Research On Quality Control Theories And Methods For Complex Workpiece Machining Process

A complex workpiece is defined as a kind of mechanical part which has complex composition, multiple quality characteristics, multiple machining processes, and high manufacturing precise. The quality control of complex workpiece machining process (CWMP) is the basis of product quality improvement and the competitive ability of enterprises. The CWMP quality control is concerned in this dissertation, the research method of this dissertation is combining the theory analysis with practice application. The core technologies, such as machining process variation propagation modeling, variation sources diagnosability analysis and identification, machining process technology capability index calculation and improvement, setup planning selection in machining process, etc., have been mainly studied. The basis of the CWMP quality control theory is established. A CWMP quality control method for assembly is proposed. The main contributions of this dissertation are as follows.(1) On the basis of the CWMP characteristic and variation sources analysis, a CWMP variation propagation modeling theory based on state space model is proposed. A machining variation propagation model is established by describing the workpiece with the differential motion vector according to the kinematic principle. This model explicitly describes the relationship between quality characteristic variations of CPMP and their variation sources, which can quantify the variation propagation effects analysis among different stages and is testified by an example.(2) To study the quality variation information propagation and identify the variation sources in the CWMP, a variation sources on-line identification method is proposed based on the complicate mutuality analysis of quality variation information in the CWMP. An identification model for the CWMP variation source is established by translating the variation propagation model into a linear mixed model. A variation source fully diagnosable model and a partially diagnosable model are presented for the CWMP by referring to the identifiability in a linear mixed model. To study the partially diagnosable CWMP, the concept of a minimal diagnosable class is proposed to identify the process variation and realize the identifiability study of partially diagnosable process. On this basis, introducing the multivariate variance components analysis method into the CWMP diagnosis, combining an engineering case, an on-line variation sources identification method for CWMP is proposed based on statistical estimates and hypothesis testing, which realizes the quick variations sources identification in the CWMP.(3) To study the problem of the process technology capability in the CWMP, an observation model is established for technology capability in the CWMP according to the process adjustment strategy, a technology capability index measuring method is presented based on Taguchi's quality loss function. Based on this, in order to achieve the highest efficiency method of process technology capability adjustment, a two-level sensitive analysis model for process technology capability impact parameter model is proposed to determine the positions and directions of the process adjustment. A technology capability adjustment method for the CWMP is proposed by combining an engineering case.(4) To study the optimization of setup planning in the CWMP, a setup variation propagation model is established based on the variation propagation model of the CWMP. A minimum cost and quality assured function is defined for an objective function to obtain an optimal setup planning selective decision and the solution of this problem is based on dynamic planning algorithm, which realizes the optimum configuration of setup planning parameters in the CWMP.(5) To study the CWMP quality control method for assembly, the core technologies are extracted according to the analysis of CWMP quality control problem for assembly, a matchable degree calculation formula is proposed. The CWMP quality control flow for the assembly is presented based on the study of the CWMP quality control flow for the single. A CWMP quality control method for assembly is presented by combining an engineering case.