Selection of a Ship Compressor Using Statistical Data Processing

Abstract

INTRODUCTION: Compressors are essential for supplying compressed air at various pressures and flow rates to the ships power systems, and to the ship as a whole. An appropriate choice of the compressor is essential for the proper operation of water transport infrastructure. Hence, it is necessary to consider this issue in detail. If all variables, the compressor selection parameters to be considered, were measured in the same scales and units, it would be possible to suggest adding up all values, but this approach is very crude. The solution is to normalize the values of the variables and then calculate a final criterion based on them.
 AIM: Finding a formal criterion by which the selection of a particular compressor will be made.
 MATERIAL AND METHODS: The process of selecting a compressor for starting the ships engines and for the general operation of the ship is discussed based on exponential and linear rationing methods. Certain parameters of compressors from domestic and foreign manufacturers are offered for statistical processing. Subsequently, data processing was carried out using maximum likelihood estimation and logistic regression.
 RESULTS: The result of the study is a single formal criterion, the final rating, instead of several qualitative parameters. A set of characteristics has been determined that describes the optimal compressor based on the above calculations and data processing. A simulation of the probability of assigning a compressor, with a certain set of characteristics, to a positive or negative class has been performed. The training of the model was done using the available training data.
 CONCLUSION: Statistical methods of data processing can be applied to an object such as a compressor. Given a set of desired compressor characteristics, it is possible to teach the machine to determine the optimal compressor.