Skip to main content

Xiaomo Jiang

Ohio State University, Civil and Environmental Engineering and Geodetic Science Department, Graduate Student

Followers

17

Following

5

Public Views

Address: Roswell, Georgia, United States

less

Saint Mary's University (Canada)

The University of Sheffield

Universidade do Porto

Marta Diaz-Guardamino

Durham University

Armando Marques-Guedes

UNL - New University of Lisbon

California Polytechnic State University at San Luis Obispo

Franco Bontempi

Università degli Studi "La Sapienza" di Roma

Università di Catania

Dionysius Siringoringo

Yokohama National University

Gaurav Bhatnagar

University of Windsor

Interests

Uploads

Papers by Xiaomo Jiang

improved OAKR approach to condition monitoring of rotating machinery

PHM Society Asia-Pacific Conference

Faults in main subsystems or components of a rotating machine often causes unscheduled shutdown, ... more Faults in main subsystems or components of a rotating machine often causes unscheduled shutdown, which may lead to not only huge economic losses, but also safety accidents. As an important part of intelligent maintenance, condition monitoring becomes a powerful tool in reducing maintenance costs through automatic fault alarming, thereby reducing potential downtime while improving system safety and reliability. An optimized auto-associative kernelregression (OAKR) model has been proposed recently and demonstrated as a promising tool for condition monitoring of various turbomachines, which is independent of fault mode and machine type. However, the fault identification accuracy of this approach largely relies on data quality in practical applications. Data incompleteness, parameter variation and system complexity often result in the inaccuracy of fault alarming for complicated rotating machinery. This paper proposes an improved OAKR method to address these issues, including utilizing ...

Tension Force Estimation of Cables with Two Intermediate Supports

International Journal of Structural Stability and Dynamics, Feb 19, 2020

The presence of intermediate supports usually imposes difficulties in identifying the tension for... more The presence of intermediate supports usually imposes difficulties in identifying the tension force of stayed cables in cable-stayed bridges or hanger cables in arch bridges. This paper establishes the partial differential equations of motion of the cable and derives two numerical models with (Model 1) and without (Model 2) considering the flexural rigidity. The effects of two intermediate supports on the identification accuracy of the cable tension force are further studied analytically and experimentally. The effects of several non-dimensional parameters (e.g. damper location, support stiffness, flexural rigidity, and mode order of the cable) on the identification accuracy of the models are also investigated. It is theoretically concluded that the simplified Model 2 provides acceptable accuracy on tension force identification when the non-dimensional parameter [Formula: see text] is greater than 90 (slender cables), whereas the accurate Model 1 can be applied for tension force identification at any scenarios. The feasibility of two models is further verified by three numerical examples and field tests on two real-world arch bridges.

Mode shape–aided cable force determination using digital image correlation

Structural Health Monitoring-an International Journal, Sep 22, 2020

Assuming the distance between two nodal points of a specific cable vibration mode as the effectiv... more Assuming the distance between two nodal points of a specific cable vibration mode as the effective length of a pinned–pinned cable, mode shape–aided cable tension estimation methods are employed to estimate the cable force. This article proposes a framework based on digital image correlation technique for remote measurement of the dynamic displacement time history of cables in cable structures. Frequency domain decomposition technique is then used to extract the cable natural frequencies and mode shapes. Identified cable mode shapes are used along with a tensioned pinned–pinned cable model to estimate the cable force. Accuracy of the proposed methodology is investigated using the experimental data coming from a laboratory-scale test setup and hanger cables of a real-world arch bridge.

Enhancing Turbine Performance Degradation Prediction with Atmospheric Factors

Annual Conference of the PHM Society, 2016

Heavy duty gas turbine engines are not only ingesting the air, but also eating a myriad of aeroso... more Heavy duty gas turbine engines are not only ingesting the air, but also eating a myriad of aerosol particles, which may have various negative effects on the turbine operation efficiency as well as the component failure. This paper attempts to develop predictive degradation models for gas turbines by integrating satellite collected atmospheric factors, on-site monitoring data, and physics-based calculated performance results. Multiple variables are analyzed and employed for predictive modeling. The vital variables are identified by using data exploratory correlation analysis and stepwise regression analysis. The performance degradation calculation is obtained from physics based thermodynamic heat balance of gas turbine. It requires balancing mass and energy of gas turbine to match measurement data through thermodynamic cycle matching. The performance degradation prior to the offline water wash is used as the predictor. Artificial neural network modeling is employed to establish the predictive models. A procedure is presented to explain the proposed methodology, and results are discussed. This paper provides an effective methodology and procedure to apply big data for the performance degradation prediction of gas turbines.

System and method for hybrid modeling risk of turbomachinery

Systems and methods are disclosed herein to improve the operation of turbomachinery. The systems ... more Systems and methods are disclosed herein to improve the operation of turbomachinery. The systems and methods include a hybrid risk model (82). The hybrid risk model (82) includes a sub-model of a physical nature (78, 102, 206) and a random sub-model (80, 208). The sub-model of a physical nature (78, 102, 206) is configured to model the physical components of a turbomachine (10). Statistical sub-model (80, 208) is configured to model history information of the turbomachine (10). The hybrid risk model (82) is configured to calculate a turbomachine parameter (84, 86, 88, 202, 204, 208).

