Abstract:Power plant on-line measured operational data are often corrupted with random and gross errors. The data reconciliation method can reduce the impact of random errors by adjusting redundant measurements to satisfy system constraints and detect gross errors together with a statistical test method. In previous studies, the data reconciliation method is mainly used to deal with measurements with random and gross errors, and its application is mainly in the data preprocessing areas. In this work, we extend the data reconciliation and gross error detection method to cover both sensor and equipment performance monitoring in power plants, through introducing equipment characteristic constraints together with characteristic parameter dominant factor models in the data reconciliation method. The validity and capability of the proposed framework are illustrated with case studies in the feed water regenerative heating system of a 1000 MW ultra-supercritical coal-fired power generation unit. Case study results show that the characteristic parameter dominant factor models have relative root mean squared errors smaller than 2.3%, whilst the distribution properties of the test statistics for the integrated sensor and equipment performance monitoring are validated with simulated test statistic samples. We also illustrate that the proposed framework can efficiently detect and identify both sensor biases and equipment faults in the system. At the same time, the ability of the data reconciliation and global test method for measurement gross error detection is also improved due to the increased system redundancy under the proposed framework. (C) 2014 Elsevier Ltd. All rights reserved.

Plant-wide Mass Balance Using Extended Support Vector Regression Based Data Reconciliation and Gross Error Detection

A MT-NT-MILP Combined Method for Gross Error Detection and Data Reconciliation

Production Tracking and Mass Balance in a Refinery Based on Manufacturing Operations Management and Data Reconciliation

A New Method to Solve Robust Data Reconciliation in Nonlinear Process

Support Vector Regression Approach for Simultaneous Data Reconciliation and Gross Error or Outlier Detection

Simultaneous Data Reconciliation and Joint Bias and Leak Estimation Based on Support Vector Regression

An NT-MT Combined Method for Gross Error Detection and Data Reconciliation

A Study of Gross Error Detection and Data Reconciliation

Steady Data Reconciliation And Gross Error Detection Based On The Assumption Of Bounded Error Distribution

A Framework for Multi-layer Data Reconciliation in Chemical Process Systems

Data Reconciliation Based on Large Scale Rigorous Model with Equation Oriented Method

Methodology of data reconciliation and parameter estimation for variable load in process system

Correntropy based data reconciliation and gross error detection for bilinear systems

Multiscale Data Rectification Method with Application to Material Balance in Petrochemical Enterprises

Industrial Processes: Data Reconciliation and Gross Error Detection

Application of Data Reconcillation Method in Material Balance in Process Industry

Data Reconciliation and Gross Error Detection for Operational Data in Power Plants

Evaluating Material Balance System Based on Simulation

New Approach to Data Rectification of Hybrid Systems

A Data Reconciliation Based Framework for Integrated Sensor and Equipment Performance Monitoring in Power Plants

INDUSTRIAL APPLICATION OF DATA RECONCILIATION FOR HYBRID SYSTEMS