Noise Model Accuracy for General Aviation Aircraft. Washington, DC: The National Academies Press. doi: 10.17226/22269.
×
Page 51

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

7 Detailed Analysis–Causes of Error: Test Examples As suggested by the Panel, we focused our analysis more on specific details of the aircraft flight profiles in the INM than on the internal workings of the INM. However, we did a review of INM equations and coefficients used for modeling the aircraft. We found that, despite the complexity of some of the equations, most aspects of aircraft performance were weak functions of all but basic assumptions about the fundamental variables of aircraft weight, engine static thrust, and the ratio of drag over lift. We assume that these variables are well known, and should be correct as included in the INM. Consequently, we examined only performance profiles to help identify causes of error. We chose three departure aircraft to explore possible causes of the discrepancies in sound level and altitude. We also provide detailed information about three selected arrival aircraft for future reference and consideration. We emphasize that in what follows our efforts focused on departures for several reasons. Departures tend to have the greater influence on DNL contours, and are, or should be, more amenable to detailed examination than are arrivals. Departures tend to follow a somewhat standard sequence of application of maximum (or close to maximum) thrust for start of roll through lift off and initial climb, to decelerated climb in favor of speed increase, which may include flap retraction, then change of thrust to climb out. Arrivals, on the other hand, can be highly variable, largely for reasons of air traffic control and maintenance of separation. Additionally, as shown by Figure 6, we expect considerably less improvement in arrival sound energy computation through correction of arrivals than we expect through correction of departures. Arrivals, in fact, are relatively well modeled by the INM. The modeled arrival profiles and actual arrival profiles do not differ by much, in terms of decibels, Figure 6. We judge that reducing what arrival discrepancies exist to be of low priority. The aircraft departures we have chosen to examine are those for the LJ35, the GLF4, and the EA50. The first two because they are two of the selected aircraft, Table 7, and are not modeled with substitution aircraft types. The EA50 is examined because we have on our team J R Engineering (JRE), the firm that developed the noise and performance profiles for the EA50.25 Should we want to better understand specifics of how profiles are developed, we have access to their engineers. The result of this analysis, as will be shown, is that it is likely most error in the INM modeling is caused by significant differences between the standard noise and performance profiles (management of thrust, flaps, speed, climb rates and associated noise-power-distance curves) and actual average practice. Additionally, the noise-power-distance curves (the built in relationships between power and noise as a function of distance from the aircraft) may be incorrect for some aircraft. In the following presentations, we first provide for each selected aircraft the counts of data points used (valid matches of measured event SEL with flight track) and then reiterate the measured and modeled SEL values and the discrepancies. Three graphics then present the measured and modeled SEL, the measured minus modeled SEL differences, and the measured and modeled altitudes. Departures 7.1 For each of the three aircraft departures, a final graphic presents all radar returns of the departure altitudes, overlaid with the standard INM altitude profile, and a “best fit” altitude profile based on a new performance profile we developed. For comparison, a polynomial fit to the radar returns is also graphed. For all three cases, the best fit is a result of iterations in the use of thrust, climb rate, speed and flaps to yield the measured altitude at the 12,000 foot track distance and to have the aircraft using the most 25 JRE has been involved in over 30 flight test programs ranging from very large jets (Lockheed C-5) to the smallest jets (Eclipse 500) to single engine GA aircraft. 34

reasonable thrust at that point that produces the best approximation to the desired SEL. The resultant “best fit” discrepancies of SEL and altitude are given. 7.1.1 LJ35 This aircraft is important not only because it is a substitute aircraft for four of the selected nine aircraft, Table 7, but because it is responsible for the largest discrepancies of all the selected departure aircraft. Table 9 and Table 10 provide basic information about the data used and the measured / modeled comparisons. Figure 7 shows the complete database of measured and modeled SEL as a function of track distance from runway takeoff end. To demonstrate the trends of the data, we plot a linear least-squares fit line.26 Figure 7 and Figure 8 suggest that the discrepancy decreases with increasing track distance. Figure 9 shows clearly that the modeled altitude profile is higher, at least initially, than most actual altitudes. Table 9 LJ35 sample size of database by airport Number of Tracks w/ SEL Airport A D BED 7 60 BWI 10 59 DEN 3 0 FXE 879 518 HPN 283 35 VNY 1191 1059 TOTAL 2373 1731 Table 10 LJ35 measured and modeled departure results summarized Designator INM Aircraft Type Measured SEL (dB) Modeled SEL (dB) Measured minus Modeled (dB) Effective Altitude Discrepancy (dB) Discrepancy if Altitude Corrected (dB) LJ35 LEAR35 86.6 94.1 -7.5 -2.5 -10.1 26 The actual change of SEL with track distance would unlikely be linear, but the linear relationship was used to provide a simple comparison of the trend with distance. 35

