Emergent contamination warning systems are critical to ensure drinking water supply security. After detecting the existence of contaminants, identifying the types of contaminants is conducive to taking remediation measures. An online classification method for contaminants, which explored abnormal fluctuation information and the correlation between 12 water quality indicators adequately, is proposed to realize comprehensive and accurate discrimination of contaminants. Firstly, the paper utilized multi-fractal detrended fluctuation analysis (MF-DFA) to select indicators with abnormal fluctuation, used multi-fractal detrended cross-correlation analysis (MF-DCCA) to measure the cross-correlation between indicators. Subsequently, the algorithm fused the abnormal probability of each indicator and constructed the abnormal probability matrix to further judge the abnormal fluctuation of indicators using D−S evidence theory. Finally, the singularity index of the cross-correlation function and th... [more]

Ice cream viscosity is one of the properties that most changes during crystallization in scraped surface heat exchangers (SSHE), and its online measurement is not easy. Its estimation is necessary through variables that are easy to measure. The temperature and power of the stirring motor of the SSHE turn out to be this type of variable and are closely related to the viscosity. Therefore, a mathematical model based on these variables proved to be feasible. The development of this mathematical relationship involved the rheological study of the ice cream base, as well as the application of a method for its in situ melting in the rheometer as a function of the temperature, and the application of a mathematical model correlating the SSHE stirring power and the ice cream viscosity. The result was a coupled model based on both the temperature and stirring power of the SSHE, which allowed for online viscosity estimation with errors below 10% for crystallized systems with a 30% ice fraction at... [more]

To implement the quality-relevant monitoring scheme for batch processes with multiple output modes, this paper presents a novel methodology based on stochastic programming. Bringing together tools from stochastic programming and ensemble learning, the developed methodology focuses on the robust monitoring of process quality-relevant variables by taking the stochastic nature of batch process parameters explicitly into consideration. To handle the problem of missing data and lack of historical batch data, a bagging approach is introduced to generate individual quality-relevant sub-datasets, which are used to construct the corresponding monitoring sub-models. For each model, stochastic programming is used to construct an optimal quality trajectory, which is regarded as the reference for online quality monitoring. Then, for each sub-model, a corresponding control limit is obtained by computing historical residuals between the actual output and the optimal trajectory. For online monitoring,... [more]

This paper deals with the monitoring of the performance of a photovoltaic plant, without using the environmental parameters such as the solar radiation and the temperature. The main idea is to statistically compare the energy performances of the arrays constituting the PV plant. In fact, the environmental conditions affect equally all the arrays of a small-medium-size PV plant, because the extension of the plant is limited, so any comparison between the energy distributions of identical arrays is independent of the solar radiation and the cell temperature, making the proposed methodology very effective for PV plants not equipped with a weather station, as it often happens for the PV plants located in urban contexts and having a nominal peak power in the 3÷50 kWp range, typically installed on the roof of a residential or industrial building. In this case, the costs of an advanced monitoring system based on the environmental data are not justified, consequently, the weather station is of... [more]

A novel active semi-supervised learning framework using unlabeled data is proposed for fault identification in labeled expensive chemical processes. A principal component analysis (PCA) feature selection strategy is first given to calculate the weight of the variables. Secondly, the identification model is trained based on the obtained key process variables. Thirdly, the pseudo label confidence of identification model is dynamically optimized with an historical, current, and future pseudo label confidence mean. To increase the upper limit of the identification model that is self-learning with high entropy process data, active learning is used to identify process data and diagnosis fault causes by ontology. Finally, a PCA-dynamic active safe semi-supervised support vector machine (PCA-DAS4VM) for fault identification in labeled expensive chemical processes is built. The application in the Tennessee Eastman (TE) process shows that this hybrid technology is able to: (i) eliminate chemical... [more]

This article describes the use of the electrical conductivity for measuring bed expansion in a continuous anaerobic biofilm reactor in order to prevent the exit of support material from the reactor with the consequent loss of biomass. The substrate used for the tests is obtained from a two-stage anaerobic digestion (AD) process at the pilot scale that treats the liquid fraction of fruit and vegetable waste (FVW). Tests were performed with the raw substrate before anaerobic treatment (S1), the effluent from the hydrolysis reactor (S2), and the effluent from the methanogenic reactor (S3) to evaluate its effect on the electrical conductivity values and its interaction with colonized support material. The tests were carried out in a 32 L anaerobic inverse fluidized bed reactor (IFBR), which was inoculated with colonized support material and using two industrial electrodes at different column positions. The results with the previously digested samples (S2 and S3) were satisfactory to detect... [more]

The concentration of volatile fatty acids (VFAs) is one of the most important measurements for evaluating the performance of anaerobic digestion (AD) processes. In real-time applications, VFAs can be measured by dedicated sensors, which are still currently expensive and very sensitive to harsh environmental conditions. Moreover, sensors usually have a delay that is undesirable for real-time monitoring. Due to these problems, data-driven soft sensors are very attractive alternatives. This study proposes different data-driven methods for estimating reliable VFA values. We evaluated random forest (RF), artificial neural network (ANN), extreme learning machine (ELM), support vector machine (SVM) and genetic programming (GP) based on synthetic data obtained from the international water association (IWA) Benchmark Simulation Model No. 2 (BSM2). The organic load to the AD in BSM2 was modified to simulate the behavior of an anaerobic co-digestion process. The prediction and generalization perf... [more]

