Flight Data Recorder Research Articles

Crack Propagation Tests for Load Sequences Developed Using Different Flight Parameters of a Trainer Aircraft

Abstract Military aircraft are subjected to highly variable and unpredictable loads due to diverse mission profiles, armament configurations, and individual piloting styles. This variability complicates the definition of precise load spectra, particularly in cases where data loss occurs due to Flight Data Recorder (FDR) malfunctions or data mishandling. This paper investigates the use of different flight parameters, such as load factor (nz ), barometric height (Hb ), and horizontal velocity (Vp ), to define load sequences for the PZL-130 “Orlik” TC-II military trainer aircraft. These sequences were then used to evaluate crack propagation using Compact Tension (CT) specimens. The results show that the incorporation of additional flight parameters improves the accuracy of crack propagation predictions when compared to direct strain measurements. This study highlights the potential of using available flight data to develop reliable load spectra for fatigue life estimation in military aircraft, even when direct load measurements are not financially feasible.

Improved modelling of low-pressure rotor speed in commercial turbofan engines: A comprehensive analysis of machine learning approaches

Accurate engine thrust modelling offers notable opportunities for reducing fuel consumption, mitigating emission effects, managing air traffic, and optimizing the preliminary design of aircraft. This study focuses on the development of machine learning models to predict the low-pressure rotor speed (N1) parameter, which is closely related to the thrust of turbofan engines, for commercial aircraft during the cruise phase. The analysis utilizes a Flight Data Recorder (FDR) dataset from 1086 actual flights involving twin-engine, narrow-body, short-to-medium haul commercial aircraft equipped with CFM56-7B turbofan engines. To achieve accurate N1 parameter predictions, the study employs machine learning models, including Deep Learning (DL), Random Forest (RF), Gradient Boosted Machines (GBM), and Artificial Neural Networks (ANN). These models are evaluated using statistically significant indicators, demonstrating that machine learning models yield highly accurate results. Among the models tested, the DL model provides the most precise estimates of the N1 parameter. This study not only advances the understanding of engine thrust modelling through machine learning but also provides practical insights for the aerospace industry. The findings underline the potential of machine learning techniques in delivering superior prediction accuracy, which can be integrated into real-time flight management systems, ultimately contributing to more sustainable and efficient aviation operations.

Flight Data Recorders: Past, Present, and Future

Flight Data Recorders: Past, Present, and Future

Automatic recognition of airliners wake turbulence using various techniques of machine intelligence

Airplanes generate severe turbulent airflow, known as trailing vortices, which can have an adverse control impact on following airplanes. Aviation authorities worldwide impose separation standards between leading and following airplanes to reduce the capacity of airports to handle large volumes of traffic. Optimizing these separations is highly sought after. These separations are optimized based on several factors such as airplane size, weight, and design in addition to weather conditions. The recognition of trailing vortices is usually done manually through examination of pilot reports and analysis of flight data recorders (FDRs). This process can be subjective and depends greatly on the analyst's experience. Automated processes based on Artificial Intelligence (AI), such as Machine Learning (ML) and Computational Intelligence (CI), have shown a potential to better deal with such problems. Results obtained from these various techniques are presented to show their degree of appropriateness. Flight event identification, such as encounters with airplane vortices, wind shear, turbulence, and uncommanded control, can help establish aircraft warning and monitoring systems to enhance automatic flying, especially during landing and take-off. The developed model can attain great assurance recognition of the aircraft wake turbulence by supporting the air traffic controllers with the decision information in actual work to ensure aircraft safety. The success rates obtained from the investigated CI models compare favorably well to results obtained by either manual and/or other CI-based techniques when applied to vortex identification.

