Krzysztof Dragan

The Development of the Non-Parametric Classification Models for the Damage Monitoring on the Example of the ORLIK Aircraft Structure

This paper presents approach for the damage growth monitoring and early damage detection in the PZL-130 ORLIK TC II turbo propeller military trainer based on the array of the PZT sensors which will be deployed in the structure of the aircraft. Special attention will be paid to the preliminary results of the statistical models which provide an automated tool to infer about the damage presence and its size. In particular the effectiveness of the selected signal characteristics will be assessed using dimensional reduction methods (PCA) and the so called averaged damage indices will be described. Verification of the several classification models based on the emerged damage indices will be presented using cross validation techniques. The preliminary results of the data collected from the subcomponents tests with the model description, as well as approach for the SHM system design will be delivered. The verification of the models results will be presented on the example of the aerospace structures.

Damage Size Estimation of the Aircraft Structure with Use of Embedded Sensor Network Generating Elastic Waves

One of the approach to develop a system of continues, automated monitoring of the health of the structures is to use elastic waves excited in a given medium by piezoelectric transducers network. Elastic waves depending on their source and the geometry of the structure under consideration can propagate over significant distance. They are also sensitive to local structure discontinuities and deformations providing a tool to detect local damage of large aerospace structures. In the paper the issue of fatigue crack growth monitoring by means of elastic guided waves actuated by a sparse array of sensors will be presented. In particular we propose signal characteristics, robust enough to detect different kinds of damages: Barely Visible Impact Damages (BVIDs) in composite materials and fatigue cracks of metallic structures. The model description and the results of specimen tests verifying damage detection capabilities of the proposed signal characteristics are delivered in the paper. Some issues concerning the proposed damage indices and its application to damage detection and its monitoring are also discussed.

Energy Correlated Damage Indices in Fatigue Crack Extent Quantification

Signals received by piezoelectric transducers (PZT) network can be influenced by many factors. Apart from environmental conditions, whose variability should be compensated, significant difference in a signal can be also caused by relative geometry changes of a designed sensors node, e.g. the damage localization and its orientation with respect to sensors location in the node. In the adopted approach a set of damage indices (DIs), carrying marginal signal information content and correlated with the total energy received by a given sensor are proposed. These are sensitive to the two main modes of guided wave interaction with a fatigue crack, i.e. its transmission and reflection from a damage. Detailed description of DIs detection capabilities are delivered in the paper. Two dimensional reduction techniques: Principal Component Analysis and Fishers Linear Discriminant are compared. The results of the data collected from specimen fatigue test are used to compare several classification models based on the emerged effective damage indices.

Damage Detection in Riveted Aircraft Elements Based on the Electromechanical Impedance Measurements

This work is devoted to the use of the electromechanical impedance method for the damage detection in a riveted aircraft element. In the first part of the paper a theoretical background of the impedance-based damage detection technique is made. Next, the description of the utilised experimental set-up is described. Then, an application of the method used to detect damage in a wing sheathing of a turboprop training aircraft is presented. A damage scheme incorporating multiple notches through the selected rivets is considered. Finally, the suitability of the described method to distinguish close and far field damages is discussed.

Structural Health Monitoring of a Composite Panel Based on PZT Sensors and a Transfer Impedance Framework

One of the ideas for development of Structural Health Monitoring (SHM) systems is based on excitation of elastic waves by a network of PZT piezoelectric transducers integrated with the structure. In the paper, a variant of the so-called Transfer Impedance (TI) approach to SHM is followed. Signal characteristics, called the Damage Indices (DIs), were proposed for data presentation and analysis. The idea underlying the definition of DIs was to maintain most of the information carried by the voltage induced on PZT sensors by elastic waves. In particular, the DIs proposed in the paper should be sensitive to all types of damage which can influence the amplitude or the phase of the voltage induced on the sensor. Properties of the proposed DIs were investigated experimentally using a GFRP composite panel equipped with PZT networks attached to its surface and embedded into its internal structure. Repeatability and stability of DI indications under controlled conditions were verified in tests. Also, some performance indicators for surface-attached and structure-embedded sensors were obtained. The DIs’ behavior was dependent mostly on the presence of a simulated damage in the structure. Anisotropy of mechanical properties of the specimen, geometrical properties of PZT network as well as, to some extent, the technology of sensor integration with the structure were irrelevant for damage indication. This property enables the method to be used for damage detection and classification.

