Validation of a crosswind change criterion for building induced airflow disturbances using a flight simulator: case studies at the Hong Kong International Airport
2016; Wiley; Volume: 23; Issue: 4 Linguagem: Inglês
10.1002/met.1598
ISSN1469-8080
Autores Tópico(s)Air Traffic Management and Optimization
ResumoMeteorological ApplicationsVolume 23, Issue 4 p. 742-748 RESEARCH ARTICLEFree Access Validation of a crosswind change criterion for building induced airflow disturbances using a flight simulator: case studies at the Hong Kong International Airport P. W. Chan, Corresponding Author P. W. Chan [email protected] Hong Kong Observatory, Kowloon, Hong KongCorrespondence: P. W. Chan, Hong Kong Observatory, 134A Nathan Road, Kowloon, Hong Kong, China. E-mail: [email protected]Search for more papers by this authorHenk Krus, Henk Krus Cyclone Fluid Dynamics BV, Waalre, The NetherlandsSearch for more papers by this author P. W. Chan, Corresponding Author P. W. Chan [email protected] Hong Kong Observatory, Kowloon, Hong KongCorrespondence: P. W. Chan, Hong Kong Observatory, 134A Nathan Road, Kowloon, Hong Kong, China. E-mail: [email protected]Search for more papers by this authorHenk Krus, Henk Krus Cyclone Fluid Dynamics BV, Waalre, The NetherlandsSearch for more papers by this author First published: 09 December 2016 https://doi.org/10.1002/met.1598Citations: 3AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL ABSTRACT At airports with significant economic growth, more and more buildings are being constructed inside and around the airport premises. These buildings may become a source of low level wind disturbances to the aircraft operating at the airport. In assessing such low level wind effects of buildings, objects and structures, a 7 kn crosswind change criterion can be used. This criterion restricts the wind variations along the glide path and above the runway to a crosswind change of less than 7 kn. It was developed in and based on the experience in the Netherlands over the past 20 years. Therefore, the question arises whether this criterion is applicable to airports and buildings elsewhere in the world, and whether it is consistent with other hazardous wind metrics available in the literature. Based on two cases of pilot reported building induced wind disturbances at Hong Kong International Airport, computer simulated wind fields and a flight simulator were employed to study the validity of the 7 kn criterion. It was found that the criterion is in general consistent with the other hazardous wind metrics available in the literature for the cases under consideration. The number of cases considered in the present study is still limited and further studies into the matter should be considered. 1 Introduction These days it is becoming more common to have a variety of commercial developments inside and around airports. This is particularly the case for an airport which is built on reclaimed land. The commercial developments allow for maximizing the benefits of the reclaimed land available. The buildings and structures are kept to the maximum heights required by airport height restrictions. These are mainly based on the consideration of potential hazards arising from collisions with obstacles (in emergency cases) and non-interference with automatic landing systems. In addition to these two constraints, it is also important to consider the effect of the buildings on the crosswind conditions along the flight paths. Hong Kong International Airport (HKIA) is built on an artificial island and includes buildings inside and outside the airport area, for example terminals, hangars, an exhibition centre, hotels, and other commercial and supporting facilities (Figure 1). Some previous studies of building effects on the wind conditions have been conducted in Hong Kong (Liu et al., 2010; Leung et al., 2012). Pilot reports have been received on the increased level of turbulence along certain flight paths. The turbulence increase is suspected to be associated with the presence of (new) buildings. The buildings include the row of hangars at the western end of the south runway (R1W, i.e. west end of runway 1, the southern runway). Complex interactions of airflow disturbances arise from the row of hangars and the terminal building, exhibition centre, as well as other buildings next to the eastern approach to the north runway (R2E, i.e. east end of runway 2, the northern runway). The locations of the buildings can be found in Figure 1. The buildings become a source of low level wind shear and turbulence to aircraft operating at HKIA, in addition to terrain effects, sea breezes and