scispace - formally typeset
Search or ask a question

Data Driven Methodology Based on Artificial Immune Systems for Damage Detection

TL;DR: A methodology for structural damage detection using a type of artificial intelligence that is called artificial immune system is presented and results show that the proposed methodology allows to detect damages in the experimental setup.
Abstract: Structural Health Monitoring is a growing area of interest given the benefits obtained from its use. This area includes different tasks in the damage identification process, the main important, is the damage detection since an early detection allows to avoid possible catastrophes in structures in service. Practical solutions require a big quantity of sensors and a robust system to process and obtain a reliable solution. In this sense, bio-inspired algorithms provide tools for an effective data analysis taking advantage of the developments provided by the nature by means of computational algorithms. As a contribution in this area, this paper presents a methodology for structural damage detection using a type of artificial intelligence that is called artificial immune system. The developed methodology includes the inspection of the structure by means of a distributed piezoelectric active sensor network at different actuation phases to define a baseline by each actuation phase using data from the structure when it is known as healthy. In a second step, same experiments are performed to the structure when its structural state is unknown to determine the presence of damage by using the developed artificial immune system. Results show that the proposed methodology allows to detect damages in the experimental setup.
Citations
More filters
Journal ArticleDOI
TL;DR: A bioinspired strategy for the detection of structural changes using an artificial immune system (AIS) and a statistical data-driven modeling approach by means of a distributed piezoelectric active sensor network at different actuation phases is introduced.
Abstract: Among all the aspects that are linked to a structural health monitoring (SHM) system, algorithms, strategies, or methods for damage detection are currently playing an important role in improving the operational reliability of critical structures in several industrial sectors. This paper introduces a bioinspired strategy for the detection of structural changes using an artificial immune system (AIS) and a statistical data-driven modeling approach by means of a distributed piezoelectric active sensor network at different actuation phases. Damage detection and classification of structural changes using ultrasonic signals are traditionally performed using methods based on the time of flight. The approach followed in this paper is a data-based approach based on AIS, where sensor data fusion, feature extraction, and pattern recognition are evaluated. One of the key advantages of the proposed methodology is that the need to develop and validate a mathematical model is eliminated. The proposed methodology is applied, tested, and validated with data collected from two sections of an aircraft skin panel. The results show that the presented methodology is able to accurately detect damage.

31 citations


Cites background from "Data Driven Methodology Based on Ar..."

  • ...[14], is rather different because it is a databased approach based on AIS (artificial immune system), where sensor data fusion, feature extraction, and pattern recognition are evaluated....

    [...]

Journal ArticleDOI
TL;DR: This work introduces a taste recognition methodology, which is composed of several steps including unfolding data, data normalization, principal component analysis for compressing the data, and classification through different machine learning models.
Abstract: Electronic tongue-type sensor arrays are devices used to determine the quality of substances and seek to imitate the main components of the human sense of taste. For this purpose, an electronic ton...

17 citations


Cites background from "Data Driven Methodology Based on Ar..."

  • ...The reason is that group scaling considers changes between sensors and does not process them independently.(30,34) In group scaling, the variables are divided into a predefined number of blocks of equal size and each block is scaled by the large mean of their standard deviations....

    [...]

Journal ArticleDOI
TL;DR: In this article, a new method of prognosis based on remaining useful life (RUL) prediction for degradation assessment was proposed to improve the accuracy of forecasting the system state, which can be effectively applied to many systems for prognosis.
Abstract: Purpose The purpose of this paper is to create a new method of prognosis based on remaining useful life (RUL) prediction for degradation assessment. Design/methodology/approach In the present paper the authors describe a new method of prognosis to improve the accuracy of forecasting the system state. This framework of forecasting integrates the model-based information and the hybrid approach, which employs the structured residuals in the first part and the particle filter in the second part. Findings The performance of the suggested fusion framework is employed to predict the RUL of battery pack in hybrid electric vehicle. The results show that the proposed method is plausible due to the good prediction of RUL, and can be effectively applied to many systems for prognosis. Originality/value In this study the authors illustrate how the suggested method can provide an accurate prediction of the RUL over conventional data-driven methods without physical model and classical particle filter with a single damage model.

4 citations

Proceedings ArticleDOI
25 May 2016
TL;DR: In this article, a new method of prognosis is developed to improve the accuracy of the system state by integrating the model-based information and the hybrid approach, which employs the structured residuals in the first one and the Particle Filter in the second.
Abstract: All industrial systems and machines are subjected to degradation processes which can be related to the operating conditions. This degradation can cause unwanted stops at any time and major maintenance work sometimes. Prognostic activity is now recognized as a key feature in maintenance strategies, it gives operators a potent tool in decision making by quantifying how much time is left until functionality is lost. The reason is to plan the heavy interventions and to manage the stock of spare parts. In addition, it can be used to improve the characterization of the material proprieties that govern damage propagation for the structure being monitored. The tool for measuring the size of damage and health status in industry is the Remaining Useful Life (RUL), which can be estimated by using four main approaches, namely experience based, model-based, data-driven and hybrid approaches. A new method of prognosis is developed in this paper to improve the accuracy of the system state. This framework of forecasting integrates the model-based information and the hybrid approach, which employs the structured residuals in the first one and the Particle Filter in the second. The performance of the suggested fusion prognostic framework is employed to predict the RUL of battery pack in Hybrid Electric Vehicle (HEV).

