For the MIMO nonlinear systems, a multivariable ORVFL neural network adaptive predictive control algorithm based on Improved Sparrow Search Algorithm was proposed in this paper. The algorithm uses the ORVFL network to approximate the nonlinear system model, and applies to the multi?step prediction of the system process. In order to improve the performance of the Sparrow Search Algorithm, the algorithm is used to optimize the system performance index online and solve the optimal control law of each sampling period. The results show that the algorithm has good control performance and good anti?model mismatch ability.

Based on the differential lever principle, a multi?stage compliant magnifying hinge was designed, and a micro?positioning platform with three translational degrees of freedom was assembled in an orthogonal way, and the symmetrical closed?loop structure formed has the characteristics of shock resistance, no cumulative error and better isotropy. The material is 7075 aluminum, and the stacked piezoelectric ceramics were used as the driving device to provide input displacement. Combined with the path search method and the static matrix method, the static balance equation and geometric deformation equation of the mechanism were obtained; the sensitivity analysis of the structural parameters was carried out, and the parameters with higher sensitivity in the mechanism are optimized with the displacement amplification ratio as the goal; a finite element analysis was performed on the assembly. The results shows that the stroke can reach 391.42 μm and the magnification is 13.05, and the platform has a high displacement magnification ratio and good positioning accuracy.

Pipeline transportation is the main means to transport oil and gas and other important materials, and the damage of buried pipelines by earthquake is difficult to repair. Based on the nonlinear dynamic theory, the ANSYS finite element software was used to establish the pipeline?soil model, and then the nonlinear contact model between pipeline and soil was established. Finally, different seismic waves were applied to the buried pipeline and compared. The results show that the upper and lower parts of the pipeline have large displacements under the influence of seismic waves; as the input seismic waves gradually increase, the displacement of the lower part of the pipeline decreases, but it is still greater than the displacement on both sides of the pipeline. It can be seen that the upper and lower parts of the pipeline will be greatly affected when an earthquake occurs.

The compliant microgrippers are directly contacted with the gripper in the microoperation field and during microassembly.And as the end microactuator in the micro?operation and micro?assembly system,it plays an important role in realizing micro?operation and micro?assembly tasks.A flexible microgripper mechanism based on straight circular flexure hinge and reed isosceles trapezoidal flexure hinge was studied in this paper.The displacement amplification and parallel displacement of gripper displacement were realized by piezoelectric ceramic driving.The mechanical pseudo?rigid body model of the flexible microgripper was established, and the kinematics model was established by using the pseudo?rigid body model method.Furthermore, the displacement amplification factor was calculated, and the relationship among the input force, input displacement and output displacement was obtained.The finite element method was used to simulate and verify the model.The results show that the prediction of theoretical values is consistent with the simulation results.

Taking the 6?PTRT parallel as the research object, the pose error model of 6?PTRT parallel robot is established, the closed?loop vector method of single branch chain is used to establish the error equation based on its input and output relationship. According to the pose error model of 6?PTRT parallel robot,the mechanism error is converted into the drive rod error,the influence of the error parameters of each drive rod length on the output pose error is analyzed by using MATLAB software;the objective function of pose error correction of 6?PTRT parallel robot is established, and the adaptive weighted particle swarm optimization algorithm based on shrinkage factor is used to optimize the error parameters of each drive rod, modify the terminal pose, and improve the kinematics accuracy, which provides a theoretical basis for the dynamics, pose calibration and trajectory planning and control of 6?PTRT parallel robot.