System Identification And Temperature Effect Research Of The Grating Ruling Engine

The spectrum instruments are widely applied in the aerospace, astronomical observation, biology research, national defense construction, and so on. As the core component of the spectrum instrument, the diffraction grating directly affects the development of the spectrum instrument. Due to the requirements of large area and high precision, the grating must be manufactured by the mechanical grating ruling engine. The mechanism of the grating ruling is extremely complex. There are many factors that influence the accuracy. This paper analyzes the critical components of system and the characteristics of the positioning system are discussed based on the system identification. The relationship between the temperature vibration and the system error is researched and a compensation system has been built.The mechanical grating ruling mainly includes the positioning system, the ruling system, and the measurement system. The detailed structure of the three systems and the working modes of the ruling engine are introduced. The positioning experiments of both the large travel mode and the short travel mode are conducted and compared. In order to understand the characteristics of the ruling engine, the core components are researched specially, including the positioning stage, the positioning guide rail, the tool holder, and the tool holder guide rail. The modal analysis and the deformation analysis are applied on the core components. The theoretical models of the coarse positioning system and the fine positioning system have been developed.The grating is installed on the positioning stage. So the position of the positioning system directly affects the performance of the grating. The system identification of the positioning system is a practical method to learn the features. The preprocessing of the original data contains data interpolation and normalization processing. There are two different methods applied for system identification. The one is the method based on the ARX model. The model’s order is determined firstly and the parameters of the ARX model are calculated secondly. The other method is the one based on the kinetic model. The kinetic model of the positioning system is built according to the real structure. Then the unknown parameters of the model are determined.As an instrument with nanoscale precision, the grating ruling engine requires a very demanding environment. The temperature vibration will obviously impact the accuracy of the system. The temperature-error model is developed to research the relationship between the temperature vibration and the system error. Both the polynomial model based on the genetic neural network and the generalized regression neutral network is employed for modeling. On the basis of the temperature-error model, the compensation system has been developed. Two controllers are used in the system, the constant PID controller and the single neuron based PID controller. Both the two controllers can keep the accuracy of the ruling engine with the extended temperature range. The positioning precision of the ruling engine with compensation system and temperature range of20+0.3℃is higher than the one of the original system with temperature range of20±0.01℃. As an adaptive controller, the single neuron based PID controller is more adaptable than the constant PID controller.