Some procedures or functions had be added to an ESTA (Expert System Shell for Text Animation) so that the ESTA and MATLAB can communicate via some data files.On this basis,a deep learning-DBN(Deep Belief Network) and two BP(back propagation) artificial neural network based on the MATLAB programming were researched by using directly DGA (Dissolved Gas Analysis) and characteristic gas method in transformer oil chromatographic analysis.The transformer fault diagnosis expert system based on a three ratio and characteristic gas method of DGA and ESTA including the DBN and two BP artificial neural network programmed in MATLAB had be created.The basic application shows the effectiveness of the expert system.
The outside rear-view mirror (OSRVM) is installed on the vehicle’s surface, which causes unwanted aerodynamic noise and wind drag during driving. It is important to use simulation methods to predict the performance of aerodynamic noise and wind drag of commercial vehicles due to the OSRVM. Considering the wind drag of the OSRVM, a combinational simulation strategy is employed to calculate external flow and interior acoustic fields of commercial vehicles, respectively. The flow field is computed a priori with an incompressible flow solver. The acoustic field was then computed based on the information extracted from the CFD solver. To obtain the interior noise level at the driver’s ears, a vibroacoustic model is used to calculate the response of the window glass structure and interior cavities, where the unsteady aerodynamic pressure loading on the two side windows’ surface is treated as the acoustic source field. The paper provides flow field and acoustic simulations for three OSRVM configuration models. The results are compared to data obtained in road sliding test measurement on the commercial vehicle. The accuracy of the hybrid simulation method is proved, and the comparative analyses verify that the OSRVM B model dramatically reduces the interior noise and wind drag of commercial vehicles.
An application research work had improved a Bayesian network rule function of Pro/3 (a production system type expert system shell). This improvement makes it possible for the same rules to ask the same evidence such as the evaluated probability or the certainty factor only once, which can avoid repetitive ask, adding simplicity and improving performance of Pro/3 system. By using it, this work had also created a transformer component fault diagnosis knowledge base model based on the Bayesian network rule so that using the model had implemented an expert system prototype system. The basic application shows the effectiveness of the improvement and the prototype system.
It makes possible that it can do basic uncertain inference and directly read/write some table contents of MySQL database with improved ESTA 4.5 (Expert System Shell for Text Animation) based on Visual Prolog 6.3 and Windows, namely adding some uncertain inference functions and some interaction functions with MySQL to the ESTA. It put forward also an uncertain inference method in the ESTA 4.5 based on Visual Prolog 5.2 and Web. It had designed and implemented a condition evaluation expert system of substation DC (Direct Current) system based on the improved ESTA and the enterprise standards (Q/GDW 607-2011) of State Grid. The test shows that the uncertain inference of improved ESTA has feasibility, the interaction with MySQL simplifies some rules of writing and enhances flexibility of it, and the expert system is effective and practical.
An application research work had improved some functions of a file scan and transformation software (FileScanner) in Pro/3(an expert system shell) by exploring transformation of semi-structured files (Html format) into structured text files. Some existing problems such as line feed failure and Chinese characters incorrectly displaying in the result file transformed had been solved by improving its Java programming. After a .Html format file be scanned and transformed,a .txt format file produced can implement effectively line feed when it is directly opened, and can display correctly Chinese characters. The structured text file transformed can directly interact with other application programs or databases so as to facilitate the analysis of semi-structured data and mining some values of the information behind the data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.