Susana Ferreiro

Application of Bayesian networks in prognostics for a new Integrated Vehicle Health Management concept

The aeronautics industry is attempting to implement important changes to its maintenance strategy. The article presents a new framework for making final decision on aeroplane maintenance actions. It emphasizes on the use of prognostics within this global framework to replace corrective and Preventive Maintenance practise for a predictive maintenance to minimize the cost of the maintenance support and to increase aircraft/fleet operability. The main objective of the article is to show the Bayesian network model as a useful technique for prognosis. The specific use case for predicting brake wear on the plane is developed based on this technique. The network allows estimate brake wear from the aircraft operational plan. This model, together with other models to make predictions for various components of the aeroplane (that should be monitored) offers a forward-looking approach of the status of the plane, allowing later the evaluation of different operational plans based on operational risk assessment and economic cost of each one of them depending on the scheduled checks.

A Bayesian network model integrated in a prognostics and health management system for aircraft line maintenance

The maintenance strategies of different industrial sectors are changing and evolving continuously in search of cost reduction, optimization of the operational reliability, the availability, and the resources used. The aeronautical industry belongs to this group. This article presents the global framework of a health management system as a new concept in aircraft line maintenance. This framework allows the transformation of the traditional maintenance (preventive and corrective, time based) into a predictive maintenance based on prognostic techniques. The article focuses in the development of the module ‘Condition View’, which has an essential role within this new model since it determines the deterioration and remaining useful life of critical aircraft parts. The case of detection and prediction of the aircraft break wear using a Bayesian net model that covers part of the functionality of ‘Condition View’ integrated in the global framework is presented.

Web services system for distributed technology upgrade within an e-maintenance framework

Nowadays, industrial maintenance is one of the most important tasks in the industry because its cost is too high, usually due to poor maintenance decisions. Traditionally, corrective maintenance and preventive maintenance are performed, but both of them, the excessive and the lacking maintenance can be harmful. In the last years, CBM (Condition Based Maintenance) technology or predictive maintenance has appeared in order to establish whether the system will fail during some future period and then take actions to avoid consequences. This paper shows the e-maintenance platform nicknamed DYNAWeb which is part of DYNAMITE project. DYNAWeb develops a CBM system based on OSA-CBM standard over MIMOSA comprising broad of capabilities like sensing and data acquisition, signal processing, health assessment, prognosis... This platform ensures the integration of all the components (software and hardware) using different technologies (sensor technologies, wireless communication technology) and providing them with agents and (Semantic) Web Services to allow the integration and the reuse among different applications.

Data Mining for Burr Detection (in the Drilling Process)

Drilling is the most important operation in aeronautic industry carried out previous to riveting. Its main problem lies with the burrs. Nowadays, there is a burr elimination task (manual task) subsequent to drilling and previous to riveting that increases manufacturing cost. It is necessary to develop a monitoring system to detect automatically and on-line when the generated burr is out of aeronautic limits, and then deburring. This system would reduce holes deburring to the holes which really are out of tolerance limits, focusing in trying to avoid false negatives. The article shows an improvement in burr generation prediction, using Data Mining techniques versus current mathematical model. It gives an overview of the process from data preparation and selection to data analysis (with machine learning algorithms) and evaluation of the models.

A Bayesian network for burr detection in the drilling process

One of the most important processes in the aeronautical sector is drilling. The main problem associated with drilling is burr. There is a tolerance level for this burr and it cannot exceed 127 microns, which would provoke structural damage and other problems. Currently, the burr elimination task is carried out visually and manually with the aim of guaranteeing quality in the process. However, it is an expensive procedure and needs to be replaced by a motorized system capable of automatically detecting in which holes the burr exceeds the permitted level and has to be eliminated or reduced. The paper presents a burr prediction model for high speed drilling in dry conditions on aluminium (Al 7075-T6), based on a Bayesian network learned from a set of experiments based on parameters taken from the internal signal of the machine and parameters from the condition process. The paper shows the efficiency and validity of the model in the prediction of the apparition of burr during the drilling and compares the results with other data-mining techniques.

Using ensembles of regression trees to monitor lubricating oil quality

This work describes a new on-line sensor that includes a novel calibration process for the real-time condition monitoring of lubricating oil. The parameter studied with this sensor has been the variation of the Total Acid Number (TAN) since the beginning of oils operation, which is one of the most important laboratory parameters used to determine the degradation status of lubricating oil. The calibration of the sensor has been done using machine learning methods with the aim to obtain a robust predictive model. The methods used are ensembles of regression trees. Ensembles are combinations of models that often are able to improve the results of individual models. In this work the individual models were regression trees. Several ensemble methods were studied, the best results were obtained with Rotation Forests.

Bayesian Network for quality control in the drilling process

Abstract Nowadays, the aeronautic industry requires the automation of certain processes to minimize economic costs and to optimize resources, ensuring at the same time the quality of these processes. One of the most important tasks in this sector is the drilling process, the main problem of which lies in the occurrence of burr. Today there is a manual burr elimination task subsequent to drilling and previous to riveting which guarantees the quality of the process, where the permissible burr size is set at under 127 microns, imposed by aeronautic industry. This task increases manufacturing costs and it must be replaced by a monitoring system in order to detect automatically and on-line when the burr is outside this limit and to reduce the number of holes to be removed. This article shows the efficacy of Bayesian networks for predicting burr generation in the drilling process, which is an easy model to interpret and to integrate into the final system. Moreover, the article provides the most influential parameters in the generation of burr in the process.

Prognostics applied to aircraft line maintenance: Brake wear prediction based on Bayesian Networks

Abstract The aeronautics industry is attempting to implement important changes to its maintenance strategy. The corrective and preventive maintenance practise is evolving to predictive maintenance to minimize the cost of the maintenance support and to increase operational reliability and availability. The main objective of the article is to show the new predictive maintenance concept and the usefulness of Bayesian networks in supporting the intelligent function “decision support”. This function provides specific models for the prediction of degradation in some of the critical components of the aircraft, the basis for the new maintenance strategy concept based on prognosis. In addition, the article discusses the development of the Bayesian network model for the specific case of predicting brake wear.