On-Line Defect Detecting Method Based on Kernel Method

2011 ◽  
Vol 474-476 ◽  
pp. 858-863 ◽  
Author(s):  
Ke Jia Xu ◽  
Bin Chen ◽  
Li Zeng
Keyword(s):  
Real Time ◽  
Kernel Method ◽  
Principal Component ◽  
Kernel Principal Component Analysis ◽  
Image Method ◽  
Speed Up ◽  
On Line ◽  
And Performance ◽  
Detecting Method ◽  
Better Than

The conflict between accuracy and speed is one of the most well-known dilemmas of the real-time defect detecting system. This paper presents a real-time defect detecting algorithm based on Kernel principal component analysis (KPCA). KPCA-based feature extraction have recently shown to be very effective for image denoising, however the Normal KPCA method is time-consuming. In our method, we propose a progressive algorithm to speed up the reconstruct process while improve accuracy. Experimental results demonstrate that our method is dramatically better than Normal KPCA Pre-image method in terms of speed and performance.

A Hyperspectral Imaging System for Identification of Faecal and Ingesta Contamination on Poultry Carcasses

2003 ◽  
Vol 11 (4) ◽  
pp. 269-281 ◽  
Author(s):  
Kurt C. Lawrence ◽  
William R. Windham ◽  
Bosoon Park ◽  
R. Jeff Buhr
Keyword(s):  
Real Time ◽  
Hyperspectral Imaging ◽  
Near Infrared ◽  
Imaging System ◽  
Principal Component ◽  
Ccd Camera ◽  
Sample Population ◽  
Multivariate System ◽  
Ratio Image ◽  
On Line

A method and system for detecting faecal and ingesta contaminants on poultry carcasses were demonstrated. A visible/near infrared monochromator, which measured reflectance and principal component analysis were first used to identify key wavelengths from faecal and uncontaminated skin samples. Measurements at 434, 517, 565 and 628 nm were identified and used for evaluation with a hyperspectral imaging system. The hyperspectral imaging system, which was a line-scan (pushbroom) imaging system, consisted of a hyperspectral camera, fibre-optic line lights, a computer and frame grabber. The hyperspectral imaging camera consisted of a high-resolution charge coupled device (CCD) camera, a prism-grating-prism spectrograph, focusing lens, associated optical hardware and a motorised controller. The imaging system operated from about 400 to 900 nm. The hyperspectral imaging system was calibrated for wavelength, distance and percent reflectance and analysis of calibrated images at the key wavelengths indicated that single-wavelength images were inadequate for detecting contaminants. However, a ratio of images at two of the key wavelengths was able to identify faecal and ingesta contaminants. Specifically, the ratio of the 565-nm image divided by the 517-nm image produced good results. The ratio image was then further processed by masking the background and either enhancing the image contrast with a non-linear histogram stretch, or applying a faecal threshold. The results indicated that, for the limited sample population, more than 96% of the contaminants were detected. Thus, the hyperspectral imaging system was able to detect contaminants and showed feasibility, but was too slow for real-time on-line processing. Therefore, a multivariate system operating at 565 and 517 nm, which should be capable of operating at real-time on-line processing speed, should be used. Further research with such a system needs to be conducted.

Real-Time On-Line Performance Diagnostics of Heavy-Duty Industrial Gas Turbines

2002 ◽  
Vol 124 (4) ◽  
pp. 910-921 ◽  
Author(s):  
S. C. Gu¨len ◽  
P. R. Griffin ◽  
S. Paolucci
Keyword(s):  
Real Time ◽  
Gas Turbines ◽  
Performance Monitoring ◽  
Engine Performance ◽  
Combined Cycle ◽  
Ex Post ◽  
Ex Post Facto ◽  
Industrial Gas Turbines ◽  
On Line ◽  
And Performance

This paper describes the results of real-time, on-line performance monitoring of two gas turbines over a period of five months in 1997. A commercially available software system is installed to monitor, analyze and store measurements obtained from the plant’s distributed control system. The software is installed in a combined-cycle, cogeneration power plant, located in Massachusetts, USA, with two Frame 7EA gas turbines in Apr. 1997. Vendor’s information such as correction and part load performance curves are utilized to calculate expected engine performance and compare it with measurements. In addition to monitoring the general condition and performance of the gas turbines, user-specified financial data is used to determine schedules for compressor washing and inlet filter replacement by balancing the associated costs with lost revenue. All measurements and calculated information are stored in databases for real-time and historical trending and tabulating. The data is analyzed ex post facto to identify salient performance and maintenance issues.