Fuzzy clustering approach for accurate embedding dimension identification in chaotic time series

Integrated Computer-Aided Engineering, 2003

Xiaomo had been heavily involved in real-time construction monitoring of a 450-meter cablestayed ... more Xiaomo had been heavily involved in real-time construction monitoring of a 450-meter cablestayed concrete bridge, assessing integral-lift steel scaffolds in high-rise building construction under wind loadings, and simulating nonlinear responses of space truss structures under wind excitations. As a research assistant in University of Wyoming from 2000 to 2001, Xiaomo Jiang had participated in investigating the effects of multiple moisture susceptibility testing, freeze-thaw cycles on mechanical properties and strength of Hot Mix Asphalt mixtures.

Bayesian Calibration for Power Splitting in Single Shaft Combined Cycle Plant Diagnostics

Volume 6: Ceramics; Controls, Diagnostics and Instrumentation; Education; Manufacturing Materials and Metallurgy; Honors and Awards, 2015

Condition monitoring and diagnostics of a combined cycle gas turbine (CCGT) power plant has becom... more Condition monitoring and diagnostics of a combined cycle gas turbine (CCGT) power plant has become an important tool to improve its availability, reliability, and performance. However, there are two major challenges in the diagnostics of performance degradation and anomaly in a single-shaft combined cycle (CC) power plant. First, since the gas turbine (GT) and steam turbine (ST) in such a plant share a common generator, each turbine&amp;#39;s contribution to the total plant power output is not directly measured, but must be accurately estimated to identify the possible causes of plant level degradation. Second, multivariate operational data instrumented from a power plant need to be used in the plant model calibration, power splitting, and degradation diagnostics. Sensor data always contain some degree of uncertainty. This adds to the difficulty of both estimation of GT to ST power split (PS) and degradation diagnostics. This paper presents an integrated probabilistic methodology for accurate power splitting and the degradation diagnostics of a single-shaft CC plant, accounting for uncertainties in the measured data. The method integrates the Bayesian inference approach, thermodynamic physics modeling, and sensed operational data seamlessly. The physics-based thermodynamic heat balance model is first established to model the power plant components and their thermodynamic relationships. The model is calibrated to model the plant performance at the design conditions of its main components. The calibrated model is then employed to simulate the plant performance at various operating conditions. A Bayesian inference method is next developed to determine the PS between the GT and the ST by comparing the measured and expected power outputs at different operation conditions, considering uncertainties in multiple measured variables. The calibrated model and calculated PS are further applied to pinpoint the possible causes at individual components resulting in the plant level degradation. The proposed methodology is demonstrated using operational data from a real-world single-shaft CC power plant with a known degradation issue. This study provides an effective probabilistic methodology to accurately split the power for degradation diagnostics of a single-shaft CC plant, addressing the uncertainties in multiple measured variables.

Remote Thermal Performance Monitoring and Diagnostics: Turning Data Into Knowledge

Volume 2: Reliability, Availability and Maintainability (RAM); Plant Systems, Structures, Components and Materials Issues; Simple and Combined Cycles; Advanced Energy Systems and Renewables (Wind, Solar and Geothermal); Energy Water Nexus; Thermal Hydraulics and CFD; Nuclear Plant Design, Licensi..., 2013

Gas turbine simple or combined cycle plants are built and operated with higher availability, reli... more Gas turbine simple or combined cycle plants are built and operated with higher availability, reliability, and performance in order to provide the customer with sufficient operating revenues and reduced fuel costs meanwhile enhancing customer dispatch competitiveness. A tremendous amount of operational data is usually collected from the everyday operation of a power plant. It has become an increasingly important but challenging issue about how to turn this data into knowledge and further solutions via developing advanced state-of-the-art analytics. This paper presents an integrated system and methodology to pursue this purpose by automating multi-level, multi-paradigm, multi-facet performance monitoring and anomaly detection for heavy duty gas turbines. The system provides an intelligent platform to drive site-specific performance improvements, mitigate outage risk, rationalize operational pattern, and enhance maintenance schedule and service offerings via taking appropriate proactive actions. In addition, the paper also presents the components in the system, including data sensing, hardware, and operational anomaly detection, expertise proactive act of company, site specific degradation assessment, and water wash effectiveness monitoring and analytics. As demonstrated in two examples, this remote performance monitoring aims to improve equipment efficiency by converting data into knowledge and solutions in order to drive value for customers including lowering operating fuel cost and increasing customer power sales and life cycle value.Copyright © 2013 by ASME

Towards Predictive Maintenance of a Heavy-Duty Gas Turbine A New Hybrid Intelligent Methodology for Performance Simulation

Annual Conference of the PHM Society

As the increasing demands of global clean power for the main purpose of lowering environmental po... more As the increasing demands of global clean power for the main purpose of lowering environmental pollution, heavy-duty gas turbines are playing an increasingly important role in energy fields because of their low emission, high thermal efficiency and flexible start-up capacity. Accurate modeling and simulation of the turbine performance is extremely needed to precisely design a large-watt industrial heavy-duty gas turbine, which is still very challenging due to the high nonlinearity, system complexity, varying conditions, and strong coupling interaction of high-dimension parameters under the harsh operation environment of the turbine. In order to improve the accuracy and efficiency of the performance simulation model for a heavy-duty gas turbine at various operating conditions, this paper presents a new hybrid intelligent methodology to adeptly integrate system thermodynamic balance mechanism with multivariate data by seamlessly combing advanced signal processing, machine learning and...

