SELECTION AND JUSTIFICATION OF THE MATHEMATICAL 6 FACTORS EXPERIMENT PLAN IN THE STUDY OF ICE STARTING QUALITIES

Abstract

The practical significance of the application of multivariate regression analysis in engineering practice as a necessary step towards improving and optimizing complex systems and processes is presented. The goal and investigation tasks have been actualized to optimize the start-up process of a high-speed subcompact diesel engine. Preparation of the plural factors experiment features of diesel start-up process based on regression analysis and the theory of mathematical design of experiment are presented. Qualitative optimizing parameters of the diesel engine cold start are determined: instantaneous acceleration of the crankshaft dn / dτ at the time of crankshaft cranking by an external energy source close to the time spent on starting the engine with optimal parameters of the starting system and starting energy consumed during the crankshaft cranking by an external energy source. Six main factors influence on certain parameters of the diesel engine start quality have been investigated. The regression equation for assessing the quality of the start-up process is substantiated. The form of regression equation is a full quadratic polynomial for the reproduction of the investigated functions, based on the previous study of some used in the study individual factors influence. The analysis of proposals for rational plan selection of a six-factor experiment to find the coefficients of the second-degree regression polynomial is carried out. The substantiation of the decision regarding the choice of the plan is given. Selection was based on the requirements of the investigation test bench and the conditions for study organizing. Main impact had the predecessors experience of similar investigation and the statistical criteria for different plans of experiment comparison used in mathematical theory of design of experiments. The points of the plan have been defined on a multidimensional cube due to the need to vary on three levels with a uniform step of all 6 factors that were identified as influential. Under the conditions of the available laboratory test bench, the current values of the factors of equivalent cold start temperature, crankshaft-cranking speed and maximum temperature of the glow plugs have a variance of the installation at different points of the plan, and the nature of the factors on energy consumption is unknown in advance. As a rational experimental design for organizing the current study, the Box-Wilson orthogonal central compositional plan was chosen, built by adding plan points on the axes of the factorial space to the full-factorial plan of the lowest order, while maintaining the requirement of orthogonality and symmetry of the plan.