Assessing the impact of climate change on aircraft engine performance during different flight phases (take-off and cruise) and the environmental consequences

PurposeClimate change significantly impacts global temperatures, posing challenges to various sectors, including aviation. The purpose of this study is to assess the impact of climate change on aircraft engine performance during different flight phases (take-off and cruise) and the environmental consequences.Design/methodology/approachThis study examines the effects of rising temperatures on aircraft engine performance using real-time data from a Boeing 787-8 equipped with GEnx-1B engines, which are collected via Flight Data Recorder of the engines and were analyzed for the take-off and cruise phases on the ground. Exhaust gas temperature (EGT), fuel flow and take-off weights were evaluated.FindingsThe analysis revealed a significant increase in EGT at the cruising altitude of 38,000 ft during the summer months compared to expected standard atmospheric values. This increase, averaging over 200 °C, is attributed to global warming. Such elevated temperatures are likely to accelerate the degradation of turbine components, resulting in increased fuel consumption: higher EGT signifies inefficient engine operation, resulting in more fuel burned per unit thrust; early engine aging: elevated temperatures accelerate wear and tear on turbine components, potentially reducing engine lifespan and increasing maintenance costs and enhanced atmospheric pollution: incomplete combustion at high EGTs generates additional emissions, contributing to local air quality concerns.Practical implicationsThe research findings have practical implications for understanding the potential operational challenges and environmental impacts of climate change on aircraft engine performance. This lets us explore mitigation strategies and adapt operational procedures to ensure sustainable regional aviation practices.Originality/valueThis research enhances environmental consequences by assessing the impact of climate change on aircraft performance.

Power Setting Estimation of Departing Civil Jet Aircraft Based on Machine Learning

This paper presents an approach for estimating the power setting, indicated as N1, the rotational speed of the low-pressure shaft, of departing civil jet aircraft to be used as input for accurate noise calculations. The method utilizes the machine learning approach random forest regression and is trained using flight data recorder data divided into three departure sections. Each section uses a unique set of three to four model features based on easily accessible data, such as position and airport meteorological data. Unlike previous methods, neither fixed configuration altitudes, aerodynamic coefficients, or engine coefficients, nor takeoff mass or wind information are required as inputs. To assess the performance of the method, noise calculations for departures of five aircraft types, including narrow- and wide-body jet aircraft, were performed. Comparing overall and aircraft-specific noise levels, calculated from estimated N1 versus recorded N1 values from flight data recorder data, revealed small differences of less than 1 dB. Overall, the present machine-learning-based approach provides reliable estimates of power settings for departing civil jet aircraft, offering improved accuracy and avoiding the need for numerous assumptions and hardly accessible data.

Testing the Thermal Properties of the Insulating Structures of A Flight Data Recorder

Testing the Thermal Properties of the Insulating Structures of A Flight Data Recorder

Fuel Burn Method Assessment Using Automatic Dependent Surveillance–Broadcast and European Reanalysis Data: Limited Flight Sample Analysis

Fuel burn during the actual route flown is an important indicator of aircraft operational efficiency. This study aims to assess and systematically evaluate the method for fuel consumed during flights using data from the automatic dependent surveillance–broadcast (ADS-B), European reanalysis (ERA5) meteorological dataset, and BADA 3 performance. A literature background and comprehensive methodology are provided for fuel estimation using track data. The airborne part of the trajectory was used to estimate the total trip fuel consumed during several flights of a commercial airliner. The calculated fuel burn is compared with measured fuel consumption from the flight data recorder (FDR). The results show that fuel consumption for the entire airborne part of the trajectory can be estimated with an average error of 1.2% and with a standard deviation of 1.3%. Detailed results of fuel burn for individual flight phases, from the initial climb to the approach, are also presented. In addition, this paper also discusses the sources of errors and the potential applications of the method for network operations and environmental monitoring.