Validation Using Bayesian Methods

Simulation Foundations, Methods and Applications, 2019

Quantitative model validation is playing an increasingly important role in performance and reliab... more Quantitative model validation is playing an increasingly important role in performance and reliability assessment of a complicated system whenever simulation is involved. This chapter discusses model validation from a Bayesian perspective, considering in particular data uncertainty. First, Bayes’ theorem is defined, then the Bayesian risk rule method is introduced. Explicit expressions for the Bayesian interval hypothesis testing approach are presented in both univariate and multivariate cases. The problem of non-normal validation data is addressed by the Box–Cox transformation. A generalized procedure is presented to implement Bayesian validation methods. Classic hypothesis testing method is utilized to conduct a comparison study. The impact of the data normality assumption and of the variation of the threshold on model assessment accuracy is investigated by using both classical and Bayesian approaches. The Bayesian methodology is illustrated with a reliability model of rotor blade...

Multivariate probabilistic modelling for seepage risk assessment in tunnel segments

International Journal of Reliability and Safety, 2014

Seepage in the lining structure of a tunnel has become an increasingly significant issue in under... more Seepage in the lining structure of a tunnel has become an increasingly significant issue in underground engineering, which impacts the performance and service life of the tunnel structure. This paper presents a generic multivariate probabilistic macro model for reliability assessment of seepage behaviour in tunnel segments consisting of pre-manufactured concrete lining rings. The model is characterised by multiple influencing factors including the tunnel line, rail operating direction, seepage joint location and tunnel operational time. A multivariate statistical approach, proportional hazard modelling technique, is exploited to empirically model the seepage occurrence probability along the tunnel segment length, meanwhile integrating impacts from different influencing factors on the tunnel section seepage. The most commonly used Weibull and log-normal distributions are employed in the development of the proportional hazards modelling. The methodology is illustrated with the seepage data collected from real-world subway tunnels. Seepage occurrence probability is also assessed over the critical factors.

Physics Informed Machine Learning Approach for Performance Degradation Monitoring of Gas Turbine

PHM Society Asia-Pacific Conference

The heavy-duty gas turbine is playing an increasingly significant role on power generation due to... more The heavy-duty gas turbine is playing an increasingly significant role on power generation due to its lower-emission, higher flexibility and thermo-efficiency. Main subsystems of the gas turbine like compressor, combustor and turbine degrade over the operating time under the harsh environmental conditions, which largely impacts the efficiency and productivity of the system. Therefore, it is critical to develop effective approaches to monitor performance degradation of a heavy-duty gas turbine for system predictive maintenance thus improving the efficiency and productivity of the machine. This paper presents a new physics informed machine learning methodology to predict the degradation of gas turbine by seamlessly integrating thermodynamic heat balancing mechanism, component characteristics, multi-source data and artificial neural network model. The mechanism-based thermodynamic model is established for multiple subsystems considering the balance of flow, mass and energy, and then in...

Integrated Bayesian probabilistic approach to improve predictive modelling

International Journal of Reliability and Safety, 2017

This paper presents an integrated Bayesian probabilistic methodology to calibrate parameters of a... more This paper presents an integrated Bayesian probabilistic methodology to calibrate parameters of a predictive model and quantitatively evaluate its validity and predictive capacity with non-normality data, considering uncertainties. Bayes network is developed to graphically represent the relationships of all model variables. Bayesian regression theory associated with Markov Chain Monte Carlo simulation and Gibbs sampling is presented to calibrate model parameters to improve model accuracy. The Bayesian method

Friction modeling and analysis of injection process in squeeze casting

Journal of Materials Processing Technology, 2017

Abstract An analytic model is presented for quantitatively investigating the impact of friction b... more Abstract An analytic model is presented for quantitatively investigating the impact of friction between the punch and the shot sleeve on dynamic match status during the injection process of squeeze casting, with the ultimate purpose of optimizing machine design and controlling injection process more accurately. By combining lubrication model and friction experiment, the friction characteristics between the punch and the shot sleeve are investigated in four cases of different clearance values. The lubrication model is validated by comparing the calculated and experimental values of the friction coefficient. The friction model of injection process is then established by means of updating incremental boundary conditions iteratively. The temperature and deformation data are collected from laboratory testing and used to compare with the simulation data obtained from both the general model and the proposed friction model. The simulation data from the friction model is approximately in agreement with the experimental data, better than the general model, implying that the proposed analytics model provides more accurate simulation of the friction coefficients.

