A comparison of nonlinear control performance assessment techniques for nonlinear processes

Assessing the quality of industrial control loops is an important auditing task for the control engineer. However, there are complications when considering the ubiquitous nonlinearities present in many industrial control loops. If one simply ignores these nonlinearities, there is the danger of over‐estimating the performance of the control loop in rejecting disturbances and thereby possibly overlooking loops that need attention. To address this problem, several techniques have been recently developed to extend the control performance assessment (CPA) of single input/single output linear systems to nonlinear systems. This article surveys these nonlinear CPA techniques and compares their performances using three case studies. These results can be used to guide control engineers to select the most suitable CPA techniques for their particular applications. © 2012 Canadian Society for Chemical Engineering     

A convolutional neural network (CNN)-based direct method to detect stiction in control valves

Control valves are considered important capital assets in any process industry. A properly maintained control valve can have a significant impact on how well the process is controlled as well as the overall cost of the plant. However, control valves can suffer from poor control performance due to valve non-linearities. One of the main reasons for non-linearity is control valve stiction. Stiction not only causes oscillations in the process variables but also shortens the life of the control valve, resulting in an economic loss for the process. In a process plant, a control engineer generally analyzes the time series plot of process value (PV), set point (SP), and controller output (OP) data and identifies stiction based on the typical shape pattern of PV/SP/OP plot. In this study, the same shape pattern methodology is adapted to identify stiction using convolutional neural network (CNN) technique. A one-dimensional convolution neural network (Conv1D) algorithm is developed, which works directly on PV/SP/OP time series data for stiction detection. The proposed CNN algorithm is tested on both simulated and industrial control loop data. The suggested method provides promising results with a combined stiction prediction accuracy of 92% (92.2% in predicting non-sticky and 91.53% in predicting sticky loops) for the industrial loops data studied.     

Control valve stiction detection using Markov transition field and deep convolutional neural network

Control valve stiction is an industrial problem that often causes oscillations in process control loops. Oscillating control loops are not capable of maintaining key process variables near or at their desired values, thus yielding low-quality products, inducing economic loss, and increasing environmental impacts. Therefore, it is of vital importance to detect stiction in industrial control valves. In this regard, the present work proposes a new method based on the Markov transition field and convolutional neural network (CNN) to identify sticky control valves in industrial control loops. The Markov transition field is employed to convert process variable (PV) and controller output (OP) into two-dimensional images, which are then utilized by CNN to learn to distinguish stiction induced oscillations from oscillations brought out by a non-stiction condition. A transfer learning strategy is adopted to improve the stiction detection capability of the proposed method. Its performance is evaluated via its application to benchmark control loops taken from the chemical, paper, mining, and metal industries. Results demonstrate that the proposed method obtains the correct verdict for the majority of the control loops studied.     

Data mining and control loop performance assessment: The multivariate case

Control loop performance assessment (CLPA) techniques assume that the data being analyzed is generated during steady state operation with fixed plant dynamics and controller parameters. However, in industrial settings one often encounters environmental and feedstock variations which can induce significant changes in the plant dynamics. Availability of data sets corresponding to fixed configurations is therefore questionable in industrial scenarios, in which case it becomes imperative to extract the same from routine plant operating data. This article proposes a technique for segmenting multivariate control loop data into portions corresponding to fixed steady state operation of the system. The proposed technique exploits the fact that changes in the operating region of the system lead to changes in variance‐covariance matrix of multivariate control loop data. The univariate interval halving technique is fused with Mahalanobis distance to develop a multivariate tool that accounts for interactions between variables. The resulting data segments can be used for reliable CLPA and/or for user defined benchmarking of control loops. A multivariate control loop performance index is also proposed that requires significantly less data as compared to one of the previously proposed techniques. The proposed technique requires only routine operating data from the plant, and is tested on benchmark systems in the literature with simulations. Experimental validation on a model predictive control system aimed at maintaining the temperature profile of a metal plate demonstrates applicability of the technique to industrial systems. The proposed technique acts as a tool for preprocessing data relevant to CLPA and can be applied to large scale interacting multivariate systems. © 2017 American Institute of Chemical Engineers AIChE J, 63: 3311–3328, 2017     

