Multicriteria Approach to the Selection of Optimal Diagnostic Characteristics of Ship Bearings Monitoring

The intensity of the use of marine vehicles, violations of operating rules, cyclic loads with alternating signs and various manifestations of the external environment, which cannot be taken into account, lead to a decrease in the accuracy of determining the current state of the material. The complex structure of vibration signals and the variety of their stochastic characteristics lead to the necessity of their ranking and the priority of their use. Optimisation parameters based on the change of the trajectory of the main diagnostic feature, when it approaches the degradation state, are proposed and investigated in detail. As parameters of optimisation of stochastic characteristics of vibration signals, the features of their trajectory change under loading are used: monotonicity, speed of change, sensitivity to determination of residual resource, deviations from additivity, and energy spent on defect development. The matrix of efficiency, in which rows are represented by statistical characteristics of vibration signals, columns - by criteria of statistical solutions of Laplace, Wald, Hurwitz, additive, multiplicative and additive multiplicative convolution, and its elements - by practical results of vibrodiagnostics of sliding bearings, is compiled. It is established that the optimum characteristic of monitoring the technical condition of the sliding bearing in the process of its operation is the peak factor. Implementation of this diagnostic parameter allows for the carrying out of monitoring without stopping the equipment.