Research Of Adaptive Backstepping Control Method Of Hybrid Convey Mechanism Based On Extreme Learning Machine

The hybrid automobile electrophoretic coating conveying mechanism effectively combines the parallel and serial mechanisms,which solves the problems of the traditional conveying mechanism such as pollution of the car body,poor bearing capacity,and low flexible productivity of multiple vehicle models,and improves the quality of electrophoretic coating.The control system is the core of the hybrid mechanism,and its performance will directly affect the quality of production and painting.However,the hybrid mechanism has the characteristics of multi-parameters,complex nonlinearity,and high coupling.It is difficult to obtain an accurate dynamic model and make high performance control.At the same time,the modeling errors of the theoretical model and the actual mechanism,as well as the uncertainty of the flow of the electrophoresis coating liquid and external random disturbances in the actual operating environment,bring difficulty to the high-performance control of the hybrid mechanism.In addition,after considering the dynamic characteristics of the drive motor,there is a mismatched disturbance in the hybrid mechanism that is not in the same channel as the control voltage of the drive motor,which brings challenges to the precise control of the hybrid mechanism.Therefore,this paper takes the electrophoretic coating hybrid convey mechanism as the research object,and uses a combination of theoretical analysis,simulation test,and prototype test to study the trajectory tracking control method of the hybrid convey mechanism under matched and mismatched disturbances respectively.The engineering application of the hybrid conveyor mechanism lays a theoretical foundation.At first,this thesis gives an overview of the hybrid mechanism and its control methods.Afert that,the structure and principle of the hybrid mechanism are introduced.The mechanism is analyzed for forward and inverse kinematics,the inverse kinematics solution and the Jacobian matrix are solved,and the obtained inverse kinematics is simulated using MATLAB.Verify its correctness.Then,based on the characteristics of the hybrid mechanism with few degrees of freedom,on the basis of kinematic analysis,the Lagrangian method is used to construct the dynamic model of the flip lifting mechanism,and the generalized force is converted into the driving force of each driving joint to realize the mechanism actual torque control,and the uncertainties such as model inaccurate parameters and external random disturbances are defined as lumped interference during modeling.A dynamic model of hybrid mechanism with lumped interference is proposed.And use MATLAB to simulate and analyze the constructed model to verify the validity of the model.After that,in response to the uncertain complex nonlinear characteristics of the hybrid mechanism,a backstepping method is used to design the controller.Based on the constructed dynamic equations,the ELM is used to approximate and compensate the lumped interference in the model,and then the ELM-based adaptive backstepping controller realizes the tracking of the expected trajectory(position,speed)of the flipping and lifting mechanism,and then uses MATLAB software to build a simulation model to verify the effect of the controller.The simulation results show that the designed controller can stabilize the hybrid mechanism accurately tracking a given trajectory,the ELM network effectively suppresses the effects of uncertainty and external interference in the system,while improving convergence speed,tracking accuracy,and controller robustness.Then,for the above-mentioned hybrid mechanism dynamic control,the characteristics of the driving motor are not considered,and due to the mismatch interference between the total interference such as friction and external interference and the control input in different channels,based on the adaptive backstepping controller,two ELM networks are used for approximation and compensation for matched and mismatched disturbances respectively,then an ELM adaptive backstepping controller with mismatched interference compensation is proposed.At the same time,a genetic algorithm is used to optimize the parameters of the adaptive backstepping controller and determine the stabilization coefficients and adaptive parameters of the backstepping subsystems.The MATLAB simulation results show that the designed ELM adaptive backstepping controller with mismatch interference compensation can achieve accurate trajectory tracking of the hybrid mechanism and effectively suppress the matched and mismatched disturbances existing in the system and improve the stability and robustness of the system.Finally,for this electrophoretic coating conveying mechanism,a distributed control system with PC host computer + PMAC motion controller as the core was constructed.The motion control experiment steps are introduced and the motion control experiment of the hybrid mechanism is carried out,which further verifies the feasibility and effectiveness of the adaptive backstepping controller based on extreme learning machine designed in this paper.