Identification and control of an industrial steam-reforming plant

Reliable mathematical models for an industrial steam-gas reformer and a steam-boiler are simulated and used to study the dynamic behavior of the two coupled processes. Simulation results of the identified system are in good agreement with the nonlinear process operation. Three control loops are considered for multivariable control system design. These are Hydrogen product temperature and quality, and boiler water level. A new multivariable control structure is obtained, which manipulates steam-to-carbon (S/C) ratio for the control of coil outlet temperature (COT), the fuel gas rate to control hydrogen product quality (conversion) and the boiler feed water to control drum level. The selected structure is then tuned using the Biggest-Log modulus-Tuning (BLT) method. Results show a very satisfactory response of the temperature and quality loops with the BLT based controllers. It is also found that the boiler water level loop is partially decoupled from the other two loops and hence does not need to be detuned according to BLT criterion. The new multivariable control structure is compared with the conventional control utilising fuel gas rate to control COT. Closed-loop simulation results show a better performance for the multivariable structure under continuous operation.     

一种对预测控制系统经济性能评估的实用方法及其工业应用

In this article, an approach for economic performance assessment of model predictive control (MPC) system is presented. The method builds on steady-state economic optimization techniques and uses the linear quadratic Gaussian (LQG) benchmark other than conventional minimum variance control (MVC) to estimate the potential of reduction in variance. The LQG control is a more practical performance benchmark compared to MVC for performance assessment since it considers input variance and output variance, and it thus provides a desired basis for determining the theoretical maximum economic benefit potential arising from variability reduction. Combining the LQG benchmark directly with benefit potential of MPC control system, both the economic benefit and the op-timal operation condition can be obtained by solving the economic optimization problem. The proposed algorithm is illustrated by simulated example as well as application to economic performance assessment of an industrial model predictive control system.     

A novel framework for integrating data mining with control loop performance assessment

Data driven control loop performance assessment techniques assume that the data being analyzed correspond to single plant‐controller configuration. However, in an industrial setting where processes are affected due to the presence of feedstock variability and drifts, the plant‐controller configuration changes with time. Also, user‐defined benchmarking of control loops (common in industrial plants) requires that the data corresponding to optimal operation of the controller be known. However, such information might not be available beforehand in which case it is necessary to extract the same from routine plant operating data. A technique that addresses these fundamental requirements for ensuring reliable performance assessment is proposed. The proposed technique performs a recursive binary segmentation of the data and makes use of the fact that changes in controller settings translate to variations in plant output for identifying regions corresponding to single plant‐controller configurations. The statistical properties of the data in each such window are then compared with the theoretically expected behavior to extract the data corresponding to optimal configuration. This approach has been applied on: (1) raw plant output, (2) Hurst exponent, and (3) minimum variance index of the process data. Simulation examples demonstrate the applicability of proposed approach in industrial settings. A comparison of the three routes is provided with regard to the amount of data needed and the efficacy achieved. Key results are emphasized and a framework for applying this technique is described. This tool is of significance to industries interested in an automated analysis of large scale control loop data for multiple process variables that is otherwise left unutilized. © 2015 American Institute of Chemical Engineers AIChE J, 62: 146–165, 2016     

基于加权偏离度统计方法的预测控制性能评估算法

针对带区域约束条件的预测控制系统性能评估问题,在考虑过程输出变量约束类型的基础上,提出了基于加权偏离度统计方法的控制性能评估算法。该方法依据控制要求的不同,将输出变量分为质量变量和约束变量,并结合工程经验合理选择变量的权重。基于系统闭环运行数据和约束设置,通过计算变量的加权偏离度得到控制系统的性能评估指标,从而为预测控制器的参数调整和性能提升提供了决策依据。系统仿真实例和工程应用证明了该评估算法对区域预测控制系统性能评估的有效性。     

基于广义最小变差基准的多变量控制性能评估方法

This paper is concerned with the control performance assessment based on the multivariable generalized minimum variance benchmark. An explicit expression for the feedback controller-invariant (the generalized minimum variance) term of the multi-variable control system is obtained, which is used as a standard benchmark for the assessment of the control performance for multi input multi output (MIMO) process. The proposed approach is based on the multivariable minimum variance benchmark. In com-parison with the minimum variance benchmark, the developed method is more reasonable and practical for the control performance assessment of multivariable systems. The approach is illustrated by a simulation example and an industrial application.     

