Random effects and latent processes approaches for analyzing binary longitudinal data with missingness: a comparison of approaches using opiate clinical trial data

The analysis of longitudinal data with non-ignorable missingness remains an active area in biostatistics research. This article discusses various random effects and latent process models which have been proposed for analyzing longitudinal binary data subject to both non-ignorable intermittent missing data and dropout. These models account for non-ignorable missingness by introducing random effects or a latent process which is shared between the response model and the model for the missing-data mechanism. We discuss various random effects and latent processes approaches and compare these approaches with analyses from an opiate clinical trial data set, which had high proportion of intermittent missingness and dropout. We also compare these random effect and latent process approaches with other methods for accounting for non-ignorable missingness using this data set.     

Production Scheduling Using Deep Reinforcement Learning and Discrete Event Simulation

Scheduling in the process industry is a highly demanding task. Having access to optimal production schedules at short notice, for instance, after spontaneous changes, offers numerous advantages in terms of robustness, economics, and ultimately customer satisfaction, as delays are minimized. In this work, we describe our initial efforts to apply and evaluate deep reinforcement learning (DRL) for optimized scheduling in a typical fill-and-finish batch production plant in the chemical industry. Our pilot study demonstrates how DRL can be implemented using an approach based on discrete event simulation. We discuss the results and benefits of DRL, compare it to mathematical programming approaches, and outline a potential path forward. Our study suggests that the application of DRL in the chemical industry is a promising research direction and that DRL can complement established methods such as process simulation and mathematical programming.     

油管内空心杆捞筒研制及其应用

为了解决修井作业过程中由于稠油井杆断脱所造成的作业施工中带来的不方便和不利因素,研制了油管内空心杆捞筒.该捞筒能够很好的代替原有的打捞工具,解决了原有打捞器牙片易碎、易脱且成功率低、成本高、不易购买等问题,为企业节省了成本.     

Computer-aided multivariable control system design for processes with time delays

A computer-aided control system design package is described which handles processes with multiple time delays and allows subsequent design of interaction compensators. The package synthesizes control system designs for complex problems in a short time while also providing dynamic simulation for the purpose of evaluating controller performance. Example problems are presented along with a case study on sensitivity to errors in the delays.     

Decoupling internal model control for multivariable systems with multiple time delays

In this paper, the decoupling internal model control (IMC) with stability is investigated for multivariable stable processes with multiple time delays. All the stabilizing IMC controllers which solve this decoupling problem and the resulting closed-loop systems are characterized in terms of the open-loop system's time delays and non-minimum phase zeros. It shows that the inclusion of some time delays and non-minimum phase zeros might be necessary to make a decoupling solution realizable and stabilizing. Based on this characterization, a control design method for best achievable performance is presented. However, owing to the high complexity of the theoretical controller, a practical controller design procedure is developed with the help of the proposed model reduction algorithm. Examples are given to illustrate our analysis and design. Significant performance improvement over the existing multivariable Smith predictor control has been achieved with the proposed approach.     

On the use of Redundant Pyrotechnic Delays

The performance of redundant pyrotechnic delays is analyzed. It is found that the average delay time of a redundant cluster of delays is shorter than the corresponding average of the individual delays. Also, the standard deviation of the cluster delay time is smaller than the corresponding individual delay standard deviation. These two effects are found to be monotonic functions of the number of delays in the cluster. The effect of the delay reliability on the cluster statistics is shown to be qualitatively similar to a reduction in the effective cluster size.     

FIR model identification of multirate processes with random delays using EM algorithm

The motivation for this article comes from our development of soft sensors for chemical processes where several challenges are encountered. For example, quality variables in chemical processes are often measured off‐line through laboratory analysis. Collection of samples and subsequent analyses inevitably introduce uncertain time delays associated with the irregularly sampled quality variables, which add significant difficulty in identification of process with multirate (MR) data. Considering the MR system with random sampling delays described by a finite impulse response (FIR) model, an Expectation–Maximization (EM)‐based algorithm to estimate its parameters along with the time delays is developed. Based on the identified FIR model, two algorithms are proposed to recover the approximate output error (OE) or transfer function model. Two simulation examples as well as a pilot‐scale experiment are provided to illustrate the effectiveness of the proposed methods. © 2013 American Institute of Chemical Engineers AIChE J, 59: 4124–4132, 2013     

Proportion integral-type active disturbance rejection generalized predictive control for distillation process based on grey wolf optimization parameter tuning