3 citations


Cites methods from "Data Driven Methodology Based on Ar..."

  • ...In this study we have illustrated how the Developed Particle Filter method can provide an accurate estimate of the RUL over conventional data-driven methods without physical model [40, 41]....

    [...]

References
More filters
Book
18 May 2001
TL;DR: Independent component analysis as mentioned in this paper is a statistical generative model based on sparse coding, which is basically a proper probabilistic formulation of the ideas underpinning sparse coding and can be interpreted as providing a Bayesian prior.
Abstract: In this chapter, we discuss a statistical generative model called independent component analysis. It is basically a proper probabilistic formulation of the ideas underpinning sparse coding. It shows how sparse coding can be interpreted as providing a Bayesian prior, and answers some questions which were not properly answered in the sparse coding framework.

8,333 citations

01 Jan 2012
TL;DR: The standardization of the IC model is talked about, and on the basis of n independent copies of x, the aim is to find an estimate of an unmixing matrix Γ such that Γx has independent components.

2,296 citations


"Data Driven Methodology Based on Ar..." refers methods or result in this paper

  • ...Previous works by the authors [12][16] have shown that group scaling presents a better performance compared with other kind of normalizations, EWSHM 2014 - Nantes, France...

    [...]

  • ...The general idea in the use of PCA is to find a smaller set of variables with less redundancy but with a minimal loss of information [16]....

    [...]

Journal ArticleDOI
TL;DR: The role of the time variable in batch process data is considered and methods suggested to predict the per cent completion of batch runs with unequal duration are discussed.
Abstract: Batch process data can be arranged in a three-way matrix (batch × variable × time). This paper provides a critical discussion of various aspects of the treatment of these multiway data. First, several methods that have been proposed for decomposing three-way data matrices are discussed in the context of batch process data analysis and monitoring. These methods are multiway principal component analysis (MPCA)—also called Tucker1—parallel factor analysis (PARAFAC) and Tucker3. Secondly, different ways of unfolding, mean centering and scaling the three-way matrix are compared and discussed with respect to their effects on the analysis of batch data. Finally, the role of the time variable in batch process data is considered and methods suggested to predict the per cent completion of batch runs with unequal duration are discussed. Copyright © 1999 John Wiley & Sons, Ltd.

226 citations


"Data Driven Methodology Based on Ar..." refers background in this paper

  • ...Artificial immune system (AIS) is an adaptive and bio-inspired computational system based on the processes, performance of the human immune system and its properties such as diversity, error tolerance, dynamic learning, adaptation, distributed computation and self-monitoring [11]....

    [...]

  • ...More information about the normalizations can be found in [11],[12]....

    [...]

Journal ArticleDOI
TL;DR: In this article, a data-driven statistical approach for damage classification is proposed, which is constructed over a distributed piezoelectric active sensor network for excitation and measurement of vibrational structural responses.
Abstract: SUMMARY Damage classification is an important issue within SHM going beyond the purely damage detection. This paper proposes a data-driven statistical approach for damage classification, which is constructed over a distributed piezoelectric active sensor network for excitation and measurement of vibrational structural responses. At different phases, a single piezoelectric transducer is used as actuator, and the others are used as sensors. An initial baseline model for each phase for the healthy structure is built by applying PCA to the data collected in several experiments. In addition, same experiments are performed with the structure in different states (damaged or not), and the dynamic responses are projected into the different baseline PCA models for each actuator. Some of these projections and damage indices are used as input features for a self-organizing map, which is properly trained and validated to build a pattern baseline model. This baseline is further used as a reference for blind diagnosis tests of structures. Both training/validation and diagnosis modes are experimentally assessed using an aluminum plate instrumented with four piezoelectric transducers. Damages are simulated by adding mass at different positions. Results show that all these damages are successfully classified both in the baseline pattern model and in further diagnosis tests. Copyright © 2012 John Wiley & Sons, Ltd.

118 citations


"Data Driven Methodology Based on Ar..." refers background or methods or result in this paper

  • ...The application of PCA starts with a matrix X which contains information from m sensors and n experimental trials [12]....

    [...]

  • ...Previous works by the authors [12][16] have shown that group scaling presents a better performance compared with other kind of normalizations, EWSHM 2014 - Nantes, France...

    [...]

  • ...In both steps, the data from the structure are collected and organized in a unfolded matrix [12] which contains the information of the signal collected by the sensors in different parts of the structure by each actuation phase....

    [...]

  • ...More information about the normalizations can be found in [11],[12]....

    [...]