Performance assessment of cascade controllers for nitrate control in a wastewater treatment process

A cascade control strategy is proposed to the benchmark simulation model 1 (BSM1) to enhance the treatment performance of nitrogen removal in a biological wastewater treatment plant. The proposed control approach consists of two control loops, a primary outer loop and a secondary inner loop. The method has two controllers of which the primary loop has a model predictive control (MPC) controller and the secondary loop has a proportional-integralderivative (PID) controller, which is a cascade MPC-PID controller. The primary MPC controller is to control the nitrate concentration in the effluent, and the secondary PID controller is to control the nitrate concentration in the final anoxic compartment. The proposed method controls the nitrate concentrations in the effluent as well as in the final anoxic reactor simultaneously to strictly satisfy the quality of the effluent as well as to remove the effects of disturbances more quickly by manipulating the external carbon dosage rate. Because the control performance assessment (CPA) technique has the features of determining the capability of the current controller and locating the best achievable performance, the other novelty of this paper is to suggest a relative closed-loop potential index (RCPI) which updates the CPA technology into a closed-loop cascade controller. The proposed method is compared with a cascade PID-PID control strategy and the original PID controller in BSM1 and an improved performance of the suggested cascade MPC-PID controller is obtained by using the CPA approach.     

Enhancing dynamic data reconciliation performance through time delays identification

In order to improve the performance of data reconciliation methods, an efficient Genetic algorithm (GA) for determining time delays has been developed. Delays are identified by searching the maximum correlation among the process variables. The delay vector (DV) is integrated within a dynamic data reconciliation (DDR) procedure based on Kalman filter through the measurements error model. The proposed approach can be satisfactorily applied not only off-line but also on-line. It was firstly validated in a dynamic process with recycles and feedback control loops. Then, the methodology was successfully applied to a highly non-linear and complex challenging control case study, the Tenessee Eastman benchmark process, demonstrating its robustness in complex industrial problems. This case study required to implement an extended Kalman filter to deal with the existing non-linearities.     

Control performance monitoring — a review and assessment

In this paper we present an overview of current status of control performance monitoring using minimum variance principles. Extensions to PID-achievable performance assessment, trade-off between performance and robustness, and trade-off between deterministic and stochastic performance objectives are discussed. Future directions are pointed out for research and practice with regard to root-cause diagnosis, plant-wide performance assessment, multivariable assessment, adequacy assessment of existing control strategies, performance assessment of model predictive control, and the use of intelligent field devices and artificial intelligence to form a systematic diagnostic methodology. A brief tutorial on performance assessment is given in the appendix with an industrial process example.     

Performance assessment of multivariate control loops on a paper-machine headbox

A large number of control loops in the pulp and paper industry are multivariate in nature. The main control objective of these loops is to reduce process variation. The satisfactory performance of these loops is therefore important for maintaining product quality. This paper introduces a multivariate performance measure of such control loops. The multivariate minimum variance control performance which provides an absolute lower bound on process variance can be estimated from routine operating data, and serves as a good benchmark to assess control loop performance. The proposed measure of performance is evaluated by application to the multiloop control of a headbox machine.     

Measurements of dynamic behavior of a multistage flash water desalination system

This study focuses on measuring the process dynamics for the multistage flash desalination process (MSF) in an industrial unit with a capacity of 4546 m³/d. This is a novel addition to the literature because previous studies are limited to theoretical analysis of process dynamics or controller tuning, as well as conceptual design of conventional or advanced control systems. The measurements evaluate the performance of seven control loops, which include the pressure, temperature, and flow rate of the heating steam; the pressure of the vacuum ejector; and flow rates of the brine recycle, make-up seawater, and cooling seawater. All measurements start from steady-state conditions. The system is then set on manual where all control units are disengaged. Subsequently, only one control valve is adjusted by ± 15% of its steady-state setting. A total of 14 experiments were performed involving simultaneous measurements of the system variables. Measurements showed non-linear behavior where increasing or decreasing the valve settings did not provide similar trends. Analysis of results shows that one of the most sensitive variables is the distillate level in the last stage: the distillate trays either were flooded or became dry. The brine level in the last flashing stage was also found to be sensitive to valve settings where level increase resulted in higher product salinity. The results and analysis presented provide a better understanding in system fault analysis which could be caused by improper operating conditions. These data are essential to propose, design, and evaluate advanced/comprehensive control systems for the MSF process.     