Temperature Uncertainty Analysis of Injection Mechanism Based on Kriging Modeling

Materials (Basel, Switzerland), Jan 17, 2017

A kriging modeling method is proposed to conduct the temperature uncertainty analysis of an injec... more A kriging modeling method is proposed to conduct the temperature uncertainty analysis of an injection mechanism in squeeze casting. A mathematical model of temperature prediction with multi input and single output is employed to estimate the temperature spatiotemporal distributions of the injection mechanism. The kriging model applies different weights to the independent variables according to spatial location of sample points and their correlation, thus reducing the estimation variance. The predicted value of the kriging model is compared with the sample data at the corresponding position to investigate the influence of the temperature uncertainty of the injection mechanism on the injection process including friction. The results indicate that the significant error is observed at a few sample points in the early injection due to the impact of the uncertainty facts. The variance mean and standard deviation obtained by the model calibrated by experimental samples reduce largely in co...

Bayesian kriging modeling for spatiotemporal prediction in squeeze casting

The International Journal of Advanced Manufacturing Technology, 2016

The deterioration of a shot sleeve in squeeze casting due to thermo-mechanical fatigue often resu... more The deterioration of a shot sleeve in squeeze casting due to thermo-mechanical fatigue often results in lowering the reliability and availability of the squeeze casting machine, thus reducing its productivity, meanwhile increasing the lifecycle maintenance cost. This paper presents an efficient Bayesian kriging meta-modeling method for spatiotemporal prediction under data uncertainty and non-normality, with the target applications for controlling the deformation, optimizing machine design, and predicting component fatigue cracking thus improving the reliability and availability of mechanical system. Spatiotemporal kriging model is established to substitute the complicated computer model by using numerically simulated data. Bayesian probabilistic approach is then developed to quantitatively evaluate the validity and predictive capacity of kriging meta-model, considering data uncertainty. The Anderson-Darling goodness-of-fit test is employed to perform the normality hypothesis test of difference values of validation data. Box-Cox transformation method is utilized to convert the non-normality data with the purpose of facilitating the overall validation assessment of meta-models with higher accuracy. Bayesian confidence measure is presented to quantify the confidence on the predictive capacity of the kriging meta-models, given the transformed data. A procedure is proposed to implement the proposed probabilistic methodology for meta-modeling and model validation with non-normality response series. The impact of data normality assumption and decision threshold parameter in quantitative model assessment is also investigated by using Bayesian inference approach. The effectiveness of the proposed methodology and procedure is demonstrated with the spatiotemporal temperature prediction in squeeze casting.

Deep probabilistic time series forecasting using augmented recurrent input for dynamic systems

Mechanical Systems and Signal Processing

The demand of probabilistic time series forecasting has been recently raised in various dynamic s... more The demand of probabilistic time series forecasting has been recently raised in various dynamic system scenarios, for example, system identification and prognostic and health management of machines. To this end, we combine the advances in both deep generative models and state space model (SSM) to come up with a novel, data-driven deep probabilistic sequence model. Specially, we follow the popular encoder-decoder generative structure to build the recurrent neural networks (RNN) assisted variational sequence model on an augmented recurrent input space, which could induce rich stochastic sequence dependency. Besides, in order to alleviate the issue of inconsistency between training and predicting as well as improving the mining of dynamic patterns, we (i) propose using a hybrid output as input at next time step, which brings training and predicting into alignment; and (ii) further devise a generalized auto-regressive strategy that encodes all the historical dependencies at current time step. Thereafter, we first investigate the methodological characteristics of the proposed deep probabilistic sequence model on toy cases, and then comprehensively demonstrate the superiority of our model against existing deep probabilistic SSM models through extensive numerical experiments on eight system identification benchmarks from various dynamic systems. Finally, we apply our sequence model to a real-world centrifugal compressor sensor data forecasting problem, and again verify its outstanding performance by quantifying the time series predictive distribution.

Towards Smart Condition Monitoring of Rotatory Machines: An Optimized Probabilistic Signal Reconstruction Methodology for Fault Prediction with Multi-source Uncertainties

IEEE Access

A Similarity Based Modeling Approach for Turbomachine Fault Prediction

Faults in the critical components of a turbomachine usually result in unplanned outage, leading t... more Faults in the critical components of a turbomachine usually result in unplanned outage, leading to huge loss of properties and life. Condition monitoring becomes a promising tool to provide automatic early alerting of potential damage in critical components thus ensuring the system safety and reliability while lowering its maintenance cost. This is still a challenging hot topic due to the data imperfection and multivariate correlation, as well as the variation of faults and components in different turbomachines. In this paper, a condition monitoring method based on similarity-based model is proposed to solve these problems in fault prediction of large turbine machinery. Bayesian wavelet multi-scale reconstruction is proposed to address the potential noise in the sensed multivariate time historical data. The advanced signal processing balances the over-denoising and under-denoising of raw multivariate signals. An optimized auto-associative kernel regression (OAKR) approach is develop...

