Papers by Keyword: Principal Component Analysis (PCA)

Abstract: During the course of building an innovative country and enhancing the independent innovation capability, universities are the main force and the important source of high-tech innovation. The evaluation on the university's innovation ability, not only may improve university's efficiency and level of scientific research, but also make a significant sense to perfect the china' scientific research innovation system. Based on Referring to the recent research achievements at home and abroad, research and design work was carried out in the following area. Firstly, the multi-university research innovation ability evaluating indicator system is designed in this paper. By the principle of science and justice, through questionnaires, expert opinion and reference to relevant research results. The paper designed the multi-university's research innovation ability evaluating indicator system. A variety of typical evaluation models and methods are studied. Then two evaluation models between PCA-BP and PCA-FNN are taken into comparison. And the results show that the research and application of PCA-FNN is proved to be a new method and made a significant attempt for the university’s evaluation of research innovation ability.

Abstract: In the process of multi index analysis and evaluation, the index value is normalized and standardized through non-dimensional processing of the original data. The index weight is determined according to each index relative principal component contribution through principal component analysis and principal component extraction. Then, a combined evaluation model will be formed through the grey relational analysis, calculation of each index according to grey weighted relative degree. Scientific applicability of the model will be evaluated through regional economic opening degrees.

Abstract: In order to analyze the water use efficiency of Jiangsu by the method of quantitive research, this article takes the water use amount per ten thousand Yuan GDP as object of study, and discusses the influencing factors of water use efficiency to give the suggestion of policy , regular and control. Theory analysis and expects consulting have been used to get 21 potential influencing factors. Then we get the regression equation which contains water use amount per ten thousand Yuan GDP as the dependent variable, and 21 mentioned factors as independent variables through introducing collected factors’ dates in the period of 1997~2010 into the stepwise regression which is on the basis of principal component analysis. As a result, the major four factors have been got: irrigation area per capita, Industry polluted water standard drainage, education cost, proportion the output value of tertiary industry accounting for of GDP. And this research works for the water use efficiency red line investigation in the future.

Abstract: The fiber characteristics directly affecting the yarn tensile properties are analyzed and the features of GA-BP artificial neural network are presented. Based on the interrelationship among the fiber characteristics, which could led to a worse predicting result, Principal component analysis (PCA) is adopted to solve this problem. 12 characteristics of cotton fiber tested by HVI or AFIS and yarn process parameters, such as combing, degree of twist, yarn count, was preprocessed by this method and 10 independent comprehensive indexes are induced and substituted into GA-BP artificial neural network as input factors. For further comparison, the 10 dimension reduced indexes and the 15 initial variables are respectively introduced into the GA-BP and BP artificial neural network to develop 4 prediction models. By comparing the accuracy in predicting yarn tenacity, it is concluded that GA-BP model has higher accuracy than BP model, and the dimension reduced indexes based on PCA would decrease the accuracy in prediction instead. So, blindly using PCA method for reducing the complex correlation among input variables was not helpful for the prediction accuracy.

Abstract: Support Vector Machine (SVM) method is suitable for machine learning. In order to detect pathological information from tongue diagnosis rapidly, noninvasively and objectively, a near infrared spectral identification model is proposed based on SVM. The tongue spectral data of healthy people and hepatitis patients were collected. Twenty two samples were obtained for individual groups, and for each group, fifteen samples were randomly selected and used as the training sets, while the other seven were taken as the prediction sets. For the data sets, The effects of the principal component number, kernel parameters, and kernel functions on the identification model were investigated respectively. The results showed that the penalty parameter c was always 0.25, not related to the values of the principal component number and kernel parameter g. The kernel parameter g decreased along with the increased number of principal components, and ultimately reached a relatively stable value. When the Radial Basis Function (RBF) was applied, the established model was the best, indicating that the SVM approach is feasible to classify and recognize tongue near infrared spectroscopy, as along as right parameters are selected. This can provide a novel tongue spectral analysis method to distinguish healthy individuals from hepatitis patients.

Abstract: A material on-line identification system based on THz time-domain spectroscopy using VC was developed in this paper. The main functions including data reading, linear motor driving, the calculation of optical parameters, the display and management of curve, the management of interface, the analysis of historical data , the maintenance and updating of database, and on-line identification based on principal component analysis were realized. As the prototype of material detection application, the material whose corresponding type characteristic spectrum has already existed in the database could be well identified.

Abstract: In order to effectively control pollution of sources of drinking water, improve the environmental quality of drinking water and guarantee the sanitation of drinking water, it is very important to assess water source quality. Main factors of drinking water were identified. Then principal component analysis was used to establish assessment model of drinking water, which could ensure that under the condition that the primitive data information was in the smallest loss, a small number of variables were used to replace the integrated multi-dimensional variables to simplify the data structure. The weightings of principal component were determinated as theirs pollution ratios. This paper was based on the theoretical study of principal component analysis, used the monitoring data on water quality of the main water resources in 2013 to evaluate and analyze the water quality of water resources. Analysis content included the main affecting factors, cause of pollution and the degree of pollution.The resulted showed that: the main affecting factors on water quality of Fo Si water source was CODMn, TP, fluoride.

Abstract: Based on the theoretical basis of life cycle theory and the artificial fish swarm algorithm, this article built the evaluation index system of carbon reduction ability of real estate development project, and put forward a comprehensive evaluation method. The method objectively determined the weight coefficient of evaluation index, and considered carbon reduction ability of real estate development project from the whole life cycle. Through example analysis, the evaluation method can effectively evaluate carbon emission reduction ability of real estate development project, and provides a reference for developers or the examination and approval department.

Abstract: Abstract. Thermal efficiency is an economical operation index of industrial boilers. There are many factors influencing thermal efficiency. It is difficult to keep the boiler in high efficient operation just using single automatic control method when environment has been changed. Therefore, the control of combustion systems is usually depended on artificial experience. To improve this situation, an operation optimization method is proposed. An identification model which can reflect the thermal efficiency is established by using principal component analysis based on historical data. When the boiler’s operation efficiency decreases, the parameters of influencing boiler efficiency can be directly got by contribution plot method, which can guide operators in real-time to adjust these parameters maintaining boiler efficient operation. Abstract. Thermal efficiency is an economical operation index of industrial boilers. There are many factors influencing thermal efficiency. It is difficult to keep the boiler in high efficient operation just using single automatic control method when environment has been changed. Therefore, the control of combustion systems is usually depended on artificial experience. To improve this situation, an operation optimization method is proposed. An identification model which can reflect the thermal efficiency is established by using principal component analysis based on historical data. When the boiler’s operation efficiency decreases, the parameters of influencing boiler efficiency can be directly got by contribution plot method, which can guide operators in real-time to adjust these parameters maintaining boiler efficient operation.

Abstract: Principal component analysis is performed on the operation and management evaluation of smart distribution network because of its objectivity, synthesis and simplification to original data information. According to the demand on the evaluation that focuses on intellectualization and reliability of distribution network, an index system for intelligent and reliable evaluation is built. The performance indicators and the principal component analysis are used to analyze the intelligent and reliable level of distribution network operation and management. The feasibility of the evaluation index system is verified by the evaluation results of some distribution networks.