thunderstorms for example. As a result, for new buildings at the airport a low level wind study has to be performed during the planning stage. Some details about the background and scope of the studies are described in a Hong Kong Observatory pamphlet (2012). Figure 1Open in figure viewerPowerPoint Buildings of Hong Kong International Airport that are considered in the present study. Following the experience in the Netherlands (Nieuwpoort et al., 2010), the wind effect of the buildings at HKIA has been assessed using the 7 kn criterion; with a crosswind of 25 kn to the runway/flight path, the resulting crosswind change along the flight path and over the runway should not exceed 7 kn over a distance of 100 m or longer (at a height lower than a few hundred feet). This criterion has been established by Nieuwpoort et al. (2010) and related studies in the Netherlands using wind tunnels, pilot reports and online and offline flight simulators. In particular, flight simulations by pilots play an important role in this study, and the threshold is based on the impacts on the workload for pilots landing aircraft in situations where there is significant variability in the crosswind. As such, the threshold also takes human factors into account. A recent reference to this criterion can be found in Krus (2016) with a comprehensive discussion of the history of the 7 kn criterion. It is noted that Nieuwpoort et al. (2010) refer to a crosswind change of 6 kn with a background crosswind of 20 kn. Here, the 7 kn criterion of crosswind change is adopted assuming a background crosswind of 25 kn. The two references are generally consistent with each other by linear scaling. It is worthwhile to study whether this criterion is consistent with the reported cases of the buildings' wind effects at HKIA and the hazardous wind metrics available in the literature, such as the F-factor (Hinton, 1993; FAA, 1996), roll angle and descent rate/vertical speed. The F-factor quantifies the effect of wind shear on aircraft performance. It takes into account longitudinal and vertical wind variations along the flight path. It does not take into account crosswinds. In addition to the work by Nieuwpoort et al. (2010), the validity of the 7 kn criterion is assessed in this paper in two case studies of building effects on the airflow along two glide paths at HKIA. The simulated flow field along the glide path is fed into a software (desktop) flight simulator to study the effect on the operation of the aircraft based on the existing hazardous wind metrics in the literature. The result is compared with the assessment based on the criterion of 7 kn crosswind change only. It is demonstrated that, at least for the two case studies at HKIA, the assessment result of the building effects on the airflow is generally consistent with the 7 criterion and the existing hazardous wind metrics in the literature. As such, the use of the 7 kn criterion is justified by these two real cases of building induced turbulence, and is applicable for wind studies at HKIA. To the knowledge of the authors, this is the first time that this approach has been used to validate the applicability of the 7 kn criterion. There is limited experience around the world about the study and about the need to be alert to low level turbulence associated with buildings. A previous study of wind effect by using a short range LiDAR has been made (Chan and Lee, 2012). However, this study was based on the conventional definition of wind shear as headwind change and/or turbulence, which is indistinguishable from other wind shear effects such as terrain-disrupted airflow. 2 Description of the tools used for the study For the study of the wind effects of the built environment, a computational fluid dynamics (CFD) model has been employed to establish how the winds are disturbed as a result of the presence of buildings. The simulated flow field is then fed into a (software) flight simulator to assess the effect of the airflow on the operation of the aircraft. CFD model The CFD model which has been used in the present study is a finite volume method solver employing primitive variables on 3D unstructured polyhedral meshes. The algorithms and procedures are presented in Ferziger and Períc (2002). The model domain is large enough to include all the relevant buildings, parts of the airport island and the surrounding waters. The height of the model is 300 m, ensuring very low blockage factors. Depending on the size of the simulation domain (which is related to the number of buildings to be considered in the simulation) and the wind direction, the number of cells is