Персона: Толоконский, Андрей Олегович
Email Address
Birth Date
Статус
Фамилия
Имя
Имя
Результаты поиска
Active Disturbance Rejection Control of Nuclear Pressurized Water Reactor for Power Generation
Control design for pressurized water reactor (PWR) is difficult due to associated non-linearity, modelling uncertainties and time-varying system parameters. Extended state observer (ESO) based active disturbance rejection control (ADRC) presents a simple and robust control solution which is almost model free and has few tuning parameters. However, conventional ESO suffers from noise over-amplification in the obtained estimates due to high-gain construction which in turn degrades the noise sensitivity of the closed-loop system and limits the achievable dynamic performance in practical scenarios. To overcome this problem, two recent techniques namely cascade ESO (CESO) and low-power higher-order ESO (LHESO) are implemented for control of PWR. Simulation analysis conducted in MATLAB illustrates the performance improvement obtained over conventional ESO based ADRC. Extensive simulation analysis is also conducted to investigate robustness towards parametric uncertainties. © 2022 IEEE.
Modelling and simulating of a multiple input and multiple output system to control the liquid level and temperature by using model predictive control
© Published under licence by IOP Publishing Ltd.In recent years, Model Predictive Control has been regarded as one of the most reliable advanced control methods widely used in industrial and nuclear processes. Model predictive control uses a system model to forecast its future response, and also treats a real-time optimization algorithm to choose the best control action that leads the predicted output to the reference. The overall design goal of a predictive control model is to compute the trajectory of a future manipulated variable input signal to optimize the future behavior of the plant output. Optimization is performed within a limited time window by providing information at the beginning of the time window. This paper proposes a predictive controller MPC model for multiple input-output systems. The design of the controller includes the development of a model of the state space system for the tank system, then the design and the simulation of the MPC controller for the developed model system. The variables to be controlled are the temperature and the liquid level of the tank by changing the MPC's parameters.
Adaptive second-order sliding-mode control for a pressurized water nuclear reactor in load following operation with Xenon oscillation suppression
© 2022 Elsevier B.V.This study proposes an adaptive second-order sliding mode control based on a twisting algorithm to control nuclear reactor core power during load-following power maneuvers. The control system was designed based on the concept of an extended equivalent control. The control technique does not require knowledge of the uncertainty or upper bound of the disturbance in the system. Additionally, the gain of the control system is a dynamic gain that increases or decreases according to the system requirements within a specified period. Consequently, chattering was attenuated. Chattering excites high-frequency dynamics and causes wear out of the control mechanism, making chattering a severe challenge in sliding-mode control. A nuclear reactor is a time-varying, complex, nonlinear, and constrained system. The characteristics of a nuclear reactor are a function of its operating power level, fuel burnup, and aging. In addition, the load-following mode of operation causes Xenon oscillation, which can further cause instability in the core. Therefore, the non-base load operation of the system, including load following, further aggravates uncertainty and disturbances in the system. To this end, an adaptive second-order sliding mode control that does not require knowledge of the uncertainty or the upper bound of the disturbance in the system was designed. The reactor core was modelled using an experimentally verified and validated multi-point kinetic model with four nodes. Simulation experiments were conducted using a multi-point kinetics model and an adaptive second-order sliding mode control. The results of the simulation experiments indicated that reactor core integrity is guaranteed, and the core is protected against peak power densities, such as linear heat generation rates (LHGRs) and lower departure from nucleate boiling ratios (DNBR). Moreover, the control system achieved the load-following objective and suppressed Xenon oscillation. The performance of the proposed control system was further compared with that of a twisting control system and classical proportional integral derivative control system to validate the effectiveness and reliability of the adaptive twisting second-order sliding mode control system.
Active disturbance rejection control of pressurized water reactor
Control design for pressurized water reactor (PWR) is difficult due to associated non-linearity, modelling uncertainties and time-varying system parameters. Extended state observer (ESO) based active disturbance rejection control (ADRC) presents a simple and robust solution which is almost model free and has few tuning parameters. However, conventional ESO suffers from noise over-amplification in the obtained estimates due to high-gain construction which in turn degrades the noise sensitivity of the closed-loop system and limits the achievable dynamic performance in practical scenarios. To overcome this problem, two recent techniques namely cascade ESO (CESO) and low-power higher-order ESO (LHESO) are implemented for control of PWR. Simulation analysis is conducted in MATLAB to illustrate the performance improvement obtained over conventional ESO based ADRC, particularly in case of time-varying disturbances. Extensive simulation analysis is also conducted to investigate robustness towards parametric uncertainties. The study also presents a comparison of conventional ESO, CESO and LHESO to highlight their advantages as well as limitations which in turn would help the users in selecting the appropriate scheme for their particular use-case.