Performance monitoring of industrial controllers based on the predictability of controller behavior

This paper focuses on performance assessment of industrial controllers. Instead of using process or controller models, it is based on process data collected at regular time intervals. Data analysis includes a set of tests that are reviewed in the paper and implemented in a software system. A methodology based on the concept of the predictability of controller errors is also proposed for performance monitoring. It considers the time series of the error and verifies the existence of predictable patterns beyond the control horizon in each one of the controlled variables of the process. The result of the analysis is given as a performance index. Examples using industrial data from a refinery are provided.     

The importance of dynamic performance and testing of industrial thermometers in process control

This paper considers the dynamic performance of instruments in control systems. Errors connected with transient measurements are discussed. It is shown that it is not always proper to assign common statistical properties to these dynamic errors. Emphasis is placed on temperature sensors. Common industrial thermometers are described and factors affecting their response discussed. An a-priori dynamic performance prediction procedure is presented. This permits preinstallation estimation of response characteristics of sensors used in industrial processes. Sensors in service also require performance assessment since they can deteriorate with use. In-situ testing is useful for this purpose and a recently develolped procedure is described.     

网络控制系统与网络化串级控制系统的结构分析

基于实际工业过程控制,引入了网络化串级控制系统的概念。针对网络控制系统中存在多个网络及多个控制回路,提出了节点设备连接阵和网络传输阵的概念。分析网络控制系统的三种典型结构形式,并用系统配置图、方框图以及节点设备连接阵和网络传输阵等方法描述了这三种不同结构的网络控制系统。在此基础上指出网络化串级控制系统的四种典型结构,并分别采用这三种方法进行描述。对网络控制系统和网络化串级控制系统结构的分析和描述,为系统的进一步分析和设计奠定了坚实的基础。     

化工过程多回路PID控制系统模式切换参数自整定

化工过程一般为多变量系统,但其主要控制方案为分散多回路PID常规控制。由于多变量系统内部存在不同程度的耦合作用,各控制回路之间存在相互影响,当其他回路进行手动/自动模式切换时,本回路等效被控对象将会发生突变,导致本回路的原有控制参数不能适应等效被控对象的变化,造成控制性能下降,甚至闭环系统不稳定。为避免这种情况的发生,从整个系统的角度研究控制回路模式切换时的稳定性,采用多变量频域Nyquist阵列设计法。基于对角优势下正Nyquist稳定性判据,从Gershgorin圆边界点的角度定量分析各个控制回路在模式切换前后的稳定性变化程度,从而确定各回路控制器增益的调整方向及程度,实现各回路的控制器参数在控制回路模式切换瞬间的自动整定,尽可能抵消控制回路模式切换对整个系统的扰动,保证整个系统的闭环稳定性。以Shell公司重油分馏塔的多回路PID控制系统为例,将3个PID控制回路依次投用时,根据Gershgorin圆边界点进行控制参数的自整定,闭环系统仍能保持一定的控制性能,否则闭环系统将不稳定。     

A real time approach to process control education with an open-ended simulation tool

Process simulators are widely used in industrial process designs and academic research. These simulation tools are also perfectly suitable for the process dynamics and control education of junior chemical engineers and students, as these tools mimetically help them with comprehending the basic theories of process control, such as process capacity, dead time, control loops, controllers and multi-unit processes such as distillation columns. At the University of Auckland, New Zealand, final year Chemical and Materials Engineering students who participate in the process dynamics and control paper are required to complete a series of simulation workshops in auxiliary sessions to help them in their understanding of process dynamics and control. This paper introduces the content of the workshops as well as reviews the student feedback on the introduction of the simulator and their perceptions of their learning of process dynamics and control as assisted by the software and instruction. Three case studies are provided in this paper to illustrate the benefits of running workshops. The motivation of this paper is to share our workshop design with other universities.     

Control Loop Design and Control Performance Study on Direct Internal Reforming Solid Oxide Fuel Cell

A solid oxide fuel cell (SOFC) stack is a complicated nonlinear power system. Its system model includes a set of partial differential equations that describe species, mass, momentum and energy conservation, as well as the electrochemical reaction models. The validation and verification of the control system by experiment is very expensive and difficult. Based on the distributed and lumped model of a one‐dimensional SOFC, the dynamic performance with different control loops for SOFC is investigated. The simulation result proves that the control system is appropriate and feasible, and can effectively satisfy the requirement of variable load power demand. This simulation model not only can prevent some latent dangers of the fuel cell system but also predict the distributed parameters' characteristics inside the SOFC system.