different in the various simulations. For a large area, such as the terminal buildings, hotels, the exhibition centre and surrounding structures, the number of predominantly hexahedral cells reaches over 6 million. Some special considerations of the meshes are required in the case of non-rectangular buildings, such as the characteristic wavy roof of the first terminal at HKIA. Flight simulator A six degree-of-freedom Boeing B-747 simulator (Robinson, 1990) is employed in the current study. Flying can be either in controls-fixed mode (passive aerodynamic stability) or using automatic control (passive and active stability). The simulation can ingest spatial or temporal wind fields into the aircraft response model. These input wind fields can be from one to four dimensional. They can be employed in the aircraft model either at a single point (the centre of gravity) of the aircraft or using a five-point approximation model to include wind gradients. The input winds can take the form of gridded products (such as outputs from CFD models) or come from other sources such as on-board recorded flight data. This simulation has been established and studied in many different scenarios. The simulation tool is described in detail in the document 'Summary of AeroTech's Aircraft Simulation Environment Toolset' found at http://www.atr-usa.com/documents/Summary_ASET_Description.pdf In the current study, the time averaged crosswind of 3D wind outputs from the CFD model is used in the simulator and applied to the centre of gravity on the aircraft. 3 A review of flight hazard factors The objective hazard criteria for the purposes of this work were determined from inputs provided by aircraft operators, operational manuals and research conducted by for example Nieuwpoort et al. (2010). All criteria described below assume dry runway conditions. F-factor Exceeding a critical F-factor threshold of 0.105 (Hinton, 1993) is considered a hazardous situation. The F-factor has been previously applied to laser radar data to study low level wind shear (Chan, 2012). Vertical speed Operational procedures state that a go-around should be initiated if a descent rate of 1000 ft min−1 (5.05 m s−1) or more is encountered during final approach (below 200 ft). A sink rate of over 360 ft min−1 (6 ft s−1, or 1.82 m s−1) on touchdown can be classified as a 'hard landing' and may require an inspection of the airframe. In the analysis, it is instructive to look at the relative descent rate (RDR). This is the aircraft's actual descent rate with the constant descent rate for a 3° glide slope subtracted. A positive value indicates a reduction in descent rate and a negative value represents an increase. In Table 1 the RDR thresholds are presented (in italics) relative to a nominal 3° glide slope constant rate of descent of 700 ft min−1 (3.54 m s−1). The actual descent rate depends on airspeed, groundspeed and wind conditions. Since there is no flare involved in the simulations the RDR results at landing are not meaningful. The flare is the transition phase between the final approach at an angle and the touchdown on the horizontal landing surface. Table 1. Hazard criteria summary. Criteria Description Threshold F-factor F-factor threshold below 450 ft above ground level 0.105 −1000 ft min−1, −5.05 m s−1 Descent rate below 200 ft above ground level Approach (−300 ft min−1, −1.52 m s−1) Vertical speed (RDR relative to 700 ft min−1, or 3.54 m s−1) −360 ft min−1 Descent rate at touchdown Touchdown (−6 ft s−1, or 1.82 m s−1) Approach See Figure 2 Bank angle Maximum bank angle below 200 ft Touchdown 5° Drift angle Maximum drift angle at touchdown 10° Bank angle (or roll angle) Previous research has defined which roll angle disturbances, as a function of height, are deemed hazardous and they are shown in Figure 2 (Nieuwpoort et al., 2010). The roll angle limits are often found from the aircraft's geometry – gear length, engine pod position and diameter etc. For example, for the Boeing B-773 aircraft on touchdown, a roll angle of greater than 4.8° may lead to an engine nacelle contacting the ground on touchdown. Figure 2Open in figure viewerPowerPoint Bank angle (roll angle) limitations that are given in Nieuwpoort et al. (2010). Drift angle Drift (or crab) angle is defined as the angle between the aircraft heading and the runway heading. A crabbed approach (wings-level) is generally recommended for safe landings of larger aircraft during high crosswinds. The Boeing B-773 autoland system will remove up to 5° of drift. The manufacturer warns that excessive drift (>10°) on touchdown may result in landing gear damage. Operational