Artificial pancreas under stable pulsatile MPC: Improving the closed-loop performance

This work presents a pulsatile Zone Model Predictive Control (pZMPC) for the control of blood glucose concentration (BGC) in patients with Type 1 Diabetes Mellitus (T1DM). The main novelties of the algorithm – in contrast to other existing strategies – are: (i) it controls the patient glycemia by injecting short duration insulin boluses for both, the basal and bolus infusions, in an unified manner, (ii) it performs the predictions and estimations (critical to anticipate both, hypo and hyperglycemia) based on a physiological individualized long-term model, (iii) it employs disturbance observers to compensate plant-model mismatches, (iv) it ensures, under standard assumptions, closed-loop stability, and (v) it can be used – under minor modifications – as an optimal basal–bolus calculator to emulate conventional therapies. Because of the latter characteristic, a significantly better performance is achieved, not only in terms of classical indexes (time in the normoglycemia zone, avoidance of hypoglycemia in the short term, avoidance of hyperglycemia in the long term) but also in terms of its applicability (use of the pump or injections). Such a performance is tested in a cohort of in-silico patients from the FDA-accepted UVA/Padova simulation platform, considering the most challenging scenarios.     

内模强化学习型模型预测控制及其在人工胰脏上的应用

在学习型模型预测控制的框架里,迭代学习控制器被用来更新模型预测控制器的设定点.在已经发表的研究成果里,学习型模型预测控制用到的是比例型的学习率,这种学习率的学习能力有限,而且怎样设计学习增益仍然是一个开放性问题.在本文中,基于内模控制理论提出的PID型的迭代学习控制器被用来更新模型预测控制器的设定点.为了方便起见,本文提出的结合算法可称为内模强化学习型模型预测控制.本文提出的算法应用在(1)型糖尿病人的人工胰脏闭环控制上.仿真结果显示,本算法对比于比例学习型模型预测控制可以达到更好的收敛性能,而且对非重复干扰有很好的鲁棒性.     

Advanced decision control system for effluent violations removal in wastewater treatment plants

This paper presents the application of control strategies for wastewater treatment plants with the goal of effluent limits violations removal as well as achieving a simultaneous improvement of effluent quality and reduction of operational costs. The evaluation is carried out with the Benchmark Simulation Model No. 2. The automatic selection of the suitable control strategy is based on risk detection of effluent violations by Artificial Neural Networks. Fuzzy Controllers are implemented to improve the denitrification or nitrification process based on the proposed objectives. Model Predictive Control is applied for the improvement of dissolved oxygen tracking.