Under conditions of automated production the matter of ensuring the required parameters of surface layer quality in machine parts at machining becomes urgent. To ensure the required quality parameters of a surface layer (QPSL) in machinery at mechanical oper-ation there are used simulators predicting their values depending on working modes. At the ensuring of QPSL in machine parts at mechanical operation occurs uncertainty connected with the absence of univocal simulators predicting QPSL depending on working modes. One of the methods to solve the uncertainties existing consists in the instruction and self-instruction of a technological system during control and application of obtained and accumulated information at adaptive control. And at the same time the problem of parametric identification at the adopted structure of a simulator is solved. The parametric identification of simulators is carried out by a self-learning technological system of control (STSC). The developed STSC is intended for ensuring the specified parameter of roughness Ra , sur-face residual stresses, surface micro-hardness and the complex quality parameter of friction surface Cx. The algorithm of system functioning is realized as a soft-ware loaded in memory of a control device.