guidance suggests that in strong crosswinds (greater than 20 kn crosswind component) touchdown should be made with a partial decrab prior to touchdown using a combination of bank angle and drift angle. The maximum bank angle is 5° and the maximum drift angle should be less than 5°. Table 1 summarizes the thresholds defined for the above acceptance criteria. Note that there are also pitch limitations on landing which, if exceeded, may result in a tail strike. A high pitch angle on touchdown is usually caused by the pilot flaring too much. Since the simulations are using the controls-fixed mode and there is no flare, pitch angle is not considered. The criteria in Table 1 may now be applied to the simulation results from Nieuwpoort et al. (2010). 4 Case studies at HKIA Introduction Two cases are studied in the present paper. The first case is a hard landing at the runway threshold R1W in strong northerly winds associated with a tropical cyclone. The northerly winds blow over a row of hangars (Figure 1) and airflow disturbances are expected over the western part of the south runway, which is downwind of the buildings. In the second case, the building complex of terminals, exhibition centre and other buildings are suspected to cause airflow disturbances on approach to the north runway from the east, namely before and over the R2E area (Figure 1). These two cases are well established cases of low level wind effects of buildings at HKIA. Before proceeding to the results, it is important to review some of the characteristics and assumptions of the simulations run. In case 1 the considered wind direction (325°) is basically perpendicular to the runway direction, and therefore only the crosswind component of the CFD wind fields was used in the simulation. Vertical and longitudinal (along the runway centreline) components were not applied. This case precludes the calculation of the F-factor. In case 2, the 3D wind field from the CFD simulation was used. The aircraft was flown with controls fixed. Aircraft attitude, settings and trim were set to the crosswind situation at a given height or position along the glide path. There is no decrabbing or compensation for roll or heading deviations. In evaluating if the wind field caused the aircraft to exceed the hazard criteria thresholds, the onset of the initial disturbance from the trim condition was considered. For case 1, the flights were flown at constant altitude. The effect of descending through the wind field was not captured in the simulator. Hence, the simulator cannot model exactly how an aircraft would respond to a given wind field. However, it can indicate if the wind field may lead to deviations of the critical parameters which could result in a significant destabilization of the approach and landing. For case 2, the Boeing B-747 simulator was flown along the glide slope starting at a height of 900 m. The aircraft is flown in a constant wind field corresponding to the first data point in the CFD simulation file. The aircraft is flown under automatic control until the first point of the wind field is encountered and then the aircraft is flown with controls fixed. The aircraft is trimmed to an airspeed of 80 m s−1 (about 156 kn), with flaps 30° and landing gear down. The benefits of this method are: (1) the aircraft is trimmed and quiescent when it encounters the wind field, and (2) there is no wind disturbance transient when the aircraft encounters the first point of the wind field on approach. The aircraft then goes on to encounter the winds and gradients in the field. Case 1: a row of hangars The meteorological conditions in which the wind effect of the hangars becomes significant to the landing aircraft in a tropical cyclone situation are found in Li and Chan (2012). They are related to the Typhoon Nuri case in August 2008. The typhoon was located to the east of Hong Kong and strong northerly winds were prevailing over the territory, particularly over the area of HKIA. The crosswind at the northern runway was higher than that at the southern runway, and it was decided for the aircraft to land at the southern runway from the west (instead of the northern runway, which is the normal landing runway of HKIA). However, a couple of aircraft were reporting that, upon descent and inside the wake of the hangars, the winds dropped significantly and the aircraft appeared to 'drop out of the sky'. After passing through the wake, the crosswind increased again and the aircraft experienced significant rolling (with a risk of a wing tip or engine nacelle striking the ground). A wind direction of 325° is considered in this case. The