Plant-wide control of the Tennessee Eastman problem
Abstract
This study focuses on the development and performance of four plant-wide control structures for the Tennessee Eastman challenge problem. The control structures are developed in a tiered fashion and without the use of a quantitative steady state or dynamic model of the process. The throughput or production rate manipulator is selected first so that it is located on the major process path. The inventory controls are arranged in an outward direction from this throughput manipulator. The four structures are described and comments are given on their effective handling of the defined disturbances and setpoint changes. One structure provides effective control under all circumstances for 50 hours of process time. The effective dynamic performance of these structures supports the strength of the tiered plant-wide control design methodology used.
References (6)
- J. Downs et al.
A plant-wide industrial process control problem
Computers chem. Engng
(1993) - P. Lyman
Plant-wide control structures for the Tennessee Eastman process
- R. Price
Design of plant-wide regulatory control systems
Cited by (473)
Statelets extraction-guided contrastive learning for detection of control performance degradation with varying degrees
2024, Advanced Engineering InformaticsDetection of control performance degradation (CPD) in time series plays a vital role in modern industry. However, most CPD detection methods tend to neglect the following recognitions: 1) control performance is often assessed over small segments of the time series rather than the long-term entire one, and 2) CPD may exhibit different degrees in practice, resulting in the time-varying data distribution problem. These two considerations necessitate the development of algorithms that can extract information-rich small segments, defined as statelets, to discriminate the control performance with varying degrees. To this end, a Statelets Extraction-based Contrastive Learning (SECL) model is proposed for the detection of CPD. First, a varying-length statelet generation algorithm is developed to split the original multivariate time series into multiple statelet candidates with different time lengths, which can avoid the problem of information-rich small segments being obscured by the long-term trend of the time series. A contrastive learning encoder is then constructed to map statelet candidates into a uniform feature space with the designed triplet loss, keeping features of similar statelets tight and features of dissimilar statelets far away. In this way, we can extract discriminative feature candidates to distinguish the control performance with varying degrees. An information entropy-based selection strategy is proposed to choose information-rich statelet features that can best identify different control performances. After that, a statelets distance-based classifier is designed by computing the match distance between features of a testing sample and the extracted statelet features. The proposed SECL model focuses on the extracted statelets that can significantly reveal changes related to control performance rather than comparing the entire time series, which is more accurate and robust in detecting CPD with varying degrees. The effectiveness of the proposed method is illustrated through a real-world thermal power plant process. The experiment results show that SECL can capture the discriminative features and achieve prominent detection performance.
EPBS_FIDMV: A fault injection and diagnosis methods validation benchmark for EPBS of EMU
2024, Control Engineering PracticeThe electro-pneumatic brake system (EPBS) is an essential part of electric multiple units. Fault diagnosis (FD) methods play an important role in the safe operation of EPBS, and have been active research direction. The literature is considerably on studies concerned with new fault diagnosis methods, while how to quantitatively validate these methods for EPBS is quite few. The main attribute is the difficulty of obtaining fault data due to the long-term normal operation. Besides, fault experiments on the real platform are hard to control, and the cost is high. Therefore, this paper presents a new open-source benchmark for validating the FD methods of EPBS. Initially, we review the existing pure datasets and benchmarks available for validating FD methods, revealing a notable absence of a benchmark specially designed for EPBS. Then, the normal and fault models of EPBS are established and implemented by co-simulation of AMESim and MATLAB/Simulink. The fault models include brake cylinder faults, train pipe faults, EP valve faults, and sensor faults. Additionally, a GUI interface is developed to visually present the validation results. The benchmark accompanying this paper is available on the website at http://gfist.csu.edu.cn/Download.html.
