Comprehensive reliability evaluation of multistate complex electromechanical systems based on similarity of cloud models

Abstract

High reliability and security have become the hallmarks of complex electromechanical systems. Owing to the difficulties in fault data collection, ambiguity and uncertainty have been inevitably associated with complex electromechanical systems. Thus, the ability to perform reliability evaluation using scarce fault data is of immense significance to these machines and is the focus of this study. A similarity based cloud model is proposed to evaluate the running state of complex electromechanical systems. By combining objective and subjective factors, the reliability of complex electromechanical systems is evaluated by calculating the similarity between the cloud models of actual and standardised states. Next, the inverter of an offshore wind turbine is used to verify the accuracy and effectiveness of the proposed approach. The cloud model based framework for reliability evaluation inherits the preponderance of the uncertainty problem, overcomes the drawbacks of the current reliability approaches, and provides a theoretical basis, as well as a practical approach for the maintenance and repair of complex electromechanical systems with missing fault data. Additionally, it also provides a new methodology for solving the uncertainty problems caused by paucity of data.

A novel self-learning framework for fault identification of wind turbine drive bearings

“…Based on the above feature extraction process, a three-dimensional feature matrix is formed, and each column in the matrix corresponds to a specific fault data feature, such as equation (7) given below…”

“…1–4 Therefore, monitoring and diagnosing the condition of the drive bearings of wind turbines is an essential way to ensure their safety and reliability, reduce their life-cycle maintenance costs, and extend their performance life. 5–12…”

“…[1][2][3][4] Therefore, monitoring and diagnosing the condition of the drive bearings of wind turbines is an essential way to ensure their safety and reliability, reduce their lifecycle maintenance costs, and extend their performance life. [5][6][7][8][9][10][11][12] Fault pattern identification is derived from fault diagnosis, which aims to identify and label the raw fault data without a priori knowledge and to establish a targeted fault pattern database of the wind turbine performance process. 10,[13][14][15][16][17] Traditional fault pattern identification is mainly divided into statistical-based methods and artificial intelligence-based methods, 18,19 such as fuzzy classifiers, [20][21][22] random forests (RFs), 23,24 and clustering methods including k-nearest neighbor (KNN) and spectral clustering.…”

Cognitive excursion analysis of uncertainty concepts based on cloud model

“…The similarity computing of cloud concept has been widely used in classification, clustering, similarity search, collaborative filtering, target recognition, watermarking technology research, system evaluation, similarity analysis of DNA sequence, data mining of stock time series and other fields [36][37][38][39][40][41][42]. At present, there are three main methods to calculate the similarity of cloud concepts.…”

Reliability analysis of complex electromechanical systems: State of the art, challenges, and prospects

“…Obviously, RA has been steadily attracting attention from scholars. A large number of valuable studies on topics ranging from simulation 1,2 to test design, 3,4 from coupled modeling 5 to decoupling, 6 from theories 7,8 to tools, 9 from improving analytical methods 10,11 to predictive methods, 12,13 and measuring behavior in the field 14 are available not only to solve many reliability problems in actual applications but also to promote research on reliability. However, these studies differ in their conditions of applicability, scopes, and objects.…”