Plant Performance Monitoring and Diagnostics: Remote, Real-Time and Automation

Volume 6: Ceramics; Controls, Diagnostics and Instrumentation; Education; Manufacturing Materials and Metallurgy, 2014

Combined cycle gas turbine plants are built and operated with higher availability, reliability, a... more Combined cycle gas turbine plants are built and operated with higher availability, reliability, and performance than simple cycle in order to help provide the customer with capabilities to generate operating revenues and reduce fuel costs while enhancing dispatch competitiveness. The availability of a power plant can be improved by increasing the reliability of individual assets through maintenance enhancement and performance degradation recovery through remote efficiency monitoring to provide timely corrective recommendations. This paper presents a comprehensive system and methodology to pursue this purpose by using instrumented data to automate performance modeling for real-time monitoring and anomaly detection of combined cycle gas turbine power plants. Through thermodynamic performance modeling of main assets in a power plant such as gas turbines, steam turbines, heat recovery steam generators, condensers and other auxiliaries, the system provides an intelligent platform and met...

improved OAKR approach to condition monitoring of rotating machinery

PHM Society Asia-Pacific Conference

Faults in main subsystems or components of a rotating machine often causes unscheduled shutdown, ... more Faults in main subsystems or components of a rotating machine often causes unscheduled shutdown, which may lead to not only huge economic losses, but also safety accidents. As an important part of intelligent maintenance, condition monitoring becomes a powerful tool in reducing maintenance costs through automatic fault alarming, thereby reducing potential downtime while improving system safety and reliability. An optimized auto-associative kernelregression (OAKR) model has been proposed recently and demonstrated as a promising tool for condition monitoring of various turbomachines, which is independent of fault mode and machine type. However, the fault identification accuracy of this approach largely relies on data quality in practical applications. Data incompleteness, parameter variation and system complexity often result in the inaccuracy of fault alarming for complicated rotating machinery. This paper proposes an improved OAKR method to address these issues, including utilizing ...

Tension Force Estimation of Cables with Two Intermediate Supports

International Journal of Structural Stability and Dynamics, Feb 19, 2020

The presence of intermediate supports usually imposes difficulties in identifying the tension for... more The presence of intermediate supports usually imposes difficulties in identifying the tension force of stayed cables in cable-stayed bridges or hanger cables in arch bridges. This paper establishes the partial differential equations of motion of the cable and derives two numerical models with (Model 1) and without (Model 2) considering the flexural rigidity. The effects of two intermediate supports on the identification accuracy of the cable tension force are further studied analytically and experimentally. The effects of several non-dimensional parameters (e.g. damper location, support stiffness, flexural rigidity, and mode order of the cable) on the identification accuracy of the models are also investigated. It is theoretically concluded that the simplified Model 2 provides acceptable accuracy on tension force identification when the non-dimensional parameter [Formula: see text] is greater than 90 (slender cables), whereas the accurate Model 1 can be applied for tension force identification at any scenarios. The feasibility of two models is further verified by three numerical examples and field tests on two real-world arch bridges.

Mode shape–aided cable force determination using digital image correlation

Structural Health Monitoring-an International Journal, Sep 22, 2020

Assuming the distance between two nodal points of a specific cable vibration mode as the effectiv... more Assuming the distance between two nodal points of a specific cable vibration mode as the effective length of a pinned–pinned cable, mode shape–aided cable tension estimation methods are employed to estimate the cable force. This article proposes a framework based on digital image correlation technique for remote measurement of the dynamic displacement time history of cables in cable structures. Frequency domain decomposition technique is then used to extract the cable natural frequencies and mode shapes. Identified cable mode shapes are used along with a tensioned pinned–pinned cable model to estimate the cable force. Accuracy of the proposed methodology is investigated using the experimental data coming from a laboratory-scale test setup and hanger cables of a real-world arch bridge.

Enhancing Turbine Performance Degradation Prediction with Atmospheric Factors

Annual Conference of the PHM Society, 2016

Heavy duty gas turbine engines are not only ingesting the air, but also eating a myriad of aeroso... more Heavy duty gas turbine engines are not only ingesting the air, but also eating a myriad of aerosol particles, which may have various negative effects on the turbine operation efficiency as well as the component failure. This paper attempts to develop predictive degradation models for gas turbines by integrating satellite collected atmospheric factors, on-site monitoring data, and physics-based calculated performance results. Multiple variables are analyzed and employed for predictive modeling. The vital variables are identified by using data exploratory correlation analysis and stepwise regression analysis. The performance degradation calculation is obtained from physics based thermodynamic heat balance of gas turbine. It requires balancing mass and energy of gas turbine to match measurement data through thermodynamic cycle matching. The performance degradation prior to the offline water wash is used as the predictor. Artificial neural network modeling is employed to establish the predictive models. A procedure is presented to explain the proposed methodology, and results are discussed. This paper provides an effective methodology and procedure to apply big data for the performance degradation prediction of gas turbines.

System and method for hybrid modeling risk of turbomachinery

Systems and methods are disclosed herein to improve the operation of turbomachinery. The systems ... more Systems and methods are disclosed herein to improve the operation of turbomachinery. The systems and methods include a hybrid risk model (82). The hybrid risk model (82) includes a sub-model of a physical nature (78, 102, 206) and a random sub-model (80, 208). The sub-model of a physical nature (78, 102, 206) is configured to model the physical components of a turbomachine (10). Statistical sub-model (80, 208) is configured to model history information of the turbomachine (10). The hybrid risk model (82) is configured to calculate a turbomachine parameter (84, 86, 88, 202, 204, 208).

