Modeling Aircraft Wing Loads from Flight Data Using Neural Networks

Allen, Michael J.; Dibley, Ryan P.

doi:10.4271/2003-01-3025

Cited by 16 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [2] neural networks were used to model wing bending moment loads, torsion loads, and control surface hingemoments of the Active Aeroelastic Wing aircraft because they can account for uncharacterised nonlinear effects. Inputs to the model include aircraft rates, accelerations, and control surface positions.…”

Section: Methods Descriptionsmentioning

confidence: 99%

“…All samples within the training data base are copied and afterwards mirrored in lateral direction as the aircraft can be assumed as being symmetrical. Only parameters with lateral character such as the bending moment, lateral acceleration, sideslip angle, rudder-angle, roll rate/acceleration and yaw rate/acceleration are copied with reversed signs, the other parameters (as the angle of attack, dynamic pressure, mass) are copied without manipulation similar to [2].…”

Section: Data Basismentioning

confidence: 99%

See 1 more Smart Citation

Comparison of real-time flight loads estimation methods

2014

View full text Add to dashboard Cite

Nowadays flight load exceedance monitoring is an important task to both: the aircraft manufacturer as well as the operator. The estimation of flight loads is required in several phases during development and operation of an aircraft. The requirements are usually different, for e.g. calculation of design loads for certification and operational loads monitoring of stress and fatigue. The ability to determine aircraft operational loads (more) precisely may reduce the time in maintenance. Being able to detect critical load exceedance events during flight or in a post-process is also an enabler for e.g. loads/fatigue monitoring at operator level. In this paper, a novel system identification method named local model networks is applied to the field of flight loads estimation and compared to approaches based on artificial neural networks as found in the literature. The presented approach tries to overcome some limitations with respect to model creation, robustness, interand extrapolation.

show abstract

Section: Methods Descriptionsmentioning

confidence: 99%

Section: Data Basismentioning

confidence: 99%

Comparison of real-time flight loads estimation methods

2014

View full text Add to dashboard Cite

show abstract

“…where v ij is the weighted value from the ith neuron node in the input layer to the jth node in the hidden layer, net j is the activation value of the jth node, y j is the output of the hidden layer, and f H is the activation function of a node. A sigmoid function is usually expressed by formula (11):…”

Section: Hidden Layer Stagementioning

confidence: 99%

“…Allen and Dibley applied neural networks to model the wing loads of aircrafts. The linear model was established first as the starting point of the network training, which improved the accuracy of the neural network model [11]. Gómez-Escalonilla et al developed a parametric full-scale fatigue monitoring system for an Airbus A330 using an artificial neural network with several strain gauges installed on some areas of the wings and fuselage [4].…”

Section: Introductionmentioning

confidence: 99%

Application of Principal Component Analysis-Assisted Neural Networks for the Rotor Blade Load Prediction

Zheng

Jiao

Cui

2021

International Journal of Aerospace Engineering

View full text Add to dashboard Cite

This paper presents a novel approach of principal component analysis- (PCA-) assisted back propagation (BP) neural networks for the problem of rotor blade load prediction. 86.5 hours of real flight data were collected from many steady-state and transient flight maneuvers at different altitudes and airspeeds. Prediction of the blade loads was determined by the PCA-BP model from 16 flight parameters measured and monitored by the flight control computer already present in the helicopter. PCA was applied to reduce the dimension of the flight parameters influencing the component load and eliminate the correlation among flight parameters. Thus, obtained principal components were used as input vectors of the BP neural network. The combined PCA-BP neural network model was trained and tested by real flight data. Comparison of this model and to a BP neural network model as well as to a multiple linear regression (MLR) model was also done. The results of comparison demonstrate that the PCA-BP model has higher prediction precision with an average error of 2.46%, while 4.49% for BP and 10.20% for MLR. The results also reveal that the PCA-BP model has a shorter convergence path than the BP model. This method not only is useful in establishing the load spectra of helicopter rotor in-service where installation of strain gauges is impractical but also can reduce the cost of installation and maintenance measured by strain gauges.

show abstract

“…5, 6, and 7, but it differs primarily in that the loads model development presented in this report is based on a higher number of flight maneuvers and is for a fixed-wing fighter aircraft at multiple subsonic and supersonic flight conditions. Neural networks were investigated early in the development of the AAW loads model with much success 8 but were abandoned, because the high extrapolation required of the leading-edge flap hinge moment (HM) predictions could not be easily analyzed for uncertainty. This report describes the processes that were used to generate an AAW loads model from flight data.…”

Section: Figure 1 Active Aeroelastic Wing Airplane (Ec04-0361-02)mentioning

confidence: 99%