Analysis of Stochastic Properties of MEMS Accelerometers and Gyroscopes Used in the Miniature Flight Data Recorder

Analysis of Stochastic Properties of MEMS Accelerometers and Gyroscopes Used in the Miniature Flight Data Recorder

Multi-modal LSTM network for anomaly prediction in piston engine aircraft

An aircraft is a highly intricate system that features numerous subsystems, assemblies, and individual components for which regular maintenance is inevitable. The operational efficiency of an aircraft can be maximised, and its maintenance needs can be reduced using an effective yet automatic AI-based health monitoring systems which are more efficient as compared to designing and constructing expensive and harder to operate engine testbeds. It has been observed that aircraft engine anomalies such as undergoing flameouts can occur due to the rapid change in the temperature of the engine. Engine oil temperature and cylinder head temperature, two measures connected to this issue, might be affected differently depending on flight modes and operational conditions which in turn hamper AI-based algorithms to yield accurate prediction on engine failures. In general, previous studies lack comprehensive analysis on anomaly prediction in piston engine aircraft using modern machine learning solutions. Furthermore, abrupt variation in aircraft sensors' data and noise result in either overfitting or unfavourable performance by such techniques. This work aims at studying conventional machine learning and deep learning models to foretell the possibility of engine flameout using engine oil and cylinder head temperatures of a widely used Textron Lycoming IO-540 six-cylinder piston engine. This is achieved through pre-processing the data extracted from the aircraft's real-time flight data recorder followed by prediction using specially designed multi-modal regularised Long Short-Term Memory network to enhance generalisation and avoid overfitting on highly variable data. The proposed architecture yields improved results with root mean square error of 0.55 and 3.20 on cylinder head and engine oil temperatures respectively averaged over three case studies of five different flights. These scores are significantly better i.e., up to 84% as compared to other popular machine learning predictive approaches including Random Forest, Decision Tree Regression, Artificial Neural Networks and vanilla Long Short-Term Memory networks. Through performance evaluation, it can be established that the proposed system is capable of predicting engine flameout 2 minutes ahead and is suitable for integration with the software system of aircraft's engine control unit.

Hidden Markov Models and Flight Phase Identification

The use of Hidden Markov Models (HMMs) in segmenting flight phases is a compelling approach with significant implications for aviation and aerospace research. It leverages the temporal sequences of flight data to delineate various phases of an aircraft’s journey, making it a valuable tool for enhancing the anal- ysis of flight performance and safety. In this work, we implement a multivariate HMM to identify 6 flight phases: taxi, takeoff, climb, cruise, approach and rollout. We reach a median global accuracy of about 97% over a sample of several thousand flights with a very low number of decoded unlikely transitions. Regarding several performance metrics, our method is competitive with existing methods in the litera- ture, such as fuzzy logic. Additionally, it provides, for each point of the flight, a probability of belonging to each phase. Even in situations where there are missing values in the data, HMMs remain effective, ensuring that no critical information is lost during the segmentation process. We show that HMMs work seamlessly with the fine granularity of Flight Data Recorder (FDR) data. HMMs offer remarkable flex- ibility and adaptability, proving particularly effective when the number or order of phases is unknown or not predetermined, as is often the case with complex flight scenarios such as helicopter flights. This adaptability is crucial for handling the diverse range of flight operations that differ from one aircraft to another. An example is given with the segmentation of an Automatic Dependent Surveillance–Broadcast (ADS-B) helicopter flight operated by the Swedish National Police.

Installation of the flight data recorder S2-3a on Su-22 and MiG-29 aircraft

The S2-3a flight data recording system developed at the Air Force Institute of Technology is intended for recording the flight parameters and the operating parameters of aircraft assemblies, as well as to store the recorded data in its memory to evaluate flight safety, piloting technique, technical condition of on-board system and air accident (air crash) causes. S2-3a flight data recording systems are operated on-board the: TS-11 ISKRA, PZL-130 TC-II ORLIK, M-28 BRYZA, MiG-29, and Su-22 aircraft, as well as the Mi-8, Mi-14, Mi-17, Mi-24, W-3 SOKÓŁ and SW-4 helicopters. The paper contains selected information on the instllation of the S2-3a recorder on Su-22 and MiG-29 aircraft AT the Military Aviation Works no. 2 S.A. In Bydgoszcz.

