SWSi-Xin Wen

Si-Xin Wen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Si-Xin Wen has a total of 25 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Real-Time Offset-Free Neural Network MPC Framework for Aeroengine Control on Embedded Systems

    by Wen-Tao Li, Si-Xin Wen, Xue-Fang Wang, Wan-Zhi Meng, Xi-Ming Sun

    Model predictive control (MPC) for aeroengines requires accurate prediction of complex nonlinear dynamics, which is challenging to achieve using traditional modeling approaches. While neural networks (NNs) offer strong nonlinear approximation capability, their embedded deployment suffers from approximation and quantization errors, leading to persistent steady-state offsets in closed-loop control. To address these limitations, this article proposes a real-time offset-free MPC framework that incorporates an NN-based prediction model with a multivariable adaptive error compensator to eliminate steady-state deviations and enhance tracking performance. Moreover, to ensure real-time feasibility, a low-rank approximation linearization method is developed to reduce computational complexity, and the entire framework is hardware-accelerated on a Zynq deep-learning processing unit (DPU) for embedded execution. Finally, hardware-in-the-loop (HIL) experiments on realistic aeroengine control scenarios demonstrate that the proposed approach achieves a faster dynamic response, improved steady-state accuracy, and a lower computation latency compared with conventional MPC, confirming its potential for practical aerospace applications.


  • $H_{\infty}$ Antidisturbance Event-Triggered Control for Aeroengine Systems via Switched Affine Models

    by Ying Zhao, Xinyu Ni, Sixin Wen, Ben Niu

    This study is focused on the issue of $H_{\infty }$ anti-disturbance event-triggered (ADET) control for aero-engine control systems (AECS) under multiple disturbances. Firstly, a switched affine model is employed to describe the AECS, taking into account the degradation of aero-engine performance and the phenomenon of multi-mode operation. Secondly, a disturbance observer is established to compensate for disturbances that arise from the modeling error, uncertainties and other factors. Then, based on the disturbance observer, a dynamic event-triggered mechanism (DETM) with respect to the affine term (AT) and an $H_{\infty }$ ADET controller are built to optimize network transmission resources and fuel flow efficiency. Furthermore, a sufficient condition is developed to enable the solvability of the $H_{\infty }$ ADET control issue for AECSs. Finally, a simulation platform of hardware-in-the-loop (SPHITL) for the AECS based on switched affine model is implemented to illustrate the effectiveness of the present $H_{\infty }$ ADET control methodology.


  • Actuator Fault-Tolerant Control for Aero-Engine Control System: A Zonotope-Based Approach

    by Shui Fu, Wentao Tang, Rui Wang, Si-Xin Wen, Xi-Ming Sun

    Efficient fault-tolerant control (FTC) is necessary for the safe operation of aero-engine control system. In this paper, a high performance active FTC method based on zonotope for actuator fault in aero-engine control systems is proposed. Parameter uncertainties are considered to describe linearization error and identification error of system model for reducing the gap between theory and practice. Firstly, a zonotopic observer satisfying the peak-bounded index is proposed to reduce the influence of uncertainties and improve the accuracy of fault estimation. Moreover, with the aid of the zonotopic observer, the range of the sliding surface affected by the estimation errors and model uncertainties can be evaluated, and the dynamic quasi-sliding mode domain (QSMD) can be obtained. As a result, the dynamic QSMD can help design the parameters of the sliding mode fault-tolerant controller, ensure the stability and convergence of the entire closed-loop control system. Meanwhile, the conservative problem caused by manual parameters setting is avoided. Finally, the feasibility of the proposed method is verified by the aero-engine Hardware-in-the-loop (HIL) experiment platform.


  • Thrust Estimation, Control, and Evaluation for Turbofan Engines

    by Si-Xin Wen, Yu Zhang, Jijun Li, Ke Wang, Kun-Zhi Liu, Xi-Ming Sun

    Given that thrust is the most core parameter of turbofan engines but cannot be measured, and that engines are subject to performance degradation, disturbances, or even failure, this paper presents a direct thrust control (DTC) system based on an onboard model and robust control. To overcome the limitations of the unscented Kalman filter (UKF) in terms of accuracy and convergence speed, the Levenberg-Marquardt (L-M) algorithm is employed to optimize the measurement update process, thereby establishing a superior onboard model for turbofan engines. Then, to address the real-time issue, the LSTM network is trained to approximate the nonlinear engine model used for the Kalman observer, substantially accelerating the computational efficiency. Finally, based on this onboard model, an H∞ gain-scheduled controller is designed for the DTC system. The comparative results of numerical simulations and hardware experiments on the Xavier embedded controller substantiate the feasibility, superiority, and real-time performance of our approaches, thus suggesting the potential for practical engineering.


