A bran-new performance evaluation model of coal mill based on GA-IFCM-IDHGF method. (31st May 2022)

• Record Type: Journal Article
• Authors: Xu, Wentao; Huang, Yaji; Song, Siheng; Cao, Gehan; Yu, Mengzhu; Cheng, Haoqiang; Zhu, Zhicheng; Wang, Sheng; Xu, Ligang; Li, Qiubai
• Is Part Of: Measurement. Volume 195(2022)
• Publication Date: 2022-05-31
• Subjects: Performance evaluation -- Coal mill -- Genetic algorithm -- Intuitionistic fuzzy clustering -- Improved IDHGF method; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2022.110954
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A bran-new performance evaluation model of coal mill based on GA-IFCM-IDHGF method is proposed in this paper. GA-IFCM method is adopted to identify the operating modes of the coal mill. The sample data collected from normal historical operational data sever as training sample to establish the GA-IFCM-DHGF assessment model and determine the security range among variables to choose a suitable method applied to performance evaluation of coal mill. To tackle the complicated and sensitive relationship between the current operational data and the importance among variables of coal mills, an improved IDH model proposed in this research takes into account the importance among variables that vary with the current operational data of coal mill. By looking for the functional relationship among normal variables and setting the principle of establishing scoring matrix, an improved IGF model is adopted to automatically obtained the scoring matrix to realize performance evaluation online. Abstract: Adopting a more scientific and precise assessment method to evaluate the current running performance of coal mills in practice work is essential to maintain the security and reliability of the coal-fired power plant. However, traditional assessment methods have ignored an important problem that the importance among variables varies with the current operation data of coal mill, which has a serious influence on performance evaluation of coal mill. In this paper, a bran-new GA-IFCM-IDHGF assessment method is proposed. Genetic algorithm (GA) is first applied to optimize initial parameters, which is fundamental and significant step to obtain more accurate the number of clusters. Secondly, intuitionistic fuzzy clustering model (IFCM) is adopted to identify the running modes of the coal mill. According to the Delphi, analytic hierarchy process, gray relations analysis and fuzzy integrated evaluation (DHGF), the security range of variables under the normal observed samples from different running modes is acquired, which is helpful to determine whether the current operation data is normal data. Subsequently, an improved Delphi, analytic hierarchy process, gray relations analysis and fuzzy integrated evaluation (IDHGF) method is used to assess the current running performance of coal mill under the abnormal monitoring data. The performance of the proposed performance assessment method is verified through its application in a self-defined step fault system and an actual industrial cases of coal mill. Compared with the traditional methods, the experimental results demonstrate the effectiveness of the proposed method.