Early-stage fault detection and diagnosis of distillation has been considered an essential technique in the chemical industry. In this paper, fault diagnosis of a distillation column is formulated as an inverse problem. The nonlinear least squares algorithm is used to evaluate fault parameters embedded in a nonlinear dynamic model of distillation once abnormal symptoms are detected. A partial least squares regression model is built based on fault parameter history to explicitly predict the development of fault parameters. With the stripper of Tennessee Eastman process as example, this novel approach is tested for step- and random-type faults and several factors affecting its efficiency are discussed. The application result shows that the hybrid inverse problem approach gives the correct change of fault parameter at a speed far faster than the base approach with only a nonlinear model.

Kernel methods are a class of learning machines for the fast recognition of nonlinear patterns in any data set. In this paper, the applications of kernel methods for feature extraction in industrial process monitoring are systematically reviewed. First, we describe the reasons for using kernel methods and contextualize them among other machine learning tools. Second, by reviewing a total of 230 papers, this work has identified 12 major issues surrounding the use of kernel methods for nonlinear feature extraction. Each issue was discussed as to why they are important and how they were addressed through the years by many researchers. We also present a breakdown of the commonly used kernel functions, parameter selection routes, and case studies. Lastly, this review provides an outlook into the future of kernel-based process monitoring, which can hopefully instigate more advanced yet practical solutions in the process industries.

Identifying abnormal process operation with spatial-temporal data remains an important and challenging work in many practical situations. Although spatial-temporal data identification has been extensively studied in some domains, such as public health, geological condition, and environment pollution, the challenge associated with designing accurate and convenient recognition schemes is very rarely addressed in modern manufacturing processes. This paper proposes a general recognition framework for identifying abnormal process with spatial-temporal data by employing a convolutional neural network (CNN) model. Firstly, motivated by the pasting case study, the spatial-temporal data are transformed into process images for capturing spatial and temporal interrelationship. Then, the CNN recognition model is presented for identifying different types of these process images, leading to the identification of abnormal process with spatial-temporal data. The specific architecture parameters of CNN... [more]

This paper presents a simultaneous state variables and system and actuator fault estimation, based on an unknown input interval observer design for a discrete-time parametric uncertain Takagi−Sugeno system under actuator fault, with disturbances in the process and measurement noise. The observer design is synthesized by considering unknown but bounded process disturbances, output noise, as well as bounded parametric uncertainties. By taking into account these considerations, the upper and lower bounds of the considered faults are estimated. The gain of the unknown input interval observer is computed through a linear matrix inequalities (LMIs) approach using the robust H ∞ criteria in order to ensure attenuation of process disturbances and output noise. The interval observer scheme is experimentally evaluated by estimating the upper and lower bounds of a torque load perturbation, a friction parameter and a fault in the input voltage of a permanent magnet DC motor.

In order to maximize the operating flexibility and optimize the system performance of a battery energy storage system (BESS), developing a reliable real-time estimation method for the state of charge (SOC) of a BESS is one of the crucial tasks. In practice, the accuracy of real-time SOC detection can be interfered with by various factors, such as battery’s intrinsic nonlinearities, working current, temperature, and aging level, etc. Considering the feasibility in practical applications, this paper proposes a hybrid real-time SOC estimation scheme for BESSs based on an adaptive network-based fuzzy inference system (ANFIS) and Coulomb counting method, where a commercially available lead-acid battery-based BESS is used as the research target. The ANFIS allows effective learning of the nonlinear characteristics in charging and discharging processes of a battery. In addition, the Coulomb counting method with an efficiency adjusting mechanism is simultaneously used in the proposed scheme to... [more]

AC (alternating current) system backup protection setting calculation is an important basis for ensuring the safe operation of power grids. With the increasing integration of modular multilevel converter based high voltage direct current (MMC-HVDC) into power grids, it has been a big challenge for the AC system backup protection setting calculation, as the MMC-HVDC lacks the fault self-clearance capability under pole-to-pole faults. This paper focused on the pole-to-pole faults analysis for the AC system backup protection setting calculation. The principles of pole-to-pole faults analysis were discussed first according to the standard of the AC system protection setting calculation. Then, the influence of fault resistance on the fault process was investigated. A simplified analytic approach of pole-to-pole faults in MMC-HVDC for the AC system backup protection setting calculation was proposed. In the proposed approach, the derived expressions of fundamental frequency current are applic... [more]

Accurate detection and effective control strategy of commutation failure (CF) of high voltage direct current (HVDC) are of great significance for keeping the safe and stable operations of the hybrid power grid. At first, a novel detection method for consecutive CF is proposed. Concretely, the 2nd and higher orders’ derivative values of direct current are summarized as the core to judge CF by analyzing the physical characteristics of the direct current waveform of the converter station in CF. Then, the Daubechies wavelet coefficient that can represent the 2nd and higher order derivative values of direct current is derived. Once the wavelet coefficients of the sampling points are detected to exceed the threshold, the occurrence of CF is confirmed. Furthermore, by instantly increasing advanced firing angle β in the inverter side, an additional emergency control strategy to prevent subsequent CF is proposed. Eventually, with simulations of the benchmark model, the effectiveness and superio... [more]