Demonstration of Fixed Flight-Path Angle Descent via Scheduled Commercial Flights

Toward implementation of fuel-saving descent operations, this study demonstrated a fixed flight-path angle descent using regularly scheduled commercial flights arriving at Kansai International Airport. This paper summarizes the authors' industry/academia collaboration research including 1) operational bottlenecks and weather analysis, 2) proposals of pilot and air traffic controllers’ operational procedures, 3) validation experiments using an A320 flight simulator, and 4) flight demonstrations conducted across 24 days using A320neo and A320ceo aircraft. The flight demonstration results were evaluated comparing flight recorder data of the new descent procedure relative to current descent operations. A total of 92% of the flights were approved to fly new approach procedure under actual air traffic operations, with an approximate average of 150 lb fuel reduction per flight (230 lb maximum).

An aviation accident report data-driven approach to scenario design for a centrifuge-based dynamic flight simulator

The use of flight simulators in investigating an aviation incident or accident related to human errors has been identified as an important part of a strategy to improve safety. This study aimed to replicate a real flight of the MiG-29 aircraft using a centrifuge-based dynamic flight simulator and to determine the simulator’s accuracy in recreating in-flight aircraft performance. A 60-second recording of the real flight of the MiG-29 aircraft, captured by the flight data recorder, was chosen for replication in the HTC-07 human training centrifuge simulator. To evaluate how accurately the simulator replicates the performance of the aircraft, the linear accelerations and angular velocities acting on a pilot during the real flight were compared with those during the replication of that flight in the simulator. The fit of these parameters was assessed using the root mean square percentage error (RMSPE) and the correlation coefficient (r). The highest replication accuracy was achieved for the vertical component of the linear acceleration (RMSPE=2068; r=0.98), while the worst result was obtained for the longitudinal component (RMSPE=14205; r=0.31). Inaccuracies were much more pronounced for the angular velocity. The roll angular velocity had the lowest replication error (RMSPE=12640). However, its correlation with the recorded velocity during the real flight was very weak (r=-0.02). Despite some inaccuracies in replicating other components of the acceleration and angular velocity vectors, the HTC-07 simulator seems valuable for investigating aviation incidents or accidents related to human factors.

The Optimization of a Model for Predicting the Remaining Useful Life and Fault Diagnosis of Landing Gear

With the development of next-generation airplanes, the complexity of equipment has increased rapidly, and traditional maintenance solutions have become cost-intensive and time-consuming. Therefore, the main objective of this study is to adopt predictive maintenance techniques in daily maintenance in order to reduce manpower, time, and the cost of maintenance, as well as increase aircraft availability. The landing gear system is an important component of an aircraft. Wear and tear on the parts of the landing gear may result in oscillations during take-off and landing rolling and even affect the safety of the fuselage in severe cases. This study acquires vibration signals from the flight data recorder and uses prognostic and health management technology to evaluate the health indicators (HI) of the landing gear. The HI is used to monitor the health status and predict the remaining useful life (RUL). The RUL prediction model is optimized through hyperparameter optimization and using the random search algorithm. Using the RUL prediction model, the health status of the landing gear can be monitored, and adaptive maintenance can be carried out. After the optimization of the RUL prediction model, the root-mean-square errors of the three RUL prediction models, that is, the autoregressive model, Gaussian process regression, and the autoregressive integrated moving average, decreased by 45.69%, 55.18%, and 1.34%, respectively. In addition, the XGBoost algorithm is applied to simultaneously output multiple fault types. This model provides a more realistic representation of the actual conditions under which an aircraft might exhibit multiple faults. With an optimal fault diagnosis model, when an anomaly is detected in the landing gear, the faulty part can be quickly diagnosed, thus enabling faster and more adaptive maintenance. The optimized multi-fault diagnosis model proposed in this study achieves average accuracy, a precision rate, a recall rate, and an F1 score of more than 96.8% for twenty types of faults.

