Papers by Keyword: Chi-Square Distribution

Paper TitlePage

Authors: Jie Su, Xu Guang Wang
Abstract: The acquisition of steady-state data is the premise for optimizing the boiler combustion parameters. Assuming the measuring parameters obey the Gaussion distribution, this paper uses the large-probability distribution interval to approximate the variation range of random variable, constructs appropriate working condition matrix whose column quadratic sum approximately obeys the Chi-square distribution, and then obtains the Chi-square distribution based steady-state data judgment criterion. The judgment threshold of this criterion could be self-adaptive. This judgment criterion has been validated by an experiment.
Authors: Yong Le Lü
Abstract: In the research of Prognostics and Health Manage-ment aiming at the airborne equipments such as aeroengines, the working state of equipments can be efficiently monitored based on the flight data acquired, recorded and transported to the ground database by the Aircraft Condition Monitoring System.Firstly, the conception of normal working performance model and the Polynomial Coefficient Auto-Regressive model are introduced in the paper to help identify the abnormality of equipments. Secondly, based on chi-square distribution model, the abnormality detection algorithm based on chi-square test of standardized error sum of squares and the abnormality detection algorithm based on chi-square test of distribution fitting are put forward to detect the equipments’ latent damage or fault. Compared to the former, the later can effectively reduce the rate of false alarm, however response unpunctually to the equipment’s abnormality. Finally, the validity of algorithms is confirmed by the results of simulations aiming at a low pressure compressor rotor vibration amplitude sequence. It is indicated that the algorithms will be good tools for condition-based maintenance and autonomic logistics in future.
Authors: Jie Su, Xu Guang Wang
Abstract: Assuming the measuring parameters obey the Gaussion distribution, this paper proposes a Chi-square distribution based steady data judgment algorithm, which includes two parts, i.e. gross error detection and steady data judgment criterion. In order to overcome the noise’s influence on the result of neutral network training, the paper introduces the RANSAC algorithm into the neutral network training and put forward a RANSAC-BP neutral network training algorithm, which culls noisy data during neutral network training and then retrain the neutral network with noise free data, thus robust to data noises. This algorithm has been validated by a simulation experiment.
Authors: T.M. Ivanova, H.U. Lubman, T.I. Savyolova, Vladimir Serebryany
Abstract: Experimental pole figures are measured by x-ray method for materials with hexagonal symmetry (Ti and Mg alloys). The Orientation Distribution Function is calculated by approximation method with central normal distribution. Texture inhomogeneities and effects of defocusing are the main sources of pole density errors. The measurement errors depend on crystal direction {hkl} and are different for maximum and minimum regions on pole figure. The influence of texture measurement errors on accuracy of the ODF calculation is investigated.
Authors: Jie Su, Xu Guang Wang
Abstract: This paper proposes a gross error judgment criterion and diagnoses the transformer fault by integrating the gross error judgment criterion and the characteristic gas ratio method. In this way it is possible to judge whether the transformer is in face of an incipient fault by examining the gross errors of the measured values of the fault characteristic gases, at the same time the fault probability could be calculated according to the remarkable level. And then in combination with the characteristic gas ratio method, the fault category and fault cause of the transformer could be figured out. The method has been validated by an actual example of fault diagnosis.
Showing 1 to 5 of 5 Paper Titles