Application of matrix approach of fuzzy logic for decision support in oil mining equipment service

The work solves the problem of building an intelligent decision support system for servicing oil production equipment. At the first stage – the choice of an intelligent model - it is shown that in the existing conditions it is difficult to obtain a training sample in digital form. On the other hand, there is an opportunity to gain knowledge of subject matter experts – masters and technologists – in the form of a set of linguistic rules. Based on this, a conclusion about the effectiveness of the use of fuzzy logic to solve this problem is made.
At the stage of constructing an intelligent model, the use of the matrix approach of fuzzy logic is proposed. To elaborate this approach an algorithm of fuzzy inference based on vector fuzzy predicates is developed. Capabilities and advantages of new algorithm are demonstrated. In particular, it is shown that the matrix representation makes possible reducing computations to solving a system of linear equations. Matrix inference also allows to explicitly determine the range of values of the analyzed parameters at which the knowledge base does not allow making a clear conclusion.
A model of a fuzzy logic machine in the form of a fuzzy combinational circuit that analyzes an external memory block is proposed for the analysis of retrospective information on the change in the values of the parameters of technological equipment over time. Specific cases allowing the transition from a state machine to a combinational circuit are shown. Article also shows how this can be done. The main advantage of this approach is the absence of the need to use the difficult to formalize concept of a fuzzy state, which leads to a simplified construction of fuzzy logical devices with memory.

At the end work contains brief conclusions about the application of the proposed methods and algorithms for building, testing, implementing a decision support system and about its effectiveness.