Torque Distribution Optimization Control For The Distributed Electric Loader

In recent years,new opportunities and challenges have been presented in energy saving and emission reduction related policies for developing the new energy construction machinery industry,the development trend of electrification is strong.Wheel loaders are the leading construction machinery product with a wide operation range and significant market share.Carbon emissions and tailpipe pollution from conventional fuel wheel loaders are avoided when using electric wheel loaders.The research of key technologies for electric wheel loaders is necessary and has significant application value in strategic development,technology reserves,and supply markets.Wheel loaders operate in harsh environments and on rugged roads,with significant load changes before and after when shoveling and loading operations,as well as drastic changes in front and rear axle loads when carrying and walking.Therefore,given the operational characteristics of wheel loaders,the study of optimal control of drive torque distribution can effectively improve the whole machine’s performance.In this thesis torque distribution optimization control is carried out to improve the performance of the distributed electric drive wheel loader(DEDWL),based on National Natural Science Foundation of China “Force Torque Distribution and Coordinated Control of Articulated Special Vehicle with Distribution Electric Drive System under Unstructured Terrain”(grant No.51875239)and State Key Laboratory of Smart Manufacturing for Special Vehicles and Transmission System“Dynamics Control Study of Distributed Electro-driven Articulated Special Vehicles based on State Estimation”(grant No.GZ2018KF005).Front and rear axle distributed electric drive wheel loaders are taken as the object of study,and the operational characteristics of the wheel loaders are combined.For the two typical operating conditions of load walking and shoveling operations,a state estimation method for the core parameters of the DEDWL and an optimized control strategy for drive force distribution are proposed.The primary research of this thesis is as follows:The state estimation method of core parameters of DEDWL is presented.Based on the adaptive Kalman filtering algorithm,the wheel slip state is identified,and the longitudinal vehicle speed state is estimated.In order to obtain information on the kinematic and tire force variables of the whole machine in real-time,a hierarchical state estimation strategy is established using a unscented Kalman filter algorithm modified by singular value decomposition and a strong tracking algorithm.In addition,the strategy of system noise segmentation is introduced,and the state estimation method of shovel mechanics variables is studied.The core parameter state estimation study provides state information for the subsequent torque distribution control.An optimized control strategy for torque distribution for the loaded travel conditions of the DEDWL is proposed.The characteristics of the load travel operation are analyzed,two control modes are divided based on road surface characteristics,load and speed information,and control modes are discriminated based on variable information from sensors and state estimation.The drive consistency optimal control model is characterized by the consistency of the drive,the minimization of the slip rate difference is set as the optimal control objective,the model-free adaptive control algorithm is employed,and the optimal distribution coefficients are solved.The economic optimal control model is characterized by the tire dissipation energy,the tire dissipation power is minimized as the optimal control objective,the Lagrange multiplier method is introduced,and the optimal distribution coefficient is found.The optimal control strategy of torque distribution for shovel operation conditions is proposed.The two core objectives of drive anti-skid control and dynamic optimization are determined by collecting shovel operation process data through actual vehicle tests and analyzing the optimized control requirements.Conventional anti-skid control schemes and optimal control schemes based on intelligent algorithms are studied separately.A conventional algorithmic allocation optimization strategy is established using logical thresholds,and drive anti-slip control is implemented.The tire loading rate is set as the optimal control objective based on the estimated shovel force in the study using intelligent algorithms.The simulated annealing particle swarm algorithm is used to solve the distribution coefficient,and the antiskid post-processing is established to avoid the loss of traction due to skidding.Ultimately,the power and economy of the DEDWL are enhanced.The effectiveness of the proposed optimal control strategy for drive torque distribution of DEDWL is validated from three aspects: model simulation validation,real-vehicle test validation,and model simulation validation incorporating real-vehicle test data.The results show that the proposed state estimation strategy is better estimated with high accuracy and follow-through.With the torque distribution control strategy proposed in this thesis,the drive consistency and economy of the distributed electric drive wheel loader are improved in the load-travel conditions.The traction force is optimized,and the drive anti-skid is realized in the shovel operation conditions.