Multivariate State Estimation Technique Research Articles

Experimental study on engine gas-path component fault monitoring using exhaust gas electrostatic signal

This paper presents the recent development in engine gas-path components health monitoring using electrostatic sensors in combination with signal-processing techniques. Two ground-based engine electrostatic monitoring experiments are reported and the exhaust gas electrostatic monitoring signal-based fault-detection method is proposed. It is found that the water washing, oil leakage and combustor linear cracking result in an increase in the activity level of the electrostatic monitoring signal, which can be detected by the electrostatic monitoring system. For on-line health monitoring of the gas-path components, a baseline model-based fault-detection method is proposed and the multivariate state estimation technique is used to establish the baseline model for the electrostatic monitoring signal. The method is applied to a data set from a turbo-shaft engine electrostatic monitoring experiment. The results of the case study show that the system with the developed method is capable of detecting the gas-path component fault in an on-line fashion.

Research on Early Fault Diagnostic Method of Wind Turbines

Challenging environmental factors combined with high and turbulent winds make serious demands on wind turbines and result in significant component fault rates. In this paper, an early fault diagnostic research is conducted upon wind turbines. Firstly, the SCADA (Supervisory Control and Data Acquisition) system is used to analyze the units’ long-hour operating data, preparing for the further modeling work. Then the MSET (Multivariate State Estimation Technique) is adopted to estimate the temperature of the gear box and to obtain a result of high accuracy; with the Moving Window Calculation (MWC), the residual value between the estimated value and the real value is studied to get the dynamic trend of its average value; according to this trend in training, we define the threshold region of the residual mean value. Considering a man-made deviation in the observation vectors, faults of the gear box are simulated and studied. When the residual mean value curve exceeds the setting thresholds, an alert will be given to remind the operators of hidden problems in the unit. Research shows that this early diagnostic method is quite effective in detecting the abnormal performance of wind turbines in a real-time manner. DOI: http://dx.doi.org/10.11591/telkomnika.v11i5.2457

A Health Monitoring Method Based on Multivariate State Estimation Technique

The monitor of lithium-ion battery health is becoming a challenge because the performance of battery is effect by many environment factors. To address this problem, a new health monitor method based on Multivariate State Estimation Technique (MSET) and Sequential Probability Ratio Test (SPRT) is proposed in this paper. In order to demonstrate the performance gain of the method, a detailed experiment is performed based on a lithium-ion battery. By the comparison of performance parameters actual residuals and healthy residuals driven from the training data based on MSET, the fault detection can be implemented based on the SPRT.

Advanced electronic prognostics through system telemetry and pattern recognition methods

Electronic Prognostics (EP) is a technique used in high-reliability and high-availability systems to actively and proactively detect faults, allowing the reduction of system downtime and unplanned repairs. The approach of Sun Microsystems to EP consists of a Continuous System Telemetry Harness (CSTH) that is coupled with Sequential Probability Ratio Test (SPRT) and Multivariate State Estimation Technique (MSET) algorithms. This approach provides a unique and complete EP solution, harnessing the rich information from sensors and system variables, and providing means for their storage and analysis. The background theory behind SPRT and MSET techniques as well as their implementation for advanced EP in enterprise servers is presented in this paper.

Uncertainty Analysis of Memory Based Sensor Validation Techniques

This paper presents an analytical derivation and analysis of the uncertainty of the Multivariate State Estimation Technique (MSET). Like all other nonparametric techniques, MSET uncertainty consists of two parts: bias and variance. Bias is a systematic error in MSET inference and practically not computable and non-removable, but when properly regularized it is usually very small with respect to the variance when properly regularized. Variance, on the other hand, represents variability of the MSET estimate due to random noise in the data and can be estimated in real time. All the derivations and results are obtained for the inferential case. The MSET cost function is also derived which shows that MSET minimizes a weighted least squares cost function with weighting affected by the MSET memory matrix. The parallels between MSET and more traditional kernel techniques, namely kernel regression, are drawn and it is shown that MSET is a special type of kernel regression algorithm. The final section presents the results of the MSET uncertainty analysis for real world data obtained from a commercial nuclear power plant.

Diagnosis of Component Failures in the Space Shuttlemain Engines Using Bayesian Belief Network: A Feasibility Study

Although the Space Shuttle is a high reliability system, the health of the Space Shuttle must be accurately diagnosed in real-time. Two problems current plague the system, false alarms that may be costly, and missed alarms which may be not only expensive, but also dangerous to the crew. This paper describes the results of a feasibility study where a multivariate state estimation technique is coupled with a Bayesian Belief Network to provide both fault detection and fault diagnostic capabilities for the Space Shuttle Main Engines (SSME). Five component failure modes and several single sensor failures are simulated in our study and correctly diagnosed. The results indicate that this is a feasible fault detection and diagnosis technique and fault detection and diagnosis can be made earlier than standard redline methods allow.