Data possibility-reliability distribution-based type-II prediction set security assessment method

Abstract

The invention belongs to the field of assessment of actual projects of large buildings, public facilities, precision equipment and the like, and particularly relates to a data possibility-reliabilitydistribution-based type-II prediction set security assessment method. The method comprises the steps of firstly defining type-II prediction sets and related concepts based on possibility and reliability distribution, and proposing a similarity measurement method for the type-II prediction sets; secondly, according to an assessment demand, determining a heterogeneous data security level scale, establishing joint distribution, and performing fusion on type-II distribution for to-be-assessed heterogeneous data by utilizing a weighted fusion method; and finally, calculating similarity measure between the fusion distribution and the security level scale, determining the position of a centroid in the projection surface of a joint scale, and judging a security level. An experimental result showsthat by applying the method to the judgment of the security level of a tailing dam, the granularity of an assessment result can be effectively reduced, the misjudgment problem is reduced, and the convenience is provided for accurate interpretation of the security level of a dam body.

Description

technical field [0001] The invention belongs to the field of evaluation of actual projects such as large-scale buildings, public facilities, and precision equipment, and is specifically a type-two prediction set safety evaluation method based on data possibility-reliability distribution. Background technique [0002] Expert prediction data and sensor monitoring data reflect the actual project (such as large buildings, public facilities, precision equipment, etc.) important safeguards that occur. Most of the existing evaluation methods are based on various uncertainties of data (such as randomness, fuzziness, possibility, etc.) respectively using probability theory, fuzzy set theory, possibility theory, etc. to determine the safety level. It has strong pertinence and wide application value. However, the main purpose of safety assessment is not only to judge the research object in time after an accident or failure occurs, but also to predict its future development trend when...

AI Technical Summary

Problems solved by technology

[0003] The credibility of the data is another important factor affecting the actual engineering evaluation results besides the uncertainty. It reflects the differences in the capabilities of data sources (such as expert experience, understanding, and sensor performance, operating conditions, etc.). The role in the assessment will inevitably lead to misjudgment of the safety level

Images