An industrial process fault diagnosis method based on independent slow feature analysis and stacked sparse autoencoder network
2024, Journal of the Franklin InstituteDeep learning, with its powerful multilayer nonlinear representation of deep neural networks, enables models trained based on deep learning to describe the true distribution of data more accurately and thus have better generalization capabilities. However, the training data are usually assumed to obey independent identical distribution. Industrial process data nowadays usually have complex characteristics such as non-Gaussian, nonlinear, dynamic, strong correlation, and high-dimensional, which make it difficult to meet the assumptions of deep learning. Using industrial process data directly would result in low model accuracy and bias in the model output. This paper proposes a fault diagnosis method based on independent slow feature analysis (ISFA) and stacked sparse autoencoder (SSAE) network. The statistical properties possessed by ISFA are used to perform preliminary feature extraction of industrial process data, and then the extracted preliminary features are input to the SSAE network for fine-grained feature extraction. The proposed two-step feature extraction method not only makes the data input to the SSAE network as close as possible to obey independent identical distribution and provides convincing fault diagnosis results, but also makes full use of the powerful feature extraction capability of the SSAE network, which makes the proposed method also significantly improve the detection accuracy of incipient faults. The feasibility and superiority of the proposed method is verified through the TE process.
Interpret what a Convolutional Neural Network learns for fault detection and diagnosis in process systems
2023, Journal of Process ControlThe focus of this work is on an interpretation strategy on what a Convolutional Neural Network (CNN) has learned for fault detection and diagnosis (FDD) in process systems. Frequency spectra of process variables obtained by Continuous Wavelet Transform (CWT) are adopted as input features. Then, a CNN structure is designed to represent the mappings from input frequency features to different operation conditions. The Layer-wise Relevance Propagation (LRP) strategy is utilized to gain the relevance of each frequency feature to the classification performance. The formulations of relevance propagation for 4 types of CNN layers are presented in detail. The relevance scores are then depicted in heatmaps, where the pixels’ colors denote the contribution degrees and the most significant frequency features are considered as the major bases that the CNN discriminates different operation situations. The proposed interpretation strategy is experimented on the Tennessee Eastman process benchmark. The testing results demonstrate the efficiency of the strategy in interpreting what the CNN has learned to distinguish normal or faulty conditions in the FDD task.
SensorSCAN: Self-supervised learning and deep clustering for fault diagnosis in chemical processes
2023, Artificial IntelligenceModern industrial facilities generate large volumes of raw sensor data during the production process. This data is used to monitor and control the processes and can be analyzed to detect and predict process abnormalities. Typically, the data has to be annotated by experts in order to be used in predictive modeling. However, manual annotation of large amounts of data can be difficult in industrial settings.
In this paper, we propose SensorSCAN, a novel method for unsupervised fault detection and diagnosis, designed for industrial chemical process monitoring. We demonstrate our model's performance on two publicly available datasets of the Tennessee Eastman Process with various faults. The results show that our method significantly outperforms existing approaches (+0.2-0.3 TPR for a fixed FPR) and effectively detects most of the process faults without expert annotation. Moreover, we show that the model fine-tuned on a small fraction of labeled data nearly reaches the performance of a SOTA model trained on the full dataset. We also demonstrate that our method is suitable for real-world applications where the number of faults is not known in advance. The code is available at https://github.com/AIRI-Institute/sensorscan.
An enhanced temporal algorithm- coupled optimized adaptive sparse principal component analysis methodology for fault diagnosis of chemical processes
2023, Process Safety and Environmental ProtectionPrincipal component analysis (PCA) is a classic fault diagnosis method widely used in chemical process data modeling. However, the limitation of PCA to handle dynamic and time-variant data has been progressively exposed with the ever-increasing development of artificial intelligence technologies and applications of automation systems in the chemical industry. In this paper, an enhanced temporal algorithm-coupled optimized adaptive sparse PCA (ETA-OASPCA) methodology is proposed with the aim to improve the conventional PCA method by introducing temporal state computation and dynamic adaptive sparsity. The ETA is constructed to extract the state transitions of the time-variant process data by introducing the “Transition” and “Hold”, acquire the region of interest by employing a Maximum Fuzzy c-means (MFCM) method, and then use the region time prediction automaton to label the possible abnormal states. The introduction of dynamic adaptive sparsity in PCA, achieved by presenting an iterative interior point optimization method to update the dynamic sparse penalty term, can impose sparsity constraints on dynamic data calculations, thus enhancing the interpretability of the PCA model. The effectiveness of the proposed method was validated through its application in the Eastman Tennessee process, showing that an average fault detection rate, false alarm rate, and latency of 99.4%, 0.37%, and 4.11 mins can be achieved, respectively.