A Hybrid Inference Approach for Health Diagnostics With Unexampled Faulty States

2012 ◽  
Author(s):  
Prasanna Tamilselvan ◽  
Pingfeng Wang
Keyword(s):  
Artificial Intelligence ◽  
Real Time ◽  
Power Transformer ◽  
Roller Bearing ◽  
Health State ◽  
Health States ◽  
System Failures ◽  
State Classification ◽  
Testing Stage ◽  
Health Diagnosis

System health diagnostics provides diversified benefits such as improved safety, improved reliability and reduced costs for the operation and maintenance of engineered systems. Successful health diagnostics requires the knowledge of system failures. However, with an increasing complexity it is extraordinarily difficult to have a well-tested system so that all potential faulty states can be realized and studied at product testing stage. Thus, real time health diagnostics requires automatic detection of unexampled faulty states through the sensory signals to avoid sudden catastrophic system failures. This paper presents a hybrid inference approach (HIA) for structural health diagnosis with unexampled faulty states, which employs a two-fold inference process comprising of preliminary statistical learning based anomaly detection and artificial intelligence based health state classification for real time condition monitoring. The HIA is able to identify and isolate the unexampled faulty states through interactively detecting the deviation of sensory data from the known health states and forming new health states autonomously. The proposed approach takes the advantages of both statistical approaches and artificial intelligence based techniques and integrates them together in a unified diagnosis framework. The performance of proposed HIA is demonstrated with a power transformer and roller bearing health diagnosis case studies, where Mahalanobis distance serves as a representative statistical inference approach.

scholarly journals Multi-Sensor Health Diagnosis Using Deep Belief Network Based State Classification

2011 ◽  
Author(s):  
Prasanna Tamilselvan ◽  
Pingfeng Wang ◽  
Byeng D. Youn
Keyword(s):  
Health State ◽  
Layer By Layer ◽  
Classification Models ◽  
Restricted Boltzmann Machines ◽  
Health States ◽  
Sensory Data ◽  
Complex Engineered Systems ◽  
State Classification ◽  
Health Diagnosis ◽  
Sensor Health

Effective health diagnosis provides multifarious benefits such as improved safety, improved reliability and reduced costs for the operation and maintenance of complex engineered systems. This paper presents a novel multi-sensor health diagnosis method using Deep Belief Networks (DBN) based state classification. The DBN has recently become a popular approach in machine learning for its promised advantages such as fast inference and the ability to encode richer and higher order network structures. The DBN employs a hierarchical structure with multiple stacked Restricted Boltzmann Machines and works through a layer by layer successive learning process. The proposed multi-sensor health diagnosis methodology using the DBN based state classification can be structured in three consecutive stages: first, defining health states and collecting sensory data for DBN training and testing; second, developing DBN based classification models for the diagnosis of predefined health states; third, validating DBN classification models with testing sensory dataset. The performance of health diagnostics using DBN based health state classification is compared with four existing classification methods and demonstrated with two case studies.

Methods for obtaining health state utility values: generic preference-based measures of health

2016 ◽  
Author(s):  
John Brazier ◽  
Julie Ratcliffe ◽  
Joshua A. Salomon ◽  
Aki Tsuchiya
Keyword(s):  
Health State Utility ◽  
Health State ◽  
Multiattribute Utility ◽  
Health States ◽  
Utility Values ◽  
Published Evidence ◽  
State Classification ◽  
Health State Utility Values ◽  
Generic Measures ◽  
Health State Classification

This chapter describes the six most widely used generic preference-based measures of health (GPBMs) (also known as multiattribute utility scales): EQ-5D, SF-6D, HUI, AQoL, 15D, and QWB. GPBMs have become the most widely used method for obtaining health state utility values. They contain a health state classification with multilevel dimensions that together describe a universe of health states and a set of values (where full health = 1 and dead = 0) for each health state obtained by eliciting the preferences (typically) of members of the general population. These measures are reviewed in terms of their content, methods of valuation, the scores they generate, and the possible reasons for the differences found. Their performance is reviewed using published evidence on their validity across conditions, and the implications for their use in policy making discussed. The chapter also reviews the generic measures available for use in populations of children and adolescents.

A Generic Sensor Network Design Framework Based on a Detectability Measure

2010 ◽  
Author(s):  
Pingfeng Wang ◽  
Byeng D. Youn ◽  
Chao Hu
Keyword(s):  
Sensor Network ◽  
Health Management ◽  
Power Transformer ◽  
Relevant Information ◽  
Mixed Integer ◽  
Health State ◽  
Design Framework ◽  
Engineering Structures ◽  
State Classification ◽  
Technological Advances

Significant technological advances in sensing and communication promote the use of large sensor networks to monitor structural systems, identify damages, and quantify damage levels. Prognostics and health management (PHM) technique has been developed and applied for a variety of safety-critical engineering structures, given the critical needs of the structure health state awareness. The PHM performance highly relies on real-time sensory signals which convey the structural health relevant information. Designing an optimal structural sensor network (SN) with high detectability is thus of great importance to the PHM performance. This paper proposes a generic SN design framework using a detectability measure while accounting for uncertainties in material properties and geometric tolerances. Detectability is defined to quantify the performance of a given SN. Then, detectability analysis will be developed based on structural simulations and health state classification. Finally, the generic SN design framework can be formulated as a mixed integer nonlinear programming (MINLP) using the detectability measure and genetic algorithms (GAs) will be employed to solve the SN design optimization problem. A power transformer study will be used to demonstrate the feasibility of the proposed generic SN design methodology.