Fuzzy clustering approach for accurate embedding dimension identification in chaotic time series

Integrated Computer-Aided Engineering, 2003

Xiaomo had been heavily involved in real-time construction monitoring of a 450-meter cablestayed ... more Xiaomo had been heavily involved in real-time construction monitoring of a 450-meter cablestayed concrete bridge, assessing integral-lift steel scaffolds in high-rise building construction under wind loadings, and simulating nonlinear responses of space truss structures under wind excitations. As a research assistant in University of Wyoming from 2000 to 2001, Xiaomo Jiang had participated in investigating the effects of multiple moisture susceptibility testing, freeze-thaw cycles on mechanical properties and strength of Hot Mix Asphalt mixtures.

Bayesian Calibration for Power Splitting in Single Shaft Combined Cycle Plant Diagnostics

Volume 6: Ceramics; Controls, Diagnostics and Instrumentation; Education; Manufacturing Materials and Metallurgy; Honors and Awards, 2015

Condition monitoring and diagnostics of a combined cycle gas turbine (CCGT) power plant has becom... more Condition monitoring and diagnostics of a combined cycle gas turbine (CCGT) power plant has become an important tool to improve its availability, reliability, and performance. However, there are two major challenges in the diagnostics of performance degradation and anomaly in a single-shaft combined cycle (CC) power plant. First, since the gas turbine (GT) and steam turbine (ST) in such a plant share a common generator, each turbine&amp;#39;s contribution to the total plant power output is not directly measured, but must be accurately estimated to identify the possible causes of plant level degradation. Second, multivariate operational data instrumented from a power plant need to be used in the plant model calibration, power splitting, and degradation diagnostics. Sensor data always contain some degree of uncertainty. This adds to the difficulty of both estimation of GT to ST power split (PS) and degradation diagnostics. This paper presents an integrated probabilistic methodology for accurate power splitting and the degradation diagnostics of a single-shaft CC plant, accounting for uncertainties in the measured data. The method integrates the Bayesian inference approach, thermodynamic physics modeling, and sensed operational data seamlessly. The physics-based thermodynamic heat balance model is first established to model the power plant components and their thermodynamic relationships. The model is calibrated to model the plant performance at the design conditions of its main components. The calibrated model is then employed to simulate the plant performance at various operating conditions. A Bayesian inference method is next developed to determine the PS between the GT and the ST by comparing the measured and expected power outputs at different operation conditions, considering uncertainties in multiple measured variables. The calibrated model and calculated PS are further applied to pinpoint the possible causes at individual components resulting in the plant level degradation. The proposed methodology is demonstrated using operational data from a real-world single-shaft CC power plant with a known degradation issue. This study provides an effective probabilistic methodology to accurately split the power for degradation diagnostics of a single-shaft CC plant, addressing the uncertainties in multiple measured variables.

Remote Thermal Performance Monitoring and Diagnostics: Turning Data Into Knowledge

Volume 2: Reliability, Availability and Maintainability (RAM); Plant Systems, Structures, Components and Materials Issues; Simple and Combined Cycles; Advanced Energy Systems and Renewables (Wind, Solar and Geothermal); Energy Water Nexus; Thermal Hydraulics and CFD; Nuclear Plant Design, Licensi..., 2013

Gas turbine simple or combined cycle plants are built and operated with higher availability, reli... more Gas turbine simple or combined cycle plants are built and operated with higher availability, reliability, and performance in order to provide the customer with sufficient operating revenues and reduced fuel costs meanwhile enhancing customer dispatch competitiveness. A tremendous amount of operational data is usually collected from the everyday operation of a power plant. It has become an increasingly important but challenging issue about how to turn this data into knowledge and further solutions via developing advanced state-of-the-art analytics. This paper presents an integrated system and methodology to pursue this purpose by automating multi-level, multi-paradigm, multi-facet performance monitoring and anomaly detection for heavy duty gas turbines. The system provides an intelligent platform to drive site-specific performance improvements, mitigate outage risk, rationalize operational pattern, and enhance maintenance schedule and service offerings via taking appropriate proactive actions. In addition, the paper also presents the components in the system, including data sensing, hardware, and operational anomaly detection, expertise proactive act of company, site specific degradation assessment, and water wash effectiveness monitoring and analytics. As demonstrated in two examples, this remote performance monitoring aims to improve equipment efficiency by converting data into knowledge and solutions in order to drive value for customers including lowering operating fuel cost and increasing customer power sales and life cycle value.Copyright © 2013 by ASME

Towards Predictive Maintenance of a Heavy-Duty Gas Turbine A New Hybrid Intelligent Methodology for Performance Simulation

Annual Conference of the PHM Society

As the increasing demands of global clean power for the main purpose of lowering environmental po... more As the increasing demands of global clean power for the main purpose of lowering environmental pollution, heavy-duty gas turbines are playing an increasingly important role in energy fields because of their low emission, high thermal efficiency and flexible start-up capacity. Accurate modeling and simulation of the turbine performance is extremely needed to precisely design a large-watt industrial heavy-duty gas turbine, which is still very challenging due to the high nonlinearity, system complexity, varying conditions, and strong coupling interaction of high-dimension parameters under the harsh operation environment of the turbine. In order to improve the accuracy and efficiency of the performance simulation model for a heavy-duty gas turbine at various operating conditions, this paper presents a new hybrid intelligent methodology to adeptly integrate system thermodynamic balance mechanism with multivariate data by seamlessly combing advanced signal processing, machine learning and...

