Research On Control Methods Of Underactuated Double Pendulum Effects Crane

As an important construction machinery,overhead crane is widely used in manufacturing,construction,logistics and many other fields.Its main control objective is to transport goods to the target position quickly and accurately,and to ensure that the swing of goods is as small as possible without residual swing.The control dimension of the overhead crane system is less than the degree of freedom to be controlled.It is a typical underactuated system,and it is vulnerable to external disturbances such as friction,wind and so on.At the same time,the system states show strong coupling and nonlinearity.In recent years,many scholars at home and abroad have achieved a series of research results on underactuated overhead crane system,but most of the existing control methods only consider the single-pendulum effects of crane system.In many cases,such as:(1)When the weight of the hook is similar to the load mass,the quality of the hook can not be ignored;(2)When the load mass distribution is not uniform and the size of the load is too large to be considered as a particle,the crane system will exhibit double-pendulum characteristic,that is,the load will swing around the hook,resulting in a more complex dynamic model of the crane system,higher coupling between states and higher underactuation.For the above problem of anti-swing control of crane with double pendulum effects,In this paper,from the perspective of practical application,the high-performance anti-sway positioning control of the double pendulum effects crane is studied in depth,and the following effective control methods are proposed:(1)Tracking control method based on LADRC(Linear Active Disturbance Rejection Control).Aiming at the existing problems of tracking control methods for double-pendulum effects crane that cannot guarantee the bounds of the tracking error,a tracking control method based on LADRC is proposed in this dissertation.Firstly,the underactuated characteristics of the system are treated by differential flattening method,and the system is approximated to a full actuated system.Then,the uncertainty of the model and external disturbances are reduced to total disturbances by using the active disturbance rejection theory,and the total disturbances are estimated in real time to eliminate the system errors at the source.Finally,the bird swarm algorithm is applied to the parameter tuning of the controller to further optimize the control performance of the system.Through extensive of simulations,the proposed method is compared with the existing methods,which verifies its excellent antiswing positioning control performance and good robustness.(2)Grouping fuzzy control method based on weight online optimization.Aiming at the problems in the existing fuzzy control methods,two grouping fuzzy control methods based on weight online optimization are designed in this dissertation.The first method uses three sub-fuzzy controllers to control the trolley,hook and load in groups.The output of the three sub-fuzzy controllers is multiplied by their respective dynamic weighting factors,and the system output is obtained.The dynamic weighting factors are optimized online by combining the fuzzy control with the bird swarm algorithm to obtain the optimal control results of the system.The second method utilizes the advantages of sliding mode control such as strong robustness and predictability of dynamic process,applies sliding mode surface to grouping fuzzy control,effectively reduces the number of rules of the fuzzy controller,and proposes an adaptive sliding mode fuzzy control method.This method can adjust the dynamic performance in real time according to the outputs of the system,and adjust the roles of each subsystems in the system control in real time.Both methods inherit the advantages of fuzzy control.They can control the system stably without the specific model.At the same time,they can solve the complex problem of designing the fuzzy rules of multi-input system and have strong applicability.The simulation results are illustrated to validate that the two proposed grouping fuzzy control methods based on weight online optimization have good control effect.(3)Control methods based on energy analysis.Three control methods based on energy analysis are proposed for thedouble-pendulum overhead crane.In the first method,a new system parameter is constructed to enhance the coupling of the system control parameters,so as to realize the antiswing positioning control from the perspective of energy analysis and control.The second method considers the influence of wind resistance and other disturbances on the system.A disturbance rejection controller based on energy analysis is proposed to overcome the mechanical friction and the air resistance of the load and realize the antiswing positioning control.Thirdly,a simple control method based on system energy analysis is proposed for the condition that the swing angle signal can not be obtained.This method can complete antiswing positioning control under the condition of only crane position feedback,and has simple structure and strong practicability.The simulation results verify the effectiveness of the above three control methods based on system energy analysis.In this paper,the control problems of the double-pendulum effects crane are studied deeply,and some preliminary research results are obtained.Considering the complex working environment of crane in actual production,there are still many practical problems to be solved,the author will continue to make further study.