Severity assessment of sudden plunging motion for jet transport aircraft in response to severe atmospheric turbulence

PurposeThe purpose of this paper is to present a comparative study of flight circumstances, dynamic stability characteristics and controllability for two transport aircraft in severe atmospheric turbulence at transonic cruise flight for the purpose to obtain the prevention concepts of injuries to passengers and crew members for pilot training in International Air Transport Association (IATA) – Loss of Control In-flight (LOC-I) program.Design/methodology/approachA twin-jet and a four-jet transport aircraft encountering severe atmospheric turbulence are the study cases for this paper. The nonlinear unsteady aerodynamic models are established through flight data mining and the fuzzy-logic modeling technique based on the flight data of flight data recorder. This method can be adopted to examine the influence of horizontal wind shear and crosswind on loss of control, dynamic stability characteristics and controllability for transport aircraft in different weights and different sizes in tracking aviation safety of existing different types of aircraft.FindingsThe horizontal wind shear or crosswind before the turbulence encounter will easily induce rolling motion and then initiate the sudden plunging motion during the turbulence encounter. The roll rate will increase the oscillatory rolling motion during plunging motion, if the rolling damping is insufficient. The drop-off altitude will be enlarged by the oscillatory rolling motion during the sudden plunging motion.Research limitations/implicationsA lack of the measurement data of vertical wind speed sensor on board to verify the estimated values of damping term is one of the research limitations for this study. The fact or condition of being severe in sudden plunging motion can be judged through the analysis of oscillatory derivatives with both dynamic stability and damping terms.Practical implicationsThe roll rate will increase the oscillatory rolling motion during plunging motion, if the rolling damping is insufficient. The drop-off altitude will be enlarged by the oscillatory rolling motion during the sudden plunging motion. The horizontal wind shear or crosswind before the turbulence encounter will easily induce rolling motion and then initiated the sudden plunging motion during the turbulence encounter. If the drift angle is large, to turn off the autopilot of yaw control first and stabilize the rudder by the pedal. When passing through the atmosphere turbulence area, the pilots do not need to amend the heading angle urgently.Social implicationsThe flight safety prevention in avoidance of injuries for passengers and cabin crews is essential for the airlines. The horizontal wind shear or crosswind before the turbulence encounter will easily induce rolling motion and then initiated the sudden plunging motion during the turbulence encounter.Originality/valueThe flight safety prevention in avoidance of injuries for passengers and cabin crews is essential. The present assessment method is an innovation to examine the loss of control problems of aviation safety and promote the understanding of aerodynamic responses of the jet transport aircraft. It is expected to provide a valuable lecture for the international training courses for IATA – LOC-I program after this paper is being published.

Performance Analysis of Self Stabilizing Fixed Wing UAV for Real Time Surveillance

This paper discusses the design and performance of a fixed-wing Unmanned Aerial Vehicle (UAV) mounted with a flight stabilizer to stabilize itself during flight with a surveillance application. This work has been carried out considering the wider applications of low weight, low-cost UAVs. The addition of a flight stabilizer reduces the requirement of a pilot at all times during the flight. The paper discusses the design parameters of the UAV and the on-board electronics used. Various design parameters were evaluated, and a configuration was selected. The flight stabilizer was programmed accordingly, and the Flight Data Recorder (FDR) was put in place to record the in-flight parameters. Flight tests were carried out in many phases, to check the surveillance footage, aircraft’s performance without stabilizer and with stabilizer. Finally, the results obtained were optimized to fly the stable aircraft for surveillance over an area.

ESTIMATION OF FORCES ACTING TO AN OBJECT AT DEPLOYMENT OF PARACHUTE BRAKING SYSTEM BY DATA RECORDER OF OWN DESIGN

The article presents the results of research carried out using the developed and constructed flight data recorder. The purpose of developing the recorder was to optimize the parachute braking systems used in imitators of air targets produced at the armament division of the Air Force Institute of Technology. The main task of the recorder was to measure and record the linear accelerations occurring during the parachute system deployment.