Figure 7 LJ35 measured and modeled departure SEL Figure 8 LJ35 measured minus modeled departure SEL 36

Figure 9 LJ35 measured and modeled departure altitude profiles Figure 10 compares the standard INM LJ35 departure profile, all the radar returns, the best fit profile and a polynomial fit. The red lines identify the 12,000 foot track distance used for all departure discrepancy calculations and the average radar altitude at that track distance. 27 The measured minus modeled SEL discrepancy with the best fit profile is now -5.2 rather than -7.5. The modeled altitude is 5 feet higher than the measured altitude. It is noteworthy that the SEL discrepancy could not be made smaller while keeping the weight, thrust and speed within reasonable performance bounds. Consequently, we conclude that the NPDs (noise-power-distance curves) associated with this aircraft are incorrect and too loud as a function of thrust. This error, assuming our analysis is correct, can go a long way to explaining why the modeled GA jets are so generally louder than the measured levels since the LJ35 is a substitution for four other aircraft and with similar SEL discrepancies, see Table 7. 27 Details of the parameters of the best-fit profile are presented in Appendix E page 73. 37

Figure 10 LJ35 departure altitudes: all radar data, INM, best fit, and polynomial fit 7.1.2 GLF4 The GLF4 was chosen for detailed analysis because its discrepancy is positive – the INM modeled SEL is lower than the measured SEL, while its altitude is higher than the measured altitudes. Hence, correcting the altitude may reduce the discrepancy. Table 11 and Table 12 provide basic information about the data used and the measured / modeled comparisons. Figure 11 shows the complete database of measured and modeled SEL as a function of track distance from runway takeoff end. Figure 11 and Figure 12 show that the discrepancy decreases with track distance. Figure 13 shows clearly that the modeled altitude profile is higher, and increasingly so at greater track distances. Table 11 GLF4 sample size of database by airport Number of Tracks w/ SEL Airport A D BED 98 319 BWI 95 23 DEN 5 0 FXE 87 75 HPN 2355 228 VNY 2978 3159 TOTAL 5618 3804 38

Table 12 GLF4 measured and modeled departure results summarized Designator INM Aircraft Type Measured SEL (dB) Modeled SEL (dB) Measured minus Modeled (dB) Effective Altitude Discrepancy (dB) Discrepancy if Altitude Corrected (dB) GLF4 GIV 87.4 83.5 3.9 -3.5 0.4 Figure 11 GLF4 measured and modeled departure SEL 39

Figure 12 GLF4 measured minus modeled departure SEL Figure 13 GLF4 measured and modeled departure altitude profiles 40

Figure 14 compares the standard INM GLF4 departure profile, all the radar returns, the best fit profile and a polynomial fit.28 The measured minus modeled SEL value with the best fit profile is now -0.2 rather than +3.9. The modeled altitude is 16 feet lower than the measured altitude. Figure 14 GLF4 departure altitudes: all radar data, INM, best fit, and polynomial fit 7.1.3 EA50 The EA50 was chosen because it is a recently added type and, as mentioned, the firm that conducted the certification tests and developed the performance information, JRE, is on our team. After Panel review of this report, we will discuss with JRE their understanding of the standard profiles that are in the INM, and their role in developing the Eclipse profiles and NPD curves. Table 13 and Table 14 provide basic information about the data used and the measured / modeled comparisons. Figure 15 shows the complete database of measured and modeled SEL as a function of track distance from runway takeoff end. Figure 15 and Figure 16 show that the discrepancy decreases slightly with track distance. Figure 17 shows clearly that the modeled altitude profile is higher, and increasingly so at greater track distances. 28 Details of the parameters of the best-fit profile are presented in Appendix F, page 74. 41

Table 13 EA50 sample size of database by airport Number of Tracks w/ SEL Airport A D BED 3 28 BWI 0 0 DEN 1 0 FXE 4 0 HPN 80 11 VNY 116 160 TOTAL 204 199 Table 14 EA50 measured and modeled departure results summarized Designator INM Aircraft Type Measured SEL (dB) Modeled SEL (dB) Measured minus Modeled (dB) Effective Altitude Discrepancy (dB) Discrepancy if Altitude Corrected (dB) EA50 ECLIPSE500 80.0 76.3 3.7 -3.9 -0.2 Figure 15 EA50 measured and modeled SEL 42