The high-purity distillation column system is strongly nonlinear and coupled, which makes it difficult to control. Active disturbance rejection control (ADRC) has been widely used in distillation systems, but it has limitations in controlling distillation systems with large time delays since ADRC employs ESO and feedback control law to estimate the total disturbance of the system without considering the large time delays. This paper designs a proportion integral-type active disturbance rejection generalized predictive control (PI-ADRGPC) algorithm to control the distillation column system with large time delay. It replaces the PD controller in ADRC with a proportion integral-type generalized predictive control (PI-GPC), thereby improving the performance of control systems with large time delays. Since the proposed controller has many parameters and is difficult to tune, this paper proposes to use the grey wolf optimization (GWO) to tune these parameters, whose structure can also be used by other intelligent optimization algorithms. The performance of GWO tuned PI-ADRGPC is compared with the control performance of GWO tuned ADRC method, multi-verse optimizer (MVO) tuned PI-ADRGPC and MVO tuned ADRC. The simulation results show that the proposed strategy can track reference well and has a good disturbance rejection performance.     

A new method to detect and quantify correlated alarms with occurrence delays

This paper proposes a new method to detect correlated alarms and quantify the correlation level to improve the management of industrial alarm systems. The method is mainly composed of three parts. First, a so-called occurrence delay is defined as the main cause leading to erroneous conclusions from existing methods to detect correlated alarms. In order to tolerate the presence of occurrence delays, a mechanism is presented to generate continuous-valued pseudo alarm signals. Second, a novel approach is given to estimate the correlation delay between alarm signals, so that the correlation delay can be separated from occurrence delays to obtain real occurrence delays (ROD). Third, a statistical test based on the ROD is proposed to determine whether two alarm signals are correlated or not, and the Pearson's correlation coefficient is applied to quantify the correlation level. Numerical examples and industrial case studies are provided to support the proposed method.     

基于趋势相似度分析的多重时滞辨识及其在加氢裂化流程中的应用

针对复杂工业流程生产单元间变量存在多重时滞且检测困难,提出一种基于趋势相似度分析的多重时滞辨识方法。选取单元间相关性强的关键变量,利用多项式最小二乘拟合后的变量导数数据定义趋势相似度,以经采样时滞平移后的趋势相似度最小,描述多重时滞辨识问题;用L2范数量化趋势相似度向量,将多重时滞辨识问题转化成L2范数最小化问题;并用改进的自适应粒子群算法快速寻优,确定各变量的最优采样时滞。所提方法被应用于加氢裂化流程中,辨识出各变量的实际采样时滞,由此建立了基于局部加权核主元回归的柴油闪点预测模型。实验结果表明：考虑多重时滞的预测模型准确率提高了19.05%,验证了所提时滞辨识方法的有效性。     

Distillation calculations using a locally paramaterized continuation method

Distillation models are large scale nonlinear systems of equations which represent heat and material balances and phase equilibria for fractionation systems. These models are often solved by Newton's method or one of its variants. Unfortunately, the method may not converge unless the starting point or initial guess for the solution is quite close to the solution which is sought.

To resolve this difficulty, at least partially, we have turned to a convergence technique called the locally parameterized continuation method. Continuation is a mechanism for tracing a path along which lie the solutions of a family of problems. In the case of distillation, the starting point on the path corresponds to the solution of an easy, but somewhat unrealistic, distillation model, and the stopping point is the solution of the difficult more realistic model, which was of principal concern. The family of problems, and hence the path, is essentially generated by the variation of an articifial or physical continuation parameter. The continuation process then corresponds to the projection of a tangent along the path, a Newton correction to find the next point on the path and so on, until the terminus is reached. Turns are accomodated through the interchange of the continuation parameter and the appropriate member of the state vector.     

基于asNMPC方法的非线性网络化实时反馈控制

在网络化控制系统中,网络的介入带来了一些不利于系统控制的问题,如数据传输中普遍存在的随机时变延时,数据丢失和数据时序颠倒等。在预测控制思想的基础上,针对前馈通道和反馈通道同时存在网络传输环节的情况,采用asNMPC(advanced-step nonlinear model predictive control)方法实现了对于非线性网络化控制系统中上述问题的补偿控制。模型失配情况下与DMC(dynamic matrix control)进行比较,仿真结果表明了asNMPC方法对于非线性网络化控制系统的有效性。     

Effect of Variables on Ignition Delays of hydrazone/nitric acid systems

The effect of relative amounts of fuel and oxidizer, composition of the oxidizer and compactness of the fuel on ignition delays of hydrazones with nitric acid has been studied. Several of the hydrazones give minimum ignition delays at the mixture ratio close to the stoichiometric ratio. Ignition delays of phenyl- and dimethylhydrazones are greatly influenced by the acid composition. Longer ignition delays are observed with the acids having higher concentrations of NO₂. A reaction mechanism leading to hypergolic ignition has been proposed for phenylhydrazones/nitric acid systems. Longer ignition delays are observed when the fuels are taken in compacted forms.     