Validation Using Bayesian Methods

Simulation Foundations, Methods and Applications, 2019

Quantitative model validation is playing an increasingly important role in performance and reliab... more Quantitative model validation is playing an increasingly important role in performance and reliability assessment of a complicated system whenever simulation is involved. This chapter discusses model validation from a Bayesian perspective, considering in particular data uncertainty. First, Bayes’ theorem is defined, then the Bayesian risk rule method is introduced. Explicit expressions for the Bayesian interval hypothesis testing approach are presented in both univariate and multivariate cases. The problem of non-normal validation data is addressed by the Box–Cox transformation. A generalized procedure is presented to implement Bayesian validation methods. Classic hypothesis testing method is utilized to conduct a comparison study. The impact of the data normality assumption and of the variation of the threshold on model assessment accuracy is investigated by using both classical and Bayesian approaches. The Bayesian methodology is illustrated with a reliability model of rotor blade...

Multivariate probabilistic modelling for seepage risk assessment in tunnel segments

International Journal of Reliability and Safety, 2014

Seepage in the lining structure of a tunnel has become an increasingly significant issue in under... more Seepage in the lining structure of a tunnel has become an increasingly significant issue in underground engineering, which impacts the performance and service life of the tunnel structure. This paper presents a generic multivariate probabilistic macro model for reliability assessment of seepage behaviour in tunnel segments consisting of pre-manufactured concrete lining rings. The model is characterised by multiple influencing factors including the tunnel line, rail operating direction, seepage joint location and tunnel operational time. A multivariate statistical approach, proportional hazard modelling technique, is exploited to empirically model the seepage occurrence probability along the tunnel segment length, meanwhile integrating impacts from different influencing factors on the tunnel section seepage. The most commonly used Weibull and log-normal distributions are employed in the development of the proportional hazards modelling. The methodology is illustrated with the seepage data collected from real-world subway tunnels. Seepage occurrence probability is also assessed over the critical factors.

Physics Informed Machine Learning Approach for Performance Degradation Monitoring of Gas Turbine

PHM Society Asia-Pacific Conference

The heavy-duty gas turbine is playing an increasingly significant role on power generation due to... more The heavy-duty gas turbine is playing an increasingly significant role on power generation due to its lower-emission, higher flexibility and thermo-efficiency. Main subsystems of the gas turbine like compressor, combustor and turbine degrade over the operating time under the harsh environmental conditions, which largely impacts the efficiency and productivity of the system. Therefore, it is critical to develop effective approaches to monitor performance degradation of a heavy-duty gas turbine for system predictive maintenance thus improving the efficiency and productivity of the machine. This paper presents a new physics informed machine learning methodology to predict the degradation of gas turbine by seamlessly integrating thermodynamic heat balancing mechanism, component characteristics, multi-source data and artificial neural network model. The mechanism-based thermodynamic model is established for multiple subsystems considering the balance of flow, mass and energy, and then in...

Integrated Bayesian probabilistic approach to improve predictive modelling

International Journal of Reliability and Safety, 2017

This paper presents an integrated Bayesian probabilistic methodology to calibrate parameters of a... more This paper presents an integrated Bayesian probabilistic methodology to calibrate parameters of a predictive model and quantitatively evaluate its validity and predictive capacity with non-normality data, considering uncertainties. Bayes network is developed to graphically represent the relationships of all model variables. Bayesian regression theory associated with Markov Chain Monte Carlo simulation and Gibbs sampling is presented to calibrate model parameters to improve model accuracy. The Bayesian method

Friction modeling and analysis of injection process in squeeze casting

Journal of Materials Processing Technology, 2017

Abstract An analytic model is presented for quantitatively investigating the impact of friction b... more Abstract An analytic model is presented for quantitatively investigating the impact of friction between the punch and the shot sleeve on dynamic match status during the injection process of squeeze casting, with the ultimate purpose of optimizing machine design and controlling injection process more accurately. By combining lubrication model and friction experiment, the friction characteristics between the punch and the shot sleeve are investigated in four cases of different clearance values. The lubrication model is validated by comparing the calculated and experimental values of the friction coefficient. The friction model of injection process is then established by means of updating incremental boundary conditions iteratively. The temperature and deformation data are collected from laboratory testing and used to compare with the simulation data obtained from both the general model and the proposed friction model. The simulation data from the friction model is approximately in agreement with the experimental data, better than the general model, implying that the proposed analytics model provides more accurate simulation of the friction coefficients.

Temperature Uncertainty Analysis of Injection Mechanism Based on Kriging Modeling

Materials (Basel, Switzerland), Jan 17, 2017

A kriging modeling method is proposed to conduct the temperature uncertainty analysis of an injec... more A kriging modeling method is proposed to conduct the temperature uncertainty analysis of an injection mechanism in squeeze casting. A mathematical model of temperature prediction with multi input and single output is employed to estimate the temperature spatiotemporal distributions of the injection mechanism. The kriging model applies different weights to the independent variables according to spatial location of sample points and their correlation, thus reducing the estimation variance. The predicted value of the kriging model is compared with the sample data at the corresponding position to investigate the influence of the temperature uncertainty of the injection mechanism on the injection process including friction. The results indicate that the significant error is observed at a few sample points in the early injection due to the impact of the uncertainty facts. The variance mean and standard deviation obtained by the model calibrated by experimental samples reduce largely in co...