An Ada-Based On-Line Computer System for AGR Power Stations

1993 ◽  
Vol 26 (3) ◽  
pp. 76-79
Author(s):  
R Clarke ◽  
F Humphries
Keyword(s):  
Real Time ◽  
Computer System ◽  
Relational Database ◽  
High Performance ◽  
Computer Systems ◽  
Case Tools ◽  
Power Stations ◽  
On Line ◽  
Replacement System ◽  
And Performance

Nuclear Electric have commissioned Ferranti International to develop a replacement system for their on-line computer systems in earlier AGR power stations. This system represents a step forward both in software and performance and takes advantage of investment made for military use in the Ada Language, CASE tools and a high-performance real-time relational database.

scholarly journals KPCA-ESN Soft-Sensor Model of Polymerization Process Optimized by Biogeography-Based Optimization Algorithm

2015 ◽  
Vol 2015 ◽  
pp. 1-10 ◽  
Author(s):  
Wen-hua Cui ◽  
Jie-sheng Wang ◽  
Shu-xia Li
Keyword(s):  
Real Time ◽  
Optimization Algorithm ◽  
Conversion Rate ◽  
Principal Component ◽  
Soft Sensor ◽  
Kernel Principal Component Analysis ◽  
Polymerization Process ◽  
Real Time Control ◽  
Sensor Model ◽  
Input Variables

For solving the problem that the conversion rate of vinyl chloride monomer (VCM) is hard for real-time online measurement in the polyvinyl chloride (PVC) polymerization production process, a soft-sensor modeling method based on echo state network (ESN) is put forward. By analyzing PVC polymerization process ten secondary variables are selected as input variables of the soft-sensor model, and the kernel principal component analysis (KPCA) method is carried out on the data preprocessing of input variables, which reduces the dimensions of the high-dimensional data. Thek-means clustering method is used to divide data samples into several clusters as inputs of each submodel. Then for each submodel the biogeography-based optimization algorithm (BBOA) is used to optimize the structure parameters of the ESN to realize the nonlinear mapping between input and output variables of the soft-sensor model. Finally, the weighted summation of outputs of each submodel is selected as the final output. The simulation results show that the proposed soft-sensor model can significantly improve the prediction precision of conversion rate and conversion velocity in the process of PVC polymerization and can satisfy the real-time control requirement of the PVC polymerization process.

Boiler Emissions and Performance Improvement due to Utilization Real-Time Intelligent Sootblowing Boiler Monitoring Place

2020 ◽  
Author(s):  
B. Chudnovsky ◽  
I. Chatskiy
Keyword(s):  
Heat Transfer ◽  
Power Plant ◽  
Real Time ◽  
Performance Improvement ◽  
Heat Balance ◽  
Integrated System ◽  
Boiler Efficiency ◽  
On Line ◽  
And Performance ◽  
Emissions And Performance

Abstract As it is well known, deposits in boilers contribute to boiler inefficiency, capacity reductions, and overheated tubes, which lead to tube failures. To improve the heat transfer inside the furnace the fouling deposits obviously should be removed. In order to take fouling into account in the overall furnace and boiler heat balance it is necessary to measure two main parameters — thickness of the deposits and their reflectivity (emissivity) in the wavelength of visible and IR region. In the present paper it is demonstrated how such measurement (see detailed description in Ref [1–3] can be used for on-line automatic sootblowing control. Results of our study demonstrate that dynamics of both parameters (contamination thickness and reflectivity) on the operated boiler can be registered in real time and then interpreted separately. The sootblowing boiler monitoring has been implemented at the 550 MW unit equipped with B&W opposite wall burners. The fouling and thickness sensors (FTR) were installed in different locations of the combustion chamber through its width and height. It was shown that dynamics of thickness and reflectivity variation just after the wall cleaning activation are quite different. Situations have been registered where changes of reflectivity have a significant impact on heat transfer, comparable and sometimes even greater than that of growing fouling thickness. Technique and device exploited in this study appears to be a very useful tool for sootblowing optimization and, as a result, for improvement of boiler efficiency and reduction of water wall erosion and corrosion. The paper presents a strategy to implement a comprehensive automatic control of soot blowing in power plant boilers. The paper will describe the existing installations where individual components are in operation, and describe an integrated system that could combine all these parts to make an integrated intelligent sootblowing system.

Sign in / Sign up

Export Citation Format

Share Document