State Estimation And Torque Optimization Control For Front/rear Axle Electric Loader

The wheel loader is an important part of engineering machinery because of its large market volume and wide operation range.Compared to traditional wheel loader,front/rear axle electric wheel loader(FREWL)can achieve various technical advantages,such as strong passing performance,high working efficiency,fine economic performance,fast control response and convenient modular design.However,to realize the above advantages,the torque optimization control system for optimal torque and state estimation system for key state parameters should be added to FREWL.At present,the study of state estimation method torque optimization control method of wheel loader have no thorough and mature published literature.Hence,aiming at the structure and operation characteristics of FREWL,this thesis evaluates the key state parameters by the improved square root untrace kalman filter method,extended kalman filter method.Meanwhile,the applicable torque optimization control method for FREWL is studied based on many optimization algorithms.Finally,the operation performance and economic performance of FREWL are improved by the state estimation system and the torque optimization control system.Based on National Natural Science Foundation of China(No.51375202)and Science and Technology Development Project of Jilin provincial(No.20160101285JC),the state estimation system and torque optimization control system of FREWL are studied in this thesis,and the main research is as follows.(1)For state estimation system,the longitudinal dynamics model of FREWL is established.By a nonlinear filter,the longitudinal velocity and acceleration of FREWL are estimated,and the longitudinal slip rate of the front/rear wheel is calculated by measuring the front/rear wheel speed.Based on the longitudinal velocity,longitudinal acceleration,longitudinal slip rate,the driving torque of the front/rear axle,and the spading force of the shovel,the longitudinal position and height of the center of gravity of FREWL are estimated by applying the improved square root untrace kalman filter method.In order to prevent the calculation divergence,the actual undiverging filter algorithm is used to restrain the occurrence of data divergence.By the information of the longitudinal velocity,the longitudinal center of gravity,the height of the center of gravity,and the longitudinal slip rate,the longitudinal force of the tire is estimated by the extended kalman filter method.(2)For the torque optimization control system,the aim of optimizing distribution is to improve the operation performance and economic performance of FREWL on the premise of satisfying the driver's desired motion as much as possible.Hence,the minimal sum of the variance and weighted mean of tire load and the highest efficiency of the front/rear motor are selected as the optimization targets of torque optimization control strategy.The constraint conditions of FREWL in the torque optimization control strategy are listed in this thesis.In order to search the most suitable optimization algorithm of the four optimization algorithms for the torque-distribution optimization control system of FREWL,the above optimization targets are solved by the four optimization algorithms including Quai-newton lagrangian multiplier method(QNLM),sequential quadratic programming(SQP),adaptive genetic algorithms(AGA),and particle swarm optimization with random weighting and natural selection(PSO-RN).(3)A Matlab/Simulink-Carsim integrated simulation platform is constructed in order to verify the effectiveness of the state estimation system and torque optimization control system.The longitudinal speed,slip ratio,center of gravity position and tire longitudinal force are estimated under the running condition of bituminous road,driving condition of bumpy road,and spading condition of bumpy road.The simulation results show that the state estimation system has good estimation accuracy and robustness.The simulation and verification of torque optimization control strategy based on tire load ratio is carried out on the driving condition of bumpy road,the spading condition of bumpy road and the stacking condition of the bumpy road.At the same time,the torque optimization control strategy based on the motor total efficiency is simulated and verified while driving straightly on the bituminous road and reciprocating on bumpy road.Compared with the no-control of FREWL,the torque optimization control system of FREWL can effectively improve operation performance and economic performance,and reduce the production of tire slippage.Considering the optimization results and solving time,SQP is the most appropriate optimization algorithm for the FREWL experimental prototype.In addition,state estimation system and torque optimization control system are suitable for most operation and driving conditions of FREWL after detailed discussion.(4)Comprehensively considering the key factors such as the size,quality and reliability and so on,mechanical systems and hydraulic systems of FREWL are designed.Then,the experimental prototype of FREWL is set up.The experimental prototype of FREWL installs wheel sensor,GPS/INS integrated navigation system and vehicle rapid control prototype system and other data measurement,acquisition and control system.The quality and center position of the experimental prototype of FREWL are measured by static measurement method.The state estimation methods of longitudinal velocity,center of gravity position and tire longitudinal force are verified experimentally by constant torque condition and acceleration/deceleration condition.The experimental verification of torque optimization control method is executed by no-loaded small power condition,no-loaded high power condition and heavy-loaded high power condition.The experimental results show that the state estimation system has high estimation precision,and can effectively satisfy the precision requirements of torque optimization control system.Compared with total motor efficiency of no-control of FREWL,the total motor efficiency by torque optimization control strategy of FREWL increases by 4.1-19.0% in various conditions.In this thesis,the state estimation system and torque optimization control system of FREWL are studied.The experiment platform of FREWL is built,and the necessity effectiveness of the state estimation method and the torque optimization control method to improve operation performance and economic performance are verified through the integrated simulation platform and experiment platform.