The Research Of Adaptive Control Strategy In NdFeB Hydrogen Decropitation

With the advantage of high magnetic energy product, coercive force and high energydensity, the NdFeB permanent magnetic materials was widely used in modern industry andelectronic technology. As a technology of crushing NdFeB alloy, hydrogen is a nonlinear,multi variable and difficult to measure the degree of alloy crushing.The quality of NdFeBmaterial decrepitation relates to the permanent magnetic material and the final productperformance.At present, the production enterprises of the domestic and foreign neodymiumiron boron material are based on the reference curve of the Japanese SHIMADZU set valuefor computer control.But the optimization of the hydrogen crushing technology is still in itsinitial stage.For the uncertainty of the established mechanism model of the absorbhydrogen, the dynamic mechanism model of absorb hydrogen process on the basis of in-depth study is proposed to use enhance the learning of neural network optimization methodand Volterra series adaptive optimization method based on NLMS algorithm for thecomplex nonlinear process.These two methods are used to optimize the mechanism model and output expectedvalue of the hydrogen content, temperature and pressure of the alloy. Finally, the validity ofthe two models is verified through the experiments.In order to solve the problem of thehydrogen absorption and hydrogen crushing process in NdFeB alloy,the adaptive controlstrategy of hydrogen decrepitation which is based on predictive control is designed byapplying intelligent optimization control theory and method. LM algorithm is proposed tooptimize the predictive control algorithm of RBF neural network. Using the RBF neuralnetwork, the predictive control algorithm can be used for solving operation. Finally, theset value curve of specific conditions of the target is given by using the adaptive controlstrategy.In addition, this paper uses the VC++6.0to finished the design of the NdFeBhydrogen decrepitation adaptive control system, realizes the adaptive control algorithm byMATLAB for setting values, realizes the data exchange by the COM interface technology, and realizes the real-time display of the set value curve. The lower computer takes PLC asthe core, realizes the real-time control of the temperature and the pressure by the datacommunication of the serial port with the upper computer.In this paper, these two kinds of optimization method of mechanism model and thepredictive control are used in the NdFeB hydrogen decrepiting process. The goal settingvalue curve can be obtained online under the specific conditions of production, by usingthe adaptive control strategy. With the application of the computer automatic co ntrolsystem and the application of the adaptive control strategy in the actual production, reducesthe cost of the whole production process and improves the production efficiency.