Fei Gao
Belief is all I need
Social
Contact:
School of Aeronautics
Shandong Jiaotong University
Jinan, Shandong 250357 China
Email: gaofei1995@hotmail.com
Hosted on GitHub Pages — Theme by orderedlist
Research Highlights
Weapon system-of-systems analysis
Equipment evaluation based on IF-WINGS under system-of-systems warfare environment
Study on equipment evaluation under complex system-of-systems (SoS) warfare environment, where various factors including tasks, capabilities and equipment interacts. By extending the weighted influence non-linear gauge system (WINGS) with intuitionistic fuzzy sets (IFSs), we introduced the IF-WINGS method to analyze the interrelationships among different elements of the complex system while considering the strength and influence intensity of the elements. Based on the IF-WINGS method, a novel equipment system evaluation method that considers the complex interrelationships among tasks, capabilities and equipment in the equipment SoS is proposed.
Rule-based system
Density-based approach for fuzzy rule interpolation
Focusing on the problem that fuzzy rule base could suffer unmatched inputs due to incomplete information, we proposed a density-based fuzzy rule interpolation approach. Instead of selecting a fixed number of rules for the unmatched inputs without considering the actual characteristics of the selected rules, the proposed method adaptively selects the closest rules that are within a certain range of the unmatched inputs, thus assuring the selected rules are with high similarity to the inputs.
Similarity measures for extended belief rule-based system
The extended belief rule-based (EBRB) system employs the Euclidean distance measure to determine the matching degrees of various inputs against the rules, which could lead to inconsistency in som cases. We studied the characteristics of 17 different similarity measures, and applied these similarity measures to the extended belief rule-based system. In addition, we introduced an ensemble method for EBRB system with different similarity measures, where an adaptive weight learning method is used to determine the weight of each EBRB system with different similarity measures and the evidential reasoning (ER)-based combination method is used to combine the inferential results of different methods.
Human reliability analysis
Dependence assessment for HRA using cloud model and BWM
In order to properly assess the dependence level of human errors, we proposed a novel dependence assessment method that integrated the cloud model, the social trust network and the best-worst method (BWM). In this study, the social trust network is used to determine the weights of different experts, the cloud model is used to represent the uncertain and subjective judgments, and the BWM is used to determine the weights of various influence factors. By combining the dependence levels of the influence factors, the overall dependence level among human errors is obtained.