winds are basically perpendicular to the runway and they flow over the hangars before reaching the touchdown zone at R1W. The wind speed distribution at a height of 10 m above ground is shown in Figure 3(a); the crosswind profiles at various heights and the approach path are shown in Figure 3(b). Landing with this wind is a tailwind landing of 7–8 kn, slightly higher than the common limit of 5 kn. The 7 kn criterion is violated (with 10.8 kn) and moreover the gradients are rather severe for this particular case. Also at 500 m after the designated touchdown point the 7 kn criterion is violated again. Figure 3Open in figure viewerPowerPoint The wind speed distribution at a height of 10 m (a) and the crosswind profiles (b) for case 1. The touchdown point is 0 m in (b). In terms of aircraft response, Figure 4(a) shows a vertical speed exceedance just after 250 s. At the same time, the roll angle criterion is also exceeded (Figure 4(b)). The drift angles following the roll deviations do not exceed the threshold (Figure 4(c)). The exceedance of the 7 kn criterion is generally consistent with the exceedance of the thresholds of the quantities given in Table 1 and matches the pilot reports of hard landings. Figure 4Open in figure viewerPowerPoint Aircraft response parameters in case 1. The time 245.5 s is the runway threshold and the aircraft is flying at 85 m s−1 (165 kn), i.e. an interval of 5 s corresponds to 425 m. The touchdown point is 258 s. (a) Vertical speed, (b) roll angle magnitude and (c) drift angle magnitude. Case 2: terminal building, exhibition centre and surrounding buildings As reported by the pilots, there can be significant wind shear/turbulence at the lower level (around 300 ft, i.e. around 100 m) when the aircraft lands at the northern runway of HKIA from the east under even moderate south to southwesterly winds (e.g. about 10 kn). Apart from the effect of terrain (mountains on Lantau Island to the south of HKIA), it is suspected that the building complex at the northeastern part of the airport island may bring about low level wind effects on the landing aircraft. Apart from the conventional headwind change (i.e. the definition of significant wind shear) and turbulence, there are other effects on the aircraft occasionally, such as rolling of the aircraft. A number of wind directions have been considered in this analysis, namely 160°, 190°, 205° and 220°. Only the results of wind direction 205° are presented in this paper. The wind speed distribution at a height of 10 m is shown in Figure 5(a) and the crosswind profiles are given in Figure 5(b). It can be seen from Figure 5(b) that the 7 kn criterion has been violated at a number of locations along the glide path on approach to the north runway from the east (i.e. landing at R2E). Figure 5Open in figure viewerPowerPoint Same as Figure 3 but for case 2. The 205° direction is a strong crosswind case with a rather strong headwind component. Figure 6 shows the aircraft response. The roll/yaw oscillation is established by the crosswind variations, or shears, with significant roll variations (−7° to +2°, see Figure 6(a)). The −7° deviation occurs at an altitude of around 40 ft of the glide path and is outside of the 'acceptable' envelope in Figure 2. In reality, however, at that height the aircraft may be initiating decrabbing and applying opposite roll control. Figure 6Open in figure viewerPowerPoint Aircraft response parameters for case 2. The x axis refers to the distance to touchdown. (a) aircraft response, (b) RDR and (c) F-factor. The resulting motion results in a decrease in RDR of −288 ft min−1 (1.45 m s−1) at 1 km (Figure 6(b)) with the aircraft descending below the glide path if not corrected. This would result in a descent rate of 988 ft min−1 (4.99 m s−1, relative to nominal 700 ft min−1, or 3.54 m s−1, descent rate), which is close to vertical speed and RDR thresholds in Table 1. A much higher RDR (−495 ft min−1, or 2.5 m s−1) is seen at landing, but, again, the control system or pilot would be attempting to correct this. However, the shears definitely have a destabilizing effect on the approach. The reduction in headwind within the last 0.5 km results in an increasing F-factor, as seen in Figure 6(c). This headwind decrease can result in an airspeed loss, which can be significant at this phase of flight. For reference, the Federal Aviation Administration's threshold for an in situ wind shear alert is 0.105. The F-factors seen at the end of this case approach this value (0.084). The reduction in headwind (and subsequent F-factor threshold exceedance) is due to the wind speed on the ground plane