  • Switching Anti-windup Synthesis for Linear Systems Subject to Asymmetric Input Magnitude and Rate Saturation: Application to Aircraft Engines

    by Ke Wang, Si-Xin Wen, Yong-Feng Gao, Xiangkui Zhang, Xi-Ming Sun

    This article investigates the anti-windup synthesis problem for linear, time-invariant systems subject to asymmetric input magnitude and rate saturation (MRS) under a switching control framework. By developing an MRS model, the original system is augmented as a linear system with magnitude-only saturation. We then partition the input space into several regions based on the signs of the control inputs. The original asymmetric saturation function can be equivalently expressed as a symmetric one in each of these regions, and the saturation levels associated with different regions are different. By utilizing these special properties, we design a separate anti-windup gain for each region of the input space. The switching from one anti-windup gain to another is activated when the input signals leave one region for another. In comparison with the single gain anti-windup design, this switching anti-windup design can lead to better performance, since the switching among multiple anti-windup gains makes fully use of the input capacity. Finally, the proposed method is applied to the aircraft engines. Hardware-in-the-loop testing demonstrates its superiority and practicality.


  • Practical Offset-Free Model Predictive Control and Its Embedded Application to Aeroengines

    by Si-Xin Wen, Zhuo-Rui Pan, Kun-Zhi Liu, Xiangkui Zhang, Xi-Ming Sun

    Model predictive control (MPC) is popular in applications with slow dynamics because of its advantages in handling constraints and multivariable optimization. But for aeroengines, it is difficult to obtain an exact prediction model, which will lead to offsets in tracking. Besides, deploying MPC to embedded controllers for real-time control is a well-known challenge. Therefore, this paper presents a switched linear MPC, which incorporates the augmented prediction models with error integrator for offset-free tracking, the sparse-based quadratic programming formula for solving MPC, and a reset strategy for achieving bumpless transfer at the switching instant. Further, on the hardware board we developed, six hardware-related acceleration strategies are explored and evaluated for real-time performance. Then, eight cases of five objects are tested, whose results indicate a significant speedup of around 50 times. At last, the hardware-in-the-loop tests of the turbofan engine and the real bench tests of the micro-turbojet engine are performed, which verifies the superiority, real-time performance, and potential for practical applications. Note to Practitioners —This paper was motivated by the problem of applying the offset-free MPC to the embedded control system for aeroengines. The difficulty of accurate modelling and the requirement to compute the embedded MPC in a limited time are two major challenges. Accordingly, we explore six hardware acceleration strategies that are often ignored to ensure the real-time performance. Moreover, we investigate the switched MPC to obtain the desired dynamic response. However, the controller switching tends to induce fluctuations that are harmful to the engine. Hence, we present a reset strategy to ensure bumpless transfer performance. Preliminary physical experiments suggest that the proposed approaches are feasible to achieve the predetermined objectives. Further, the comparisons with the previous methods highlight our superiority. Overall, this paper contributes to optimize the computational performance of microcontrollers and enhance bumpless transfer performance for MPC. In our future work, the proposed approaches will be applied to flight experiments and more industrial devices.


  • Open Access

    Aeroengine thrust estimation and embedded verification based on improved temporal convolutional network

    by Wanzhi Meng, Zhuorui Pan, Sixin Wen, Pan Qin, Ximing Sun


  • Data-Driven Switched Model Predictive Control Without Terminal Ingredients

    by Zhi-Min Wang, Kun-Zhi Liu, Si-Xin Wen, Xi-Ming Sun

    In this paper, we propose a switched model predictive control scheme based on the data-driven method without any terminal ingredients. Compared with conventional model-based model predictive control schemes, we utilize a series of input-output data to describe the system dynamics according to the behavioral systems theory. In order to improve system performance, multiple cost functions regulated by a suitable switching strategy are considered. For the nominal case with noise-free data, we prove that the proposed data-driven switched MPC schemes can ensure the exponential stability of the system provided that the switching signal satisfies the average dwell-time condition. Moreover, the robust stability is also analysed mathematically for the robust case considering bounded measurement noise. The effectiveness of the control algorithm is verified by the speed control of a commercial high bypass turbofan engine on a hardware-in-loop platform. The results of the experiment show the advantages of the proposed switched model predictive control scheme over conventional non-switched model predictive control schemes. Note to Practitioners —In this paper, we consider the problem of designing a control law to improve the performance of the system with multiple requirements that need a tradeoff. Since the performance criteria can easily be formulated in the cost function in model predictive control algorithm, it is a natural thought to design multiple cost functions which can be switched according to different performance requirements. However, the modelling process of the system is usually complicated and time-consuming for conventional model predictive control. In this paper, the data-driven method replaces the counterpart of the model in conventional model predictive control. The input-output measurements of the system can be directly used to forecast the dynamics of the system. The feasibility and stability of the control scheme are analyzed mathematically. We show that the proposed control law can ensure the above closed-loop performance of the controlled system, provided that the parameters of the controller are suitable and the switch of the cost function is sufficiently slow. The effectiveness of the proposed control scheme is demonstrated by the application of a high bypass ratio turbofan engine on a hardware-in-loop experiment platform. The results show that the turbofan engine can achieve a relatively rapid response with smaller overshoot.