曝光台 注意防骗
网曝天猫店富美金盛家居专营店坑蒙拐骗欺诈消费者
purpose of the simulations is to estimate the probability of a TCV as a result of one
aircraft turning unexpectedly from an ILS localizer or en route track toward another
SEPARATION SAFETY MODELING
5-24
aircraft on an adjacent ILS approach or en route track. The simulated blunder is assumed
to be the result of a problem that the crew is unable to resolve regardless of instructions
from air traffic control (ATC). The blundering aircraft is assumed to turn to a heading 30
degrees from the heading of the aircraft prior to the blunder. This type of blunder is called
a worst-case blunder. The aircraft on the parallel approach, called the evader, is assumed
to continue the ILS approach until contacted by ATC.
Before the simulation of the blunder can begin, the two aircraft must be placed on their
respective ILS approaches. The range of the blundering aircraft from threshold is first
determined. The range can be chosen randomly in a predetermined interval or it can be set
to a desired constant such as 3 nm. Then the lateral and vertical deviations of the
blundering aircraft from the glide path are chosen at random from the CRM distributions.
The CRM distributions have thicker tails than a Normal distribution and have standard
deviations that grow larger as the range grows larger. The speed of the blundering aircraft
is also chosen randomly at this time.
The lateral and vertical position of the evading aircraft, as well as its speed, are then
chosen at random. Then the range of the evading aircraft is chosen. The range may be
chosen at random, or the range may be chosen so that the simulated blunder is at-risk.
The blunder is initiated by rolling the blundering aircraft into a standard rate of turn
toward the evading aircraft. Since actual flight dynamics are used in the ASAT fast-time
simulation, the flight path will differ from the flight paths of the blundering aircraft in the
real-time simulation. Pilot reaction time is measured in the real-time simulation from the
time that the controller is alerted by the PRM that the blundering aircraft will enter the
Non-Transgression Zone (NTZ) in ten seconds. Radar error is applied to the positions of
the two aircraft during the maneuver to produce a random effect on the NTZ alert time.
ATC response time is chosen randomly from its distribution and added to the NTZ alert
time to determine the time ATC begins its transmission to the evading aircraft. Pilot
response time is randomly selected from its distribution and added to the cumulative time
to find the time at which the pilot begins his or her evasion maneuver. Then the
distributions of roll rate, rate of climb, maximum bank angle, and heading change are used
to find random values for the evasion maneuver.
The slant distance between the two aircraft is computed twenty times per simulated
second from the start of the blunder. The blunder is stopped when it is determined that
the Closest Point of Approach (CPA) has occurred. The value of the CPA is added to the
set of CPA data and if a TCV occurred it is also added to the file of TCV data. The
model then begins the process again by selecting positions of the aircraft for the next
blunder.
APPROACHES TO COLLISION RISK ANALYSIS
5-25
Simulation Implementation
Before a simulation can begin, the ASAT model must be configured appropriately. For
example, in the MPAP simulations, the runways and approaches in the ASAT system are
set to match those in the real-time situation. In these simulations, the distance between
runways is set and runway threshold stagger is also set. Offset approaches can also be
modeled. Other parameters that are adjusted prior to simulation are field elevation,
altimeter setting, wind velocity, and glide slope angle. Parameters such as visibility, and
ceiling are not pertinent to the Monte Carlo simulation.
Empirical distributions of ATC response times and pilot response times are obtained from
the real-time simulation performed at the FAA Technical Center. Similarly, empirical
distributions of aircraft performance data, indicated airspeeds, roll rates, maximum bank
angles, and rates of climb are derived from data acquired during a real-time simulation.
Special software tools have been developed by ASF-450 for the efficient collection of
aircraft data from the flight track data recorded during the real-time simulation. Since
several different aircraft simulators are used in the real-time simulation, differences in
aircraft characteristics such as roll rates, indicated airspeeds, and rates of climb are
expected. The empirical data is statistically tested for similarities and the data is pooled
into larger sets whenever possible. Mathematical probability curves are fit to the empirical
中国航空网 www.aero.cn
航空翻译 www.aviation.cn
本文链接地址：
a concept paper for separation safety modeling(33)