being zero. Of key importance in this case is the rate at which the speed decays to zero close (less than 50 ft above ground) to the ground. Note that a Boeing B-747 is about 63 ft high to the top of the tail fin. The results of the various wind directions for this case are summarized in Table 2. It can be seen that the various thresholds in Table 1 are largely violated or rather close to violation (e.g. the case of 190°). Again, the criteria in Table 1 are generally consistent with the 7 kn criterion of crosswind change based on this reported case of low level wind effects of buildings. Table 2. Event summary for southerly winds for various wind directions in case 2. Event (wind direction/wind speed at a height of 10 m above ground) Peak values Summary Roll angle RDR F-factor 160°/25 kn −6.2° (790 m distance) Negligible Negligible Roll disturbance 190°/28.86 kn −3.8° (800 m distance) −242 ft min−1, or −1.22 m s−1 0.066 Vertical speed and roll disturbance 205°/35.35 kn −6.9° (244 m distance) −317 ft min−1, or 1.6 m s−1 (954 m distance) 0.084 Strong vertical speed and roll disturbance 220°/50 kn −4° (820 m distance) −454 ft min−1, or −2.29 m s−1 (1.73 km) 0.116 Strong vertical speed disturbance 5 Conclusions This paper provides an initial first study into the validity of the 7 kn crosswind change criterion for assessing low level building induced turbulence based on pilot reported cases at Hong Kong International Airport. The criterion is found to be generally consistent with the hazardous wind metrics available in the literature, as presented in Table 1. As such, there is some evidence that the 7 kn criterion can be used in the planning stage assessment of the effect of new buildings/structures on the airflow along the flight paths. Turbulence was not taken into account in this study of simulator approaches. The aircraft responses are due to changes in the averaged wind conditions only; turbulence will add to the effects. Controls-fixed means that the stability of the aircraft during approach is determined by the aerodynamic design of the aircraft only; if one or more of the criteria are met, adding gusts and deficits to the average wind profiles will definitely raise the cockpit workload. This paper is based on the two presented cases. More cases of low level building induced turbulence should be considered in order to substantiate the 7 kn criterion further. Also, the crosswind gradient is another quantity that is worth considering in assessing the effects of buildings on the conditions during approach and landing (see Krus, 2016). This is currently an active field of research. References Chan PW. 2012. Application of LIDAR-based F-factor in windshear alerting. Meteorol. Z. 21: 193– 204. Chan PW, Lee YF. 2012. Application of short-range Lidar in wind shear alerting. J. Atmos. Oceanic Technol. 29: 207– 220. FAA. 1996. Airborne windshear warning and escape guidance systems for transport airplanes. Technical Standard Order TSO-C117a. Ferziger J, Períc M. 2002. Computational Methods for Fluid Dynamics, 3rd rev. edn. Springer: Berlin. Hinton DA. 1993. Airborne derivation of microburst alerts from ground-based Terminal Doppler Weather Radar Information – a Flight Evaluation. NASA Technical Memo 108990. Hong Kong Observatory. 2012. Low level wind effect at airports – information for pilots and planners of new buildings. http://www.hko.gov.hk/aviat/articles/llw_pamphlet_3col_v17.pdf (accessed 29 September 2016). Krus H. 2016. Criteria for crosswind variations during approach and touchdown at airports. In Advances in Simulation of Wing and Nacelle Stall, Notes on Numerical Fluid Mechanics and Multidisciplinary Design, R Radespiel, R Niehuis, N Kroll, K Behrends. (eds), Vol. 131. Springer; 167– 187. Leung DYC, Lo WY, Chow WY, Chan PW. 2012. Effect of terrain and building structures on the airflow in an airport. J. Zhejiang Univ. Sci. A 13: 461– 468. Li L, Chan PW. 2012. Numerical simulation study of the effect of buildings and complex terrain on the low-level winds at an airport in typhoon situation. Meteorol. Z. 21: 183– 192. Liu CH, Man ACS, Leung DYC, Chan PW. 2010. Computational fluid dynamic simulation of the wind flow over an airport terminal building. Zhejiang Univ. Sci. A 11(6): 389– 401. Nieuwpoort AMH, Gooden JHM, de Prins JL. 2010. Wind criteria due to obstacles at and around airports. NLR-TP-2010-312. Robinson PA. 1990. The modelling of turbulence and downbursts for flight simulators. Report 339, University of Toronto Institute for Aerospace Studies. Citing Literature Volume23, Issue4October 2016Pages 742-748 FiguresReferencesRelatedInformation
Referência(s)