An improved control vector iteration approach for nonlinear dynamic optimization. II. Problems with path constraints简

This paper considers dealing with path constraints in the framework of the improved control vector iteration （CVI） approach. Two available ways for enforcing equality path constraints are presented, which can be directly incorporated into the improved CVI approach. Inequality path constraints are much more difficult to deal with, even for small scale problems, because the time intervals where the inequality path constraints are active are unknown in advance. To overcome the challenge, the ll penalty function and a novel smoothing technique are in-troduced, leading to a new effective approach. Moreover, on the basis of the relevant theorems, a numerical algo-rithm is proposed for nonlinear dynamic optimization problems with inequality path constraints. Results obtained from the classic batch reaCtor operation problem are in agreement with the literature reoorts, and the comoutational efficiency is also high.     

Communication delays and data losses in distributed adaptive high‐gain EKF

In this work, we consider distributed adaptive high‐gain extended Kalman filtering for nonlinear systems subject to data losses and delays in communications. Specifically, we consider a class of nonlinear systems that consist of several subsystems interacting with each other via their states. A local adaptive high‐gain extended Kalman filter is designed for each subsystem and the distributed estimators communicate to exchange the information. Each subsystem estimator takes the advantage of a predictor accounting for the delays and data losses simultaneously. The predictor of each subsystem is used to generate state predictions of interacting subsystems for interaction compensation. To get a reliable prediction, the predictors are designed based on a prediction‐update algorithm. The convergence of the proposed distributed state estimation is ensured under sufficient conditions handling communication delays and data losses. Finally, a chemical process example is used to evaluate the applicability and effectiveness of the proposed design. © 2016 American Institute of Chemical Engineers AIChE J, 62: 4321–4333, 2016     

Initiation of chain and thermal explosions by the reactor surface. Criterion for the participation of branching chains in a thermal explosion

The combustion of hydrogen and silane is studied. It is established that the chain initiation reaction on quartz in the zone of hydrogen and silane combustion is manifested as an autocatalytic reaction which is able to initiate a chain explosion and participate in the initiation of a thermal explosion. It is shown that in the case of an oxyhydrogen gas, the assumption of a branching-chain nature of the third limit is inconsistent with Semenov’s law, which includes double exponential dependences of the chain reaction rate on time and temperature. A criterion for the participation of branching chains in complex processes is proposed based on the presence or absence of short delays of a thermal explosion (≈1 sec). According to the criterion, the explosion of an oxyhydrogen gas at atmospheric pressure with delays markedly exceeding 1 sec proceeds without the participation of branching chains and is consistently explained by the joint action of autocatalytic processes on the reactor wall and gas-phase processes. __________ Translated from Fizika Goreniya i Vzryva, Vol. 43, No. 5, pp. 44–51, September–October, 2007.     

Homotopy continuation methods for computer-aided process design

Homotopy continuation methods have been used by the authors and others to solve difficult chemical engineering flowsheeting and design problems involving the solution of simultaneous nonlinear equations. Such methods can fail when: (1) the homotopy path, which one follows from the solution of a simple problem to the solution of the original (difficult) problem, returns to a second solution of the simple problem; (2) the homotopy path strikes an interior boundary of the domain of definition of the original (vector-valued) function; and (3) the homotopy path goes off to infinity. For the first two modes of failure, the use of an affine homotopy is discussed here as a possible remedy. Failure due to an unbounded homotopy path is the subject of current research.     

Control of multivariable systems containing time delays using a multivariable smith predictor

A multivariable version of the classical Smith predictor is applied to a double effect evaporator containing time delays in the control variables and some of the output variables. As in the classical approach, the multivariable Smith predictor eliminates time delays from the characteristic equation of the closed-loop system. Simulation and experimental results for the pilot-scale evaporator are quite promising and demonstrate the feasibility of the multivariable Smith predictor for practical control problems.     

A dead time compensation approach for multirate observer design with large measurement delays

In our previous work, we developed a multirate observer design method in linear systems with asynchronous sampling based on a Luenberger observer design coupled with inter-sample predictors. In this article, the problem of multirate multidelay observer design is addressed where both asynchronous sampling and possible measurement delays are accounted for. The proposed observer adopts an available multirate observer design in the time interval between two consecutive delayed measurements. A dead time compensation approach is developed to compensate for the effect of delay and update past estimates when a delayed measurement arrives. The stability and robustness properties of the multirate observer will be preserved under nonconstant, arbitrarily large measurement delays. A mathematical example and a gas-phase polyethylene reactor example demonstrate good performance of the proposed observer in the presence of nonuniform sampling and nonconstant measurement delays. © 2018 American Institute of Chemical Engineers AIChE J, 65: 562–570, 2019