Probabilistic Design of Smart Sensing Functions for Structural Health Monitoring and Prognosis

2013 ◽  
Author(s):  
Abdulaziz T. Almaktoom ◽  
Zequn Wang ◽  
Pingfeng Wang
Keyword(s):  
Health Management ◽  
Power Transformer ◽  
Relevant Information ◽  
Mixed Integer ◽  
Health State ◽  
Design Framework ◽  
Engineering Structures ◽  
Structural Health ◽  
State Classification ◽  
Technological Advances

Significant technological advances in sensing and communication promote the use of large sensor networks to monitor structural systems, identify damages, and quantify damage levels. Prognostics and health management (PHM) technique has been developed and applied for a variety of safety-critical engineering structures, given the critical needs of the structure health state awareness. The PHM performance highly relies on real-time sensory signals which convey the structural health relevant information. Designing an optimal structural sensor network (SN) with high detectability is thus of great importance to the PHM performance. This paper proposes a generic SN design framework using a detectability measure while accounting for uncertainties in material properties and geometric tolerances. Detectability is defined to quantify the performance of a given SN. Then, detectability analysis will be developed based on structural simulations and health state classification. Finally, the generic SN design framework can be formulated as a mixed integer nonlinear programming (MINLP) using the detectability measure and genetic algorithms (GAs) will be employed to solve the SN design optimization problem. A power transformer study will be used to demonstrate the feasibility of the proposed generic SN design methodology.

scholarly journals BattleNet: Capturing Advantageous Battlefield in RTS Games (Student Abstract)

2020 ◽  
Vol 34 (10) ◽  
pp. 13849-13850
Author(s):  
Donghyeon Lee ◽  
Man-Je Kim ◽  
Chang Wook Ahn
Keyword(s):  
Artificial Intelligence ◽  
Decision Making ◽  
Real Time ◽  
Large Scale ◽  
Outcome Predictor ◽  
Short Term ◽  
Rts Game

In a real-time strategy (RTS) game, StarCraft II, players need to know the consequences before making a decision in combat. We propose a combat outcome predictor which utilizes terrain information as well as squad information. For training the model, we generated a StarCraft II combat dataset by simulating diverse and large-scale combat situations. The overall accuracy of our model was 89.7%. Our predictor can be integrated into the artificial intelligence agent for RTS games as a short-term decision-making module.

Estimating Utility Values for Health States of DFU Patients Using EQ-5D-5L and cTTO

2021 ◽  
pp. 153473462110143
Author(s):  
Morteza Arab-Zozani ◽  
Hossein Safari ◽  
Zoha Dori ◽  
Somayeh Afshari ◽  
Hosein Ameri ◽  
...  
Keyword(s):  
Foot Ulcer ◽  
Diabetic Foot Ulcer ◽  
Economic Modeling ◽  
Health State ◽  
Health States ◽  
Utility Values ◽  
The Mean ◽  
The Difference ◽  
Diabetes Center ◽  
Health State Utility Values

Health-state utility values of diabetic foot ulcer (DFU) patients are necessary for clinical praxis and economic modeling. The purpose of this study was to estimate utility values in DFU patients using the EuroQol-5-dimension-5-level (EQ-5D-5L) and composite time trade-off (cTTO). The EQ-5D-5L and cTTO were used for estimating utility values. Data were collected from 228 patients referred to the largest governmental diabetes center in the South of Iran, Yazd province. When appropriate, independent sample t-test or analysis of variance test was used to test the difference in the utility values in each of the demographic and clinical characteristics of the patients. Finally, the BetaMix was used to identify predictors of the utility values. The means of EQ-5D-5L and cTTO values were 0.55( SD 0.21) and 0.67( SD 0.23), respectively. Anxiety and pain were the most common problems reported by the patients. The difference between the mean EQ-5D-5L values was significant for age, grade of ulcer, number of comorbidities, and having complications. In addition, variables of gender, age, grade of ulcer, and having complications were significant predictors of the EQ-5D-5L. The difference between the mean cTTO values was significant for age, employment status, grade of ulcer, number of comorbidities, and having complications. Moreover, variables of gender, age, grade of ulcer, number of comorbidities, and developing complications were significant predictors of cTTO. The current study provided estimates of utility values for DFU patients for clinical praxis and economic modeling. These estimates, similar to utilities reported in other studies, were low. Identifying strategies to decrease anxiety/depression and pain in patients is important to improve the utility values.

Sign in / Sign up

Export Citation Format

Share Document