Application of artificial neural networks for damage indices classification with the use of Lamb waves for the aerospace structures

Lamb waves (LW) are used for damage detection and health monitoring due to the long range propagation ability and sensitivity to structural integrity changes as well as their robustness in different applications. However, due to the dispersive character and multimode nature of LWs, analysis of the acquired ultrasonic signals is very complex. It becomes even more difficult when applied to a complex structure, for instance, an aircraft component with riveted joints and stringers characterized by difficult geometries. Therefore, numerous approaches to the evaluation of damage indices have been proposed in the literature. In this paper, comparison a number of damage indices applied to LWs testing in aircraft aluminum panels. The damage indices, known from the literature have been selected from the application point of view. Artificial neural network has been used for the classification of fatigue cracks and artificial damages induced in the specimens taken from a real aircraft structure. Article presents results of simulation, data analysis and data classification obtained using selected and dedicated neural network. The main aim of the presented research was to develop signal processing and signal classification methods for an aircraft health monitoring system. The article presents a part of the research carried out in the project named SYMOST.

In-Service Flaw Detection and Quantification in the Composite Structures of Aircraft

In-Service Flaw Detection and Quantification in the Composite Structures of Aircraft Taking into consideration the increased usage of composites for aircraft structures there is a necessity for gathering information about structural integrity of such components. During the manufacturing of composites as well as during in service and maintenance procedures there is a possibility for damage occurrence. There is a large number of failure modes which can happen in such structures. These failure modes affect structural integrity and durability. In this work modern approach for detection of composites damage detection such as: delaminations, disbonds, foreign object inclusion and core damage has been presented. This detection is possible with the use of advanced P-C aided Non Destructive Testing methods. In the article nondestructive testing results for the composite vertical tail skins on MiG-29 aircraft will be delivered as well as some results of F-16 horizontal stabilizer and W-3 helicopter main rotor blades. Moreover some results of the composite honeycomb and sandwich structures will be presented based on the materials used in the construction of gliders and small aircraft. Factors affecting structural integrity and durability of the composites will be highlighted as well as necessity of the inspection with the use of modern NDT techniques. At the end some effort with Structural Health Monitoring connected with possibility of condition monitoring of composites will be presented.

Artificial neural network ensembles for fatigue damage detection in aircraft

Neural networks are commonly recognized tools for the classification of multidimensional data obtained in structural health monitoring (SHM) systems. Their configuration for a given scenario is, however, a challenging task, which limits the possibilities of their practical applications. In this article the authors propose using the neural network ensemble approach for the classification of SHM data generated by guided wave sensor networks. The overproduce and choose strategy is used for designing ensembles containing different types and sizes of neural networks. The proposed method allows for a significant increase of the state assessment reliability, which is illustrated by the results obtained from the practical industrial case of a full-scale aircraft test. The method is verified in the process of detecting fatigue cracks propagating in the aircraft load-carrying structure. The long-term experiments are performed in variable environmental conditions with a net of structure-embedded piezoelectric sensors.

3D Reconstruction of Ultrasonic B-Scans for Nondestructive Testing of Composites

The paper presents an approach to 3D reconstruction of a sequence of ultrasonic B-Scans for the purpose of facilitating nondestructive testing of composites. The results of ultrasonic testing of a carbon fiber reinforced polymer specimen with barely visible impact damage was used for algorithm testing. 3D visualisation of damage based on image thresholding, contour extraction and volume rendering facilitates interpretation of ultrasonic data and can be useful in the assessment of a flaw size and location, including its depth. Accuracy of the 3D reconstruction of the internal damage of the tested specimen was verified on the basis of reference data acquired with the X-ray computed tomography. Owing to the low computational complexity of the proposed algorithm it could be applied during ultrasonic inspections of composite structures.

A method to compensate non-damage-related influences on Damage Indices used for pitch-catch scheme of piezoelectric transducer based Structural Health Monitoring

The risk of false calls of structural health monitoring systems is as much important for their application as their damage detection capabilities. Structural health monitoring based on guided waves propagation is particularly vulnerable to false calls. Signals acquired for piezoelectric transducer networks can be changed by many factors other than damage, for example, environmental factors or those related to the transducers’ aging or degradation of the transducers’ bonding with the monitored structure. A lot of studies were devoted to examine non-damage-related influences on structural health monitoring systems based on Lamb waves and to compensate the undesired effects. Most of compensation methods act on the level of the signal, that is, for a given factor influencing performance of piezoelectric transducers, signals are transformed to match the corresponding baselines. After such compensation procedure, the Damage Indices are calculated for the purpose of damage detection. In order to compensate the impact of all of non-damage-related factors, all of them need to be, at least, recognized. In this article, a different technique to compensate changes of Damage Indices values caused by factors other than damage is proposed. The method does not involve any operation on signals, but on the Damage Indices themselves. The factors causing Damage Indices’ changes neither have to be measured nor are even known. The capabilities of the method have been evaluated on the example of fatigue cracks detection in laboratory specimen tests and using results obtained during Full-Scale Fatigue Test of an aircraft.