Unstable Approach Detection and Analysis Based on Energy Management and a Deep Neural Network

The study of managing risk in aviation is the key to improving flight safety. Compared to the other flight operation phases, the approach and landing phases are more critical and dangerous. This study aims to detect and analyze unstable approaches in Taiwan through historical flight data. In addition to weather factors such as low visibility and crosswinds, human factors also account for a large part of the risk. From the accidents studied in the stochastic report of the Flight Safety Foundation, nearly 70% of the accidents occurred during the approach and landing phases, which were caused by improper control of aircraft energy. Since the information of the flight data recorder (FDR) is regarded as the airline’s confidential information, this study calculates the aircraft’s energy-related metrics and investigates the influence of non-weather-related factors on unstable approaches through a publicly available source, automatic dependent surveillance-broadcast (ADS-B) flight data. To evaluate the influence of weather- and non-weather-related factors, the outliers of each group classified by weather labels are detected and eliminated from the analysis by applying hierarchical density-based spatial clustering of applications with noise (HDBSCAN), which is utilized for detecting abnormal flights that are spatial anomalies. The deep learning method was adopted to detect and predict unstable arrival flights landing at Taipei Songshan Airport. The accuracy of the prediction for the normalized total energy and trajectory deviation of all flights is 85.15% and 82.11%, respectively. The results show that in different kinds of weather conditions, or not considering the weather, the models have similar good performance. The input features were analyzed after the model was obtained, and the flights detected as abnormal are discussed.

In-flight loss of consciousness in a fighter aircrew – G-LOC or No G-LOC conundrum

The differential diagnosis for inflight loss of consciousness in a fighter pilot is G-induced Loss of Consciousness (G-LOC) as it is physiological and 10–20% of fighter pilots may experience it during their career. However, it is very difficult to establish the diagnosis in many cases. Three cases of in-flight episodes of loss of consciousness (LOC) have been discussed in the paper highlighting how to investigate such cases to establish the diagnosis of G-LOC. Three cases have been discussed in the paper where two cases were considered as a case of G-LOC based on the circumstantial evidence and data from the flight data recorder. However, one case was diagnosed to be of LOC (Inv). One case did not benefit from the high-G training as he repeatedly experienced “GLOC” at very low G levels while wearing Anti-G Suit and performing an Anti-G Straining Maneuver (AGSM). He was recommended unfit for fighter flying. Another aircrew was experiencing G-LOC due to incorrect technique of AGSM as he had not undergone “high-G training.” After correction of technique, he could successfully meet the qualifying requirements of 9G for high G training. The third case was considered as a case of inflight LOC, not due to G exposure. Subsequently, he was diagnosed to have Focal Cortical Dysplasia. The paper describes the approach and aeromedical disposition of in-flight LOC among fighter aircrew. The paper also discusses the need for “G tolerance standard” at entry and high G training for fighter aircrew. The first case highlights that not all case of in-flight LOC among aircrew is G-LOC, even if it occurs in conjunction with high-G exposure. Ruling out the presence of any potential cause for inflight LOC is extremely important before labeling a recurrent case of inflight LOC as G-LOC. The second case re-iterates the fact that there are a set of people who will not be able to endure high G exposures due to inherent individual characteristics. These people need to be identified and screened at initial entry into fighter flying itself. The cause for the recurrent episodes of inflight G-LOC in the third case was identified as improper AGSM. The problem could be identified and corrected in the Dynamic Flight Simulator. This highlights the significance of high-G training using High Performance Human Centrifuge before the commencement of operational flying and high-G sorties and also establishes it as a diagnostic tool for such cases. In-flight LOC in a fighter pilot poses a challenge in diagnosis and differentiation from G-LOC. The paper discusses an approach to such a case in detail.