Font Size: a A A

Design And Development Of Power Plant Safety Assessment Expert System Based On B/S Structure

Posted on:2018-09-29Degree:MasterType:Thesis
Country:ChinaCandidate:J L SunFull Text:PDF
GTID:2382330515457623Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of computer and internet technology,enterprise training and examination methods have changed greatly.The use of network test system can shorten the test cycle,reduce the repeated the work of training officer,improve the work efficiency.And training officer must conduct safety assessment for the power plant staff and foreign workers every year to increase the safety awareness of employees.According to the general requirements of safety assessment of power plant,this paper studied the power plant safety assessment expert system based on B/S structure.First of all,As for the complicated construction and the maintenance problems of test library,the system realized the function which can make up questions automatically.The domain knowledge ontology is constructed by using protege.Then the system uses the Jena which can parse the ontology to parse the built domain knowledge ontology.According to the meta knowledge point pattern and the similarity algorithm,the system can generate question stem and disturbance term.Then,this paper studies the realization of genetic algorithm in the process of generating test paper.Which involves the creation of the default Configuration object,setting the fitness function,the construction of genes,the construction of chromosomes,defining the size of population,the initialization of population,the evolution of population and so on.Through transforming the constraints which contain the difficulty,chapters,question types and so on into the fitness function,system can ensure the quality of the paper.Finally,IK Analyzer segmentation method is used in the paper,as for the subjective questions which can not be directly marked.By using word segmentation in the employee's answer and the standard answer,then the system can obtain the similar degree of employee's answer and the standard answer through an improved cosine similarity algorithm,so this part's score can be got.The objective part can be sentenced automatically according to the contrast between the employee's answer and the standard answer.So,the system has basically realized automatic sentence.
Keywords/Search Tags:ontology, jena, genetic algorithm, automatic sentence
PDF Full Text Request
Related items