Process Certification For Manufacturing Process Of Rolling Bearing Based On Poor Information

Rolling bearing is an important part in mechanical system equipment, and the reliability is needed to be realized and guaranteed through the manufacturing process at the design stage. Therefore, it is a problem must be solved that ensuring the process stability of bearing parts in the manufacturing. Traditionally, the traditional statistical methods were adopted mostly to analyze the stability of manufacturing process of rolling bearing. However, the general manufacturing process has the more complex property, and it is not reliable to analyze the stability of it with the help of traditional statistical methods. So, in this paper, the poor information theory is used to study the process certification for manufacturing process of rolling bearing, researching the output errors and ensuring the stability.According to grey bootstrap maximum-entropy method, the distribution of output errors and the adjustment for the machining errors of machine tool was studied in mechanical manufacturing process. In order to analyze the error distribution, the bootstrap methodology was applied to get a large number of sample data by bootstrap resampling from the current information vector of small size, and then the grey system theory was used to establish a grey bootstrap model(GBM) of dynamic prediction. The probability distribution of output errors could be obtained by using the maximum entropy method. Based on the probability distribution of output errors, the machining errors of machine tool were adjusted so that the machining errors of products could satisfy the requirements. The machining errors of machine tool can be adjusted accurately by using grey bootstrap maximum entropy method, and the forecasting accuracy is high.Based on the grey system theory, the stability evaluation of manufacturing could be put into effect via grey relation analysis of the two data sequences in the manufacturing process. According to these two data sequences obtained in the manufacturing process with certain property, the data sequences could be sorted, so the sorting data figure was achieved. The grey relation between two data sequences was established by means of the distribution features of sorting data figure. And the stability evaluation for manufacturing process can be realized through calculation and analysis the grey confidence level. Computer simulation experiment and actual case indicate that through analyzing the grey relationship of two data sequences, if the grey confidence level is not less than 90%, the manufacturing system is stable; otherwise, the manufacturing system is not stable. The method proposed is very good at testing the stability of the manufacturing system, with accuracy up to 100%.Based on the poor information theory, this paper proposed the mutation probability obtained by the bootstrap maximum-entropy method to evaluate the manufacturing process. The probability density functions of the two data sequences with specific attribute could be established in the manufacturing process by means of bootstrap maximum-entropy method. Through calculating the intersection area of the probability density functions, the mutation probability could be applied to evaluate the manufacturing system. Simulation experiment and case study show that it can examine the manufacturing process whether there is mutation or not through analysis of the mutation probability, and the forecasting accuracy is high.When appearing the system errors in the manufacturing process of rolling bearing, the non-sorting grey relationship was used to dynamically evaluate the stability in the process of manufacturing by using the grey system theory. Via establishing grey relationship of two non-sorting data sequences and calculating the grey confidence level, the real-time monitoring of manufacturing process could be carried out. Engineering cases show that using the non-sorting grey relationship for dynamic evaluation of the stability of manufacturing process is feasible.This paper studies manufacturing process of rolling bearing via the grey system theory, implementing the static and dynamic evaluation of stability of manufacturing process, and verifying the feasibility of this method. This topic research makes up for analyzing the manufacturing process of rolling bearing with poor information theory, for providing a new method to research the mechanical manufacturing system, having significant academic and application value.