Bayesian kriging modeling for spatiotemporal prediction in squeeze casting

The International Journal of Advanced Manufacturing Technology, 2016

The deterioration of a shot sleeve in squeeze casting due to thermo-mechanical fatigue often resu... more The deterioration of a shot sleeve in squeeze casting due to thermo-mechanical fatigue often results in lowering the reliability and availability of the squeeze casting machine, thus reducing its productivity, meanwhile increasing the lifecycle maintenance cost. This paper presents an efficient Bayesian kriging meta-modeling method for spatiotemporal prediction under data uncertainty and non-normality, with the target applications for controlling the deformation, optimizing machine design, and predicting component fatigue cracking thus improving the reliability and availability of mechanical system. Spatiotemporal kriging model is established to substitute the complicated computer model by using numerically simulated data. Bayesian probabilistic approach is then developed to quantitatively evaluate the validity and predictive capacity of kriging meta-model, considering data uncertainty. The Anderson-Darling goodness-of-fit test is employed to perform the normality hypothesis test of difference values of validation data. Box-Cox transformation method is utilized to convert the non-normality data with the purpose of facilitating the overall validation assessment of meta-models with higher accuracy. Bayesian confidence measure is presented to quantify the confidence on the predictive capacity of the kriging meta-models, given the transformed data. A procedure is proposed to implement the proposed probabilistic methodology for meta-modeling and model validation with non-normality response series. The impact of data normality assumption and decision threshold parameter in quantitative model assessment is also investigated by using Bayesian inference approach. The effectiveness of the proposed methodology and procedure is demonstrated with the spatiotemporal temperature prediction in squeeze casting.

Deep probabilistic time series forecasting using augmented recurrent input for dynamic systems

Mechanical Systems and Signal Processing

The demand of probabilistic time series forecasting has been recently raised in various dynamic s... more The demand of probabilistic time series forecasting has been recently raised in various dynamic system scenarios, for example, system identification and prognostic and health management of machines. To this end, we combine the advances in both deep generative models and state space model (SSM) to come up with a novel, data-driven deep probabilistic sequence model. Specially, we follow the popular encoder-decoder generative structure to build the recurrent neural networks (RNN) assisted variational sequence model on an augmented recurrent input space, which could induce rich stochastic sequence dependency. Besides, in order to alleviate the issue of inconsistency between training and predicting as well as improving the mining of dynamic patterns, we (i) propose using a hybrid output as input at next time step, which brings training and predicting into alignment; and (ii) further devise a generalized auto-regressive strategy that encodes all the historical dependencies at current time step. Thereafter, we first investigate the methodological characteristics of the proposed deep probabilistic sequence model on toy cases, and then comprehensively demonstrate the superiority of our model against existing deep probabilistic SSM models through extensive numerical experiments on eight system identification benchmarks from various dynamic systems. Finally, we apply our sequence model to a real-world centrifugal compressor sensor data forecasting problem, and again verify its outstanding performance by quantifying the time series predictive distribution.

Towards Smart Condition Monitoring of Rotatory Machines: An Optimized Probabilistic Signal Reconstruction Methodology for Fault Prediction with Multi-source Uncertainties

IEEE Access

A Similarity Based Modeling Approach for Turbomachine Fault Prediction

Faults in the critical components of a turbomachine usually result in unplanned outage, leading t... more Faults in the critical components of a turbomachine usually result in unplanned outage, leading to huge loss of properties and life. Condition monitoring becomes a promising tool to provide automatic early alerting of potential damage in critical components thus ensuring the system safety and reliability while lowering its maintenance cost. This is still a challenging hot topic due to the data imperfection and multivariate correlation, as well as the variation of faults and components in different turbomachines. In this paper, a condition monitoring method based on similarity-based model is proposed to solve these problems in fault prediction of large turbine machinery. Bayesian wavelet multi-scale reconstruction is proposed to address the potential noise in the sensed multivariate time historical data. The advanced signal processing balances the over-denoising and under-denoising of raw multivariate signals. An optimized auto-associative kernel regression (OAKR) approach is develop...

Plant Performance Monitoring and Diagnostics: Remote, Real-Time and Automation

Volume 6: Ceramics; Controls, Diagnostics and Instrumentation; Education; Manufacturing Materials and Metallurgy, 2014

Combined cycle gas turbine plants are built and operated with higher availability, reliability, a... more Combined cycle gas turbine plants are built and operated with higher availability, reliability, and performance than simple cycle in order to help provide the customer with capabilities to generate operating revenues and reduce fuel costs while enhancing dispatch competitiveness. The availability of a power plant can be improved by increasing the reliability of individual assets through maintenance enhancement and performance degradation recovery through remote efficiency monitoring to provide timely corrective recommendations. This paper presents a comprehensive system and methodology to pursue this purpose by using instrumented data to automate performance modeling for real-time monitoring and anomaly detection of combined cycle gas turbine power plants. Through thermodynamic performance modeling of main assets in a power plant such as gas turbines, steam turbines, heat recovery steam generators, condensers and other auxiliaries, the system provides an intelligent platform and met...