Chenkun Yin

Terminal iterative learning control based station stop control of a train

The terminal iterative learning control (TILC) method is introduced for the first time into the field of train station stop control and three TILC-based algorithms are proposed in this study. The TILC-based train station stop control approach utilises the terminal stop position error in previous braking process to update the current control profile. The initial braking position, or the braking force, or their combination is chosen as the control input, and corresponding learning law is developed. The terminal stop position error of each algorithm is guaranteed to converge to a small region related with the initial offset of braking position with rigorous analysis. The validity of the proposed algorithms is verified by illustrative numerical examples.

A High-Order Internal Model Based Iterative Learning Control Scheme for Nonlinear Systems With Time-Iteration-Varying Parameters

In this technical note, we propose a new iterative learning control (ILC) scheme for nonlinear systems with parametric uncertainties that are temporally and iteratively varying. The time-varying characteristics of the parameters are described by a set of unknown basis functions that can be any continuous functions. The iteratively varying characteristics of the parameters are described by a high-order internal model (HOIM) that is essentially an auto-regression model in the iteration domain. The new parametric learning law with HOIM is designed to effectively handle the unknown basis functions. The method of composite energy function is used to derive convergence properties of the HOIM-based ILC, namely the pointwise convergence along the time axis and asymptotic convergence along the iteration axis. Comparing with existing ILC schemes, the HOIM-based ILC can deal with nonlinear systems with more generic parametric uncertainties that may not be repeatable along the iteration axis. The validity of the HOIM-based ILC under identical initialization condition (i.i.c.) and the alignment condition is also explored.

Data-Driven Modeling for UGI Gasification Processes via an Enhanced Genetic BP Neural Network With Link Switches

In this brief, an enhanced genetic back-propagation neural network with link switches (EGA-BPNN-LS) is proposed to address a data-driven modeling problem for gasification processes inside United Gas Improvement (UGI) gasifiers. The online-measured temperature of crude gas produced during the gasification processes plays a dominant role in the syngas industry; however, it is difficult to model temperature dynamics via first principles due to the practical complexity of the gasification process, especially as reflected by severe changes in the gas temperature resulting from infrequent manipulations of the gasifier in practice. The proposed data-driven modeling approach, EGA-BPNN-LS, incorporates an NN-LS, an EGA, and the Levenberg-Marquardt (LM) algorithm. The approach cannot only learn the relationships between the control input and the system output from historical data using an optimized network structure through a combination of EGA and NN-LS but also makes use of the networks gradient information via the LM algorithm. EGA-BPNN-LS is applied to a set of data collected from the field to model the UGI gasification processes, and the effectiveness of EGA-BPNN-LS is verified.

Iterative learning control based automatic train operation with iteration-varying parameter

An iterative learning control (ILC) based on Automatic Train Operation (ATO) is proposed to address train speed tracking problem under iteration-varying operation condition, that is, the air resistance coefficients of the train at any two consecutive iterations are completely different. Using the data measured by the sensors, repetitive requirement of traditional ILC is partially relaxed in the proposed method. The effectiveness of the proposed method is verified by theoretical analysis and numerical simulation.

Iterative learning control for train trajectory tracking under speed constrains with iteration-varying parameter:

An iterative learning control based on Automatic Train Operation is proposed for repetitively running train to deal with trajectory tracking problem under iteration-varying operation condition and certain speed constrains. Iteration-varying operation condition considered in this paper focuses on the air resistance coefficient of the train, which may be completely different at any two consecutive operation processes due to different weather conditions. To eliminate the influence of the resistance coefficient caused by the different weather conditions, data measured by the temperature sensors is used. In order to meet the speed constrains, the Barrier Composite Energy Function scheme is used to guarantee that the speed constraints are not violated. The effectiveness of the proposed method is verified by theoretical analysis and numerical simulation.

Iterative learning control design with high-order internal model for nonlinear systems

In this work we focus on iterative learning control (ILC) design for tracking iteration-varying reference trajectories that are generated by high-order internal models (HOIM). An HOIM can be formulated as a polynomial operator between consecutive iterations to describe the changes of desired trajectories in the iteration domain. The classical ILC for tracking iteration-invariant reference trajectories, on the other hand, is a special case of HOIM where the polynomial renders to a unity coefficient or a special first order internal model. By inserting the HOIM into P-type ILC, the tracking performance along the iteration axis is investigated for a class of continuous-time nonlinear systems. Utilizing of conventional time-weighted norm method guarantees validity of proposed algorithm in a sense of data-driven control.

An ILC scheme for a class of nonlinear systems with time-varying parameters subject to second-order internal model

In this paper, we propose a new iterative learning control (ILC) scheme, which is devoted to dealing with unknown parameters that are both time varying and iteration varying. In particular, we consider iteration-varying parameters that are generated by a second-order internal model. By incorporating the internal model into the parametric learning law, the ILC scheme can handle more generic nonlinear systems and more generic parametric uncertainties, comparing with existing ILC schemes that are first order in essence. We further explore the conditions under which the new ILC scheme can guarantee learning convergence. Utilizing the information of previous two iterations and the method of composite energy function (CEF), we are able to derive pointwise convergence along the time axis and asymptotic convergence along the iteration axis.

Corrections to “A High-Order Internal Model Based Iterative Learning Control Scheme for Nonlinear Systems With Time-Iteration-Varying Parameters” [Nov 10 2665-2670]

In the above titled paper (ibid., vol. 55, no. 11, pp. 2665-2670, Nov. 10), six errors were found. The corrections are presented here.

ILC based perimeter control for an urban traffic network

Macroscopic fundamental diagram (MFD) that describes traffic flow in an urban road network can be used to design perimeter control method to regulate the traffic flow from a macroscopic level. Most of the perimeter control algorithms are regarded as a kind of model-based feedback control method, whose performance is hardly to improve in practice due to the model uncertainty. By noticing the repetitive nature of urban traffic flow, an iterative learning control (ILC) based perimeter control method is proposed for an urban region. Since the repetitive information of the controlled system is fully utilized, an improved tracking performance is guaranteed by theoretical analysis, and simulation results verify the effectiveness of the proposed perimeter control method.

Data-driven modeling for fixed-bed intermittent gasification processes by enhanced lazy learning incorporated with relevance vector machine

An enhanced lazy learning approach incorporated with relevance vector machine (ELL-RVM) is proposed for modeling of the fixed-bed intermittent gasification processes inside UGI gasifiers. The online measured temperature of produced crude gas plays a dominant role during gasification processes. However, it is difficult to formulate the dynamics of gasifiers temperature via first principles due to the complexity of UGI gasification process, especially severe changes in the temperature versus infrequent manipulation of the gasifier and noise in the temperature data collected from practical fields. Noticing that the changes of some input variables of UGI gasification process are small but impactful, a novel weighted-neighbour selection method, which is based on minimizing dynamic cost functions for different outputs coordinately, is adopted to enhance the lazy learning approach. The sparseness and short test time of RVM is fully utilized in design and implementation of the proposed online modeling algorithm under the Bayesian learning framework. The effectiveness of ELL-RVM for modeling UGI gasification processes is verified by a series of experiments based on the data collected from practical fields.