Adaptive Efficiency Optimization Control of VLF-PM Motors Considering Operation Uncertainties With DCEE

For the mechatronic system of variable leakage flux-permanent magnet (VLF-PM) motors, motor parameters that vary with uncertain operating conditions lead to reduced control efficiency, which fails to meet the requirements of energy management. To solve this problem, an adaptive efficiency optimization control (EOC) considering operation uncertainties with enhanced dual control for exploration and exploitation (DCEE) is innovatively proposed. First, an N-dimensional flux linkage estimator is designed, which enables the quantification of the uncertainty of parameter variations caused by external operating conditions. Additionally, an optimization rule with dual effects is established based on the estimated N-dimensional flux linkage distribution, allowing the optimal current angle to adapt to different external operating conditions. With this method, the impact caused by errors in traditional parameter estimation is overcome. Meanwhile, model-based active uncertainty learning avoids the training of large amounts of data and eliminates a significant computational burden. Finally, a comparison is made with existing EOC methods, which proves the effectiveness and advantages of the proposed method.