Estimation of the Most Suitable Temperature for Cooling Oil on a Spindle Using Inverse Analysis of Neural Network.

As higher precision machining is required, reducing thermal deformation of a machine tool becomes more important. Today, there are much study about the thermal deformation of machine tool ; a compulsory cooling of main spindle or some ball screws, a feed back control of thermal expansion, an isolation of heat source and etc. In case of the compulsory cooling of main spindle, however, it is difficult to decide the most suitable temperature of cooling in order to reduce thermal deformation of machine tool, because there are many non-linear factors in the thermal deformation of machine tool ; these are an irregular ambient temperature fluctuation, a property of temperature dependence on a material, change of pre-stress on a spindle bearing and etc. Therefore, in this report, a new method for estimating the most suitable temperature of cooling oil on a spindle was established. Inverse analysis using neural network was used for calculating the most suitable temperature, because neural network is suitable for controlling non-linear phenomenon. At first, in the experiment, a bench lathe was used, some structural temperatures and cooling oil temperature of a spindle were measured by T-type thermocouples for study data of the neural network, and thermal behavior of the bench lathe was also measured by a dial gauge for study data of the neural network. Next, the neural network was used to learn the relationship between the measured temperatures and thermal deformation on the bench lathe. Then, the most suitable temperature of cooling oil on a spindle was calculated by inverse analysis using pervious neural network. At last, in the experiment of the bench lathe, this method was evaluated about the various experiment conditions. It is concluded from the results that the proposed method was able to estimate the most suitable temperature of cooling oil on a spindle and effective in order to reduce thermal deformation.