Situational assessment method based on fuzzy dynamic Bayesian network-adaptive particle filtering

Abstract

The invention discloses a situational assessment method based on fuzzy dynamic Bayesian network-adaptive particle filtering, and the method comprises the steps of 1, data anti-interference and fuzzification processing, 2, construction of a Bayesian network, 3, establishment of a fuzzy dynamic Bayesian network, 4, segmentation of the fuzzy dynamic Bayesian network, and 5, combination with self-adaptive particle filtering. Compared with a traditional situational assessment method, the method has the advantages that the linear regression model is adopted to fill the data, data fuzzification is achieved by combining attribute deletion and concept tree lifting technologies so as to overcome the incompleteness of the collected data, and the accuracy of the algorithm is improved. Compared with astatic reasoning algorithm, the method has the advantage that the real-time performance of reasoning is improved by intervening time characteristics, and prediction processing is carried out on each group after segmentation of the fuzzy dynamic Bayesian network in combination with self-adaptive particle filtering, so that the rate and the real-time performance of the algorithm are improved.

Description

Technical field [0001] The present invention relates to the field of computer software engineering and situation assessment, specifically, it is mainly a situation assessment method based on fuzzy dynamic Bayesian network-adaptive particle filter. Background technique [0002] The current situation assessment has the following shortcomings: First, the data source is single, incomplete data sources and small changes in data will have a significant impact on the analysis results; second, the situation analysis results obtained from different data sources are single; third, the wrong judgment of the dynamic situation, Situation assessment is only performed on the information at a single moment, without considering the correlation and mutual complementation of the information at the previous and subsequent moments. Fourth, the real-time performance is not high. When the parameter information dimension is too high, the inference speed is slow. Therefore, there is an urgent need for a ...

AI Technical Summary

Problems solved by technology

[0002] The current situation assessment has the following shortcomings: first, the data source is single, incomplete data sources and small changes in data will have a significant impact on the analysis results; second, the situation analysis results obtained from different data sources are single; third, the dynamic situation is wrongly judged, Situation assessment is only performed on information at a single moment, without considering the correlation of information at previous and subsequent moments and the mutual complementarity of information; Fourth, the real-time performance is not high. When the parameter information dimension is too high, the reasoning speed is slow

Images