基于积分型切换函数的自适应神经网络控制

针对一类具有未知函数控制增益的非线性系统,利用RBF神经网络的逼近能力,依据滑模控制原理,提出了一种直接自适应神经网络控制器设计新方案。通过引入积分型切换函数及逼近误差自适应补偿项,监督控制用饱和函数代替符号函数,根据李雅普诺夫稳定性理论,证明了闭环系统是全局稳定的,跟踪误差收敛到零。该算法应用于连续搅拌型化学反应器CSTR（Continuous Stirred Tank Reactor）,仿真结果显示,该算法能很好地使CSTR跟踪给定的温度信号,表明了该控制策略的有效性。     

Intelligent process control using neural fuzzy techniques

In this paper, we combine the advantages of fuzzy logic and neural network techniques to develop an intelligent control system for processes having complex, unknown and uncertain dynamics. In the proposed scheme, a neural fuzzy controller (NFC), which is constructed by an equivalent four-layer connectionist network, is adopted as the process feedback controller. With a derived learning algorithm, the NFC is able to learn to control a process adaptively by updating the fuzzy rules and the membership functions. To identify the input–output dynamic behavior of an unknown plant and therefore give a reference signal to the NFC, a shape-tunable neural network with an error back-propagation algorithm is implemented. As a case study, we implemented the proposed algorithm to the direct adaptive control of an open-loop unstable nonlinear CSTR. Some important issues were studied extensively. Simulation comparison with a conventional static fuzzy controller was also performed. Extensive simulation results show that the proposed scheme appears to be a promising approach to the intelligent control of complex and unknown plants, which is directly operational and does not require any a priori system information.     

A nonlinear control scheme for imprecisely known processes using the sliding mode and neural fuzzy techniques

This article considers the regulation control of nonlinear chemical processes whose dynamics are imprecisely known. A nonlinear control scheme that incorporates a sliding mode controller (SMC) and a neural fuzzy strategy is proposed to deal with this kind of processes. The sliding mode controller designed on the base of a previously known process model is implemented to keep system’s trajectory around the desired manifold. For extra and/or unknown dynamics that cannot be captured before the SMC design stage, an intelligent scheme of utilizing a neural fuzzy strategy is then used to provide an adaptive ability to accommodate the perturbation, which therefore is able to force the system output back to and maintain in the desired set point. The effectiveness and applicability of the proposed scheme are demonstrated through the control of a continuous stirred tank reactor with existing simultaneously the unmodeled side reactions, measuring error, and extra matched and unmatched disturbances. The potential use of a sliding observer along with the proposed scheme is also investigated in the work. Extensive simulation results reveal that the incorporation of the model-based SMC and the intelligent neural fuzzy technique appears to be an effective and promising approach to the nonlinear control of chemical processes whose dynamics are imprecisely known.     

连续CMAC-RLS在CSTR系统中的仿真研究

基于RLS算法的连续CMAC是在对CMAC的结构进行相应改变的基础上提出的,收敛速度快,收敛性得到了证明,可以有效的提高CMAC的性能,但连续CMAC-RLS算法的仿真和应用研究还比较少,有待于更多的发展.将RLS算法引入CMAC网络从而构成的基于RLS算法的连续CMAC网络,并给出了完整的连续CMAC-RLS算法的数学描述.随后讨论了CSTR系统,并采用基于RLS算法的连续CMAC对具有代表性的强非线性系统CSTR进行建模.仿真结果表明,关于连续CMAC网络对强非线性系统具有良好的建模效果,可应用于较复杂的工业系统建模.     

Dynamic synergistic modelling for cobalt removal process in zinc hydrometallurgy and the research of parameter estimation based on unscented Kalman filter

The cobalt removal process with arsenic salt of zinc hydrometallurgy has serious non-linearity, uncertainty, and mutual coupling. Its accurate dynamic modelling has always been a challenging problem. On the basis of in-depth analysis of cobalt removal process and reaction mechanism, considering the cascade relationship between the reactors, a dynamic synergistic continuously stirred tank reactor (SCSTR) mechanism model of the cobalt removal process was constructed. Aiming at the unknown parameters in the SCSTR model, the idea of the Kalman filter was introduced, and the unknown parameters were characterized as unknown states; a method of estimating the unknown model parameters was developed using the augmented state equation and the unscented Kalman filter (UKF) algorithm. Simulation results with industrial data of a zinc smeltery showed that the parameter estimation model has high accuracy, and the estimated parameters can be used in the SCSTR model. An intensive simulation analysis of the dynamic characteristics of the complete SCSTR model was carried out to verify the influence of different input disturbances on the output ion concentration of each reactor, which demonstrated the excellent dynamic performance and potential of the model. Ultimately, according to the industrial calculation analysis, the SCSTR model has a guiding effect on the addition of zinc powder in the reactors, overcomes the blindness in the production process, and provides a momentous basis for the optimization control of the cobalt removal process.     

融合浸出过程流程特性的梯形LSTM终点pH值模型预测控制

在锌冶炼浸出过程中，pH值的稳定控制是影响生产成本和产品质量的关键。针对浸出过程连续反应搅拌釜(CSTR)机理复杂、入口条件波动大、反应大滞后的问题，提出了融合流程特性的梯形长短期记忆网络模型预测控制(TLSTM-MPC)算法。首先，分析了影响浸出过程pH值稳定控制的关键因素。结合CSTR的级联特性，提出一种表征CSTR物理特性的梯形长短期记忆网络结构(TLSTM)模型，使预测模型具有较好的物理可解释性。针对入口条件波动和反应过程大滞后的问题，提出了TLSTM-MPC算法，并采用粒子群算法进行优化求解。最后，采用某大型浸出过程现场运行数据进行验证。结果表明，该算法的超调量和调节时间相对传统控制方法分别降低了51.2%和57.4%，且在入口波动的情况下能够快速的稳定pH值。可见，该方法对浸出过程的稳定、高质量生产具有较大的工程实践应用价值。     

Neuro-adaptive practical prescribed-time control for pure-feedback nonlinear systems without accurate initial errors

This article studies the problem of practical prescribed-time tracking for pure-feedback nonlinear systems, where the transient behavior, steady-state precision, settling time as well as the rate of convergence can be preset irrespective of initial conditions. With the help of a time-dependent function and a state-dependent function, a simple coordinate transformation is established to construct a neuro-adaptive controller to achieve the tracking purpose. In particular, the proposed controller does not need to update the design parameters in accordance with different initial conditions, which makes the controller more versatile. The effectiveness of the proposed method is illustrated through two practical systems, namely, a continuous stirred tank reactor (CSTR), and a robotic system.