Search

Summary: As a deep neural networks (DNNs) model compression method, learning‐compression (LC) algorithm based on pre‐trained models and matrix decomposition increases training time and ignores the structural information of models. In this manuscript, a tensor decomposition‐based direct LC (TDLC) algorithm without pre‐trained models is proposed. In TDLC, the pre‐trained model is eliminated, and tensor decomposition is first applied to LC algorithm to preserve the structural features of the model. There are two key steps in TDLC. An optimal rank selection method is first proposed in compression‐step (C‐step) of TDLC to find global optimal ranks of tensor decomposition. Second, TDLC utilizes cyclical learning rate, which is different from traditional monotonically learning rates schedule, to improve the generalization performance of uncompressed models in learning‐step (L‐step). TDLC obtains the optimal compression model by alternately optimizing L‐step and C‐step. TDLC is compared with 16 state‐of‐the‐art compression methods in experiments part. Extensive experimental results show that TDLC produces high‐accuracy compression models with high compression rate. Comparing with TDLC‐pre‐trained, TDLC notably achieves 30% training time shorten and 11% parameter reduction in Resnet32, while improving accuracy by 0.2%. [ABSTRACT FROM AUTHOR]

Summary: This article deals with state estimation of complex nonlinear discrete fractional‐order systems with unknown noise statistics by means of an adaptive fractional‐order Unscented Kalman filter (AFUKF). Firstly, in order to alleviate the communication burden of fractional‐order Unscented Kalman filter, short‐term memory effect is utilized to decide an appropriate memory length. Then aiming at the problem of filtering divergence and accuracy degradation caused by unknown statistical characteristics of noise, based on the maximum a posterior (MAP) principle, a noise statistical estimator is introduced to estimate and correct the statistical characteristics of noise in real‐time. Finally, the unbiasedness of the proposed algorithm is analyzed to verify that the estimated mean and covariance of noise are unbiased. The effectiveness and accuracy of AFUKF are demonstrated via simulation experiments. [ABSTRACT FROM AUTHOR]

Supercapacitors undergo wide terminal voltage fluctuations, which makes it difficult to maintain a stable bus voltage and develop a satisfactory dynamic response. To solve this problem, a fast adaptive bus voltage regulation strategy is proposed in this paper. Firstly, the dual‐loop controller with an external voltage loop and an internal current loop is applied to the supercapacitor energy storage systems. Secondly, an adaptive proportional–integral controller is designed for the outer voltage loop to accommodate the variations of the system characteristics. Furthermore, in order to improve the dynamic response, the set‐point automatic modulation is integrated into the voltage control loop to compensate the stochastic load demanded fluctuations. The simulation and experimental results show that the proposed method ensures the stability of the bus voltage and provides a lower fluctuation and shorter settling time compared with the conventional method. [ABSTRACT FROM AUTHOR]

Pneumatic brake systems are crucial for the operation of trains. However, due to switching characteristics of on/off solenoid valves, the precise pressure control and low switching activities of valves are difficult to guarantee simultaneously during braking. To address the issue, a hybrid model predictive control (MPC) method is proposed for implementing the multi‐objective optimisation in this paper, i.e. the precise pressure tracking and the valve switching reduction. In order to model the hybrid behaviour caused by continuous dynamics of the pressure and discrete features of solenoid valves, the mixed logic dynamical system theory is applied. Then, the braking control problem is recast as a hybrid MPC problem, which jointly optimises the pressure tracking precision and the valve switching. A mixed integer linear program is formulated to solve the MPC problem in an efficient way. Both simulation and experiment results show that the accurate pressure control and low switching activities of valves can be achieved by the proposed method when compared with existing methods. [ABSTRACT FROM AUTHOR]