Figure 16 EA50 measured minus modeled departure SEL Figure 17 EA50 measured and modeled departure altitude profiles 43

Figure 18 provides the comparison of the standard INM EA50 departure profile with all the radar returns, the best fit profile and a polynomial fit.29 The measured minus modeled SEL value with the best fit profile is now +0.4 rather than +3.7. The modeled altitude is 7 feet higher than the measured altitude. Figure 18 EA50 departure altitudes: all radar data, INM, best fit, and polynomial fit Arrivals 7.2 The same basic information and graphs provided above for departures in Section 7.1 are provided below for the selected arrival aircraft of C56X, FLG5 and F900. These are chosen because of the five arrival aircraft, they are the three that do not use substitution aircraft. As discussed above, alternative arrival profiles have not been developed. Examination of the data on the following three aircraft suggests several generalizations. First, SEL discrepancies tend to decrease with increasing track distance. Second, measured SEL values and hence, discrepancies are widely scattered – somewhat more so than is the case with the departure SELs. Finally, measured arrival altitudes tend to be close to the modeled altitudes at the shorter track distances, higher than the modeled altitudes at the middle distances, and lower than modeled at the furthest track distances. All this points to the complexity of arrivals and the likelihood that best fit arrival profiles will need to be a complex sequence of changes in thrust, descent rate and speed changes. 29 Details of the parameters of the best-fit profile are presented in Appendix G, page 75. 44

7.2.1 C56X Table 15 C56X sample size of database by airport Number of Tracks w/ SEL Airport A D BED 69 290 BWI 49 69 DEN 14 0 FXE 87 39 HPN 2388 64 VNY 853 910 TOTAL 3460 1372 Table 16 C56X measured and modeled arrival results summarized Designator INM Aircraft Measured SEL (dB) Modeled SEL (dB) Measured minus Modeled (dB) Effective Altitude Discrepancy (dB) Discrepancy if Altitude Corrected (dB) C56X CNA560XL 85.3 87.8 -2.6 1.01 -1.54 Figure 19 C56X measured and modeled arrival SEL 45

Figure 20 D56X measured minus modeled arrival SEL Figure 21 C56X measured and modeled arrival altitude profiles 46

7.2.2 GLF5 Table 17 GLF5 sample size of database by airport Number of Tracks w/ SEL Airport A D BED 23 85 BWI 25 9 DEN 3 0 FXE 35 29 HPN 1054 86 VNY 1107 1138 TOTAL 2247 1347 Table 18 GLF5 measured and modeled arrival results summarized Designator INM Aircraft Measured SEL (dB) Modeled SEL (dB) Measured minus Modeled (dB) Effective Altitude Discrepancy (dB) Discrepancy if Altitude Corrected (dB) GLF5 GV 83.0 85.2 -2.2 0.90 -1.28 Figure 22 GLF5 measured and modeled arrival SEL 47

Figure 23 GLF5 measured minus modeled arrival SEL Figure 24 GLF5 measured and modeled arrival altitude profiles 48

7.2.3 F900 Table 19 F900 sample size of database by airport Number of Tracks w/ SEL Airport A D BED 0 0 BWI 24 10 DEN 0 0 FXE 0 0 HPN 0 0 VNY 325 340 TOTAL 349 350 Table 20 F900 measured and modeled arrival results summarized Designator INM Aircraft Measured SEL (dB) Modeled SEL (dB) Measured minus Modeled (dB) Effective Altitude Discrepancy (dB) Discrepancy if Altitude Corrected (dB) F900 F10062 83.9 88.1 -4.3 1.51 -2.78 Figure 25 F900 measured and modeled arrival SEL 49

Figure 26 F900 measured minus modeled arrival SEL Figure 27 F900 measured and modeled arrival altitude profiles 50

Next: 8 Potential Solutions »
Integrated Noise Model Accuracy for General Aviation Aircraft Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s Airport Cooperative Research Program (ACRP) Web-Only Document 19: Integrated Noise Model Accuracy for General Aviation Aircraft assesses the predictive accuracy of the Integrated Noise Model, identifies causes for deviations between actual and predicted values, identifies potential solutions to improve the model’s accuracy, and describes the steps needed for implementation

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!