Research On Dynamic Error Tracing And Prediction Modeling Of High Precision Grating Measuring System

Error is one of the key indicators to evaluate the quality of the measurement results.Inaccurate measurements can lead to the failure of entire systems and cause large economic losses.Grating measurement system is one of the main means of modern precision displacement measurement technology.With the rapid development of moden measurement technology,the precision of the grating measurement system has been to nano level.How to guarantee the high precision demand is the research focus of modern precision measurement technology.As an effective means to reduce the error and improve the accuracy,the error correction has been widely used,and such corrections are critical for precision measurements because they not only enhance measurement stability but also improve the accuracy of the instrument.Dynamic error prediction is a forward research method of error correction theory,and the error tracing is a reverse research method,they are effective measure to ensure the accuracy of measurement.The research contents of this paper are as follows:(1)Research on modeling and predicting method of dynamic measurement error;Aiming to solve the problem of low model accuracy in traditional dynamic measurement error prediction,this study employs the support vector machine(SVM)to predict the dynamic measurement error.The cuckoo search(CS)algorithm and firefly algorithm(FA)are adopted to optimize the key parameters to avoid the local minimum value which can occurs when using the traditional method of parameter optimization.It provides a theoretical basis for error prediction modeling of practical high precision grating measuring system(2)Research on error tracing method of dynamic measurement;Based on whole-system dynamic accuracy,the decomposition method of dynamic measurement error based on Empirical Mode Decomposition(EMD)and tracing method by linear neural network are proposed.The EMD is adopted to decompose the measuring errors of the dynamic systems into the component errors and analysis of the various error sources of module further back to the sensor system internally generated the error.The feasibility of the proposed method is verified by the simulation measurement system,which provides a theoretical basis for the decomposition and tracing of the actual grating measurement system.(3)Dynamic measurement error tracing and prediction modeling of high precision sensing system;The error sources of high precision grating measurement system are analyzed comprehensively from three aspects of light,electricity,and environment.Error signal of measurement system is acquired by contrast method.Using CS-SVM and FA-SVM method to predicting the error signal acquisition,and the results are also compared with those obtained from the SVM optimized by a grid search(GS)and the particle swarm optimization(PSO)method.The experiments show that the prediction model proposed in this paper has certain advantages.The error signal is decomposed into the individual error by EMD,the error spectrum is obtained by the Hilbert transform.Each single signal by fitting the linear neural network,according to the characteristics of each error source of the grating measurement system,,the individual and the system error and its spectrum corresponding to each other,to achieve the error tracing of high precision grating measurement system.