INTELLIGENT APPROACH FOR DIAGNOSIS ELECTRICAL EQUIPMENT OF INDUSTRIAL FACILITIES

Abstract

This paper is devoted to actual problem of ensuring of trouble-free functioning and a high level of fault tolerance of industrial facilities’ electrical equipment (EE) by diagnosing them during their operation. However, when diagnosing equipment, a number of challenges arise related to the following problems: a large number of monitored parameters, the presence of not only statistical information but the values of linguistic variables that are used by staff on duty, insufficient and incomplete input information. Proposed approach to diagnosing EE based on a composition of fuzzy cognitive model with a fuzzy vertices, links and fuzzy-functional cognitive model with functional links for generating intermediate information, and as well as fuzzy-production model of an upper level for taking final diagnostics decisions. The proposed approach is based on the functional model of EE diagnosing in IDEF0 notation which reflects the sequence of all necessary functions related to the diagnosing and provides the identification of the key data flows and a more comprehensive understanding of the process. The paper also provides a formulation of the fuzzy-set method that is based on the construction of functional dependencies formed in clear mapped set “the occurrence frequency of for equipment's dangerous situation”. This method’s novelty lies in the new formulation of the problem in accordance with the latest scientific achievements and its solution that was not previously considered in the well known sources of literature. The suggested method is allows to calculating the predicted values of controlled parameters of EE in case of statistic and fuzzy source data. The proposed approach to the EE diagnosis at industrial objects makes it possible to improve the accurate and complete EE diagnosis through the knowledge of the staff on duty, thereby resulting in timely repair and maintenance.