Abstract
The main parameters (technological conditions and geometric factors) the optimization of which is used to simulate electroplating processes with specified characteristics (coating thickness nonuniformity over the part surface and process duration) are considered. The mathematical statement of the electroplating process optimization problem written in the form of additive convolution of partial criteria is presented. The composition of the process mathematical model’s system of equations and the algorithms for finding its solution are described. It is pointed out that the parameters adopted for calculations according to the mathematical model may differ considerably from the actual parameters. To eliminate the need of repeatedly solving the optimization problem, it is proposed to use the experience, knowledge and intuition of a specialist in electroplating technology who directly conducts the technological process. The proposed approach can be implemented using a fuzzy production model of knowledge that takes into account technological conditions and geometric factors in an integrated manner. The aim of the developed model is to obtain close-to-optimal experimental values of partial criteria taking into account the deviations of the found parameter values from their actual values by adjusting them. The input and output variables with terms and membership functions are defined. The key electrochemical regularities required for constructing the production model knowledge base, which contains a system of rules based on conditional statements in the form of "IF ... THEN ..." are formulated. The application of the developed model is considered taking as an example the problem of correcting the results of modeling and optimizing the nickel plating process in a sulfate electrolyte for parts with complex surface shapes. For the selected forms of parts, the optimization problem was solved, the results of which were implemented in the plant without and with correction according to the developed knowledge model. The obtained results were compared by calculating the relative deviation of the predicted value of the criterion from its experimental value. It is shown that the correction of the found optimal parameters has an effect on the target criterion with increasing the weight of the first term (the coating thickness nonuniformity) due to its being calculated for an object with distributed coordinates and, as a consequence, due to a nonlinear dependence.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call