Industrial archived process data represent a convenient source of information for data-driven models, such as artificial neural network (ANN), that can be used for safety and efficiency improvement like early or even predictive fault detection and diagnosis (FDD). Nonetheless, most of the data used for model generation are representative of the process nominal states and therefore are not enough for classification problems intended to determine abnormal process conditions. This work proposes the use of techniques to augment the original real data standards, dismissing the need for experiments that could jeopardize process safety. It uses the Monte Carlo technique to artificially increase the number of model inputs coupled to the nearest neighbor search (NNS) by geometric distances to consistently classify the generated patterns in normal or faulty statuses. Finally, a radial basis function neural network is trained with the augmented data. The methodology was validated by a study case... [more]

The application of empirical Bayes for lot inspection in sequential sampling plans is usually conducted to estimate the proportion of defective items in the lot rather than for hypothesis testing of the variables’ process mean. In this paper, we propose the use of empirical Bayes in a sequential sampling plan variables’ process mean testing under a squared error loss function and precautionary loss function, for which the prediction is performed to estimate a sequence of the mean when the data are normally distributed in the case of a known mean and unknown variance. The proposed plans are compared with the sequential sampling plan. The proposed techniques yielded smaller average sample number (ASN) and provided higher probability of acceptance (Pa) than the sequential sampling plan.

Diesel particulate filter is one of the most effective after-treatment techniques to reduce Particulate Matters (PM) emissions from a diesel engine, but the blocking Diesel Particulate Filter (DPF) will seriously affect the engine performance, so it is necessary to study the fault diagnosis of blocking DPF. In this paper, a simulation model of an R425DOHC diesel engine with wall-flow ceramic DPF was established, and then the model was verified with experimental data. On this basis, the fault diagnosis of the blocking DPF was studied by using spectral analysis on instantaneous exhaust pressure. The results showed that both the pre-DPF mean exhaust pressure and the characteristic frequency amplitude of instantaneous exhaust pressure can be used as characteristic parameters of monitoring the blockage fault of DPF, but it is difficult to monitor DPF blockage directly by instantaneous exhaust pressure. In terms of sensitivity, the characteristic frequency amplitude of instantaneous exhaust... [more]

In the wireless sensor and actuator networks (WSANs) of industrial field monitoring, maintaining network connectivity with coverage perception plays a decisive role in many industrial process scenarios. The mobile actuator node is responsible for collecting data from the sensing nodes and performing diverse specific collaborative operation tasks. However, the failure of the nodes usually causes coverage vulnerability and partition of the network. Urgent and time-sensitive applications expect a minimum coverage loss to complete an instant connectivity restoration. This paper presents a hybrid coverage perception-based connectivity restoration algorithm, which is designed to restore network connectivity with minimal coverage area loss. The algorithm uses a backup node, which is selected nearby the critical node, to ensure a timely restoration when the critical node encounters failure. In the process of backup node migration, the optimal destination will be reselected to maintain the best... [more]

This work deals with the Leak Detection and Isolation (LDI) problem in water pipelines based on some heuristic method and assuming only flow rate and pressure head measurements at both ends of the duct. By considering the single leak case at an interior node of the pipeline, it has been shown that observability is indeed satisfied in this case, which allows designing an observer for the unmeasurable state variables, i.e., the pressure head at leak position. Relying on the fact that the origin of the observation error is exponentially stable if all parameters (including the leak coefficients) are known and uniformly ultimately bounded otherwise, the authors propose a bank of observers as follows: taking into account that the physical pipeline parameters are well-known, and there is only uncertainty about leak coefficients (position and magnitude), a pair of such coefficients is taken from a search space and is assigned to an observer. Then, a Genetic Algorithm (GA) is exploited to minim... [more]

The carburizing and coking of ethylene cracking furnace tubes are the important factors that affect the energy efficiency of ethylene production. To realize the diagnosis and prediction of the different coking degrees of cracking furnace tubes, and then take corresponding treatment measures, is of great significance for improving ethylene production and prolonging the service life of the furnace tube. Therefore, a fusion diagnosis and prediction method based on artificial bee colony (ABC) and adaptive neural fuzzy inference system (ANFIS) is proposed, which also introduces a coking-time factor (CTF). The actual data verification shows that the method not only improves the training efficiency and diagnosis accuracy of the coking diagnosis and inference system of the cracking furnace tube, but also realizes the prediction of the development trend of the coking degree of the furnace tube.

This paper addresses the preparation of electrically conductive electrospun mats on a base of styrene-isoprene-styrene copolymer (SIS) and multiwall carbon nanotubes (CNTs) and their application as active sensing elements for the detection of vegetable oil impurities dispersed within water. The most uniform mats without beads were prepared using tetrahydrofuran (THF)/dimethyl formamide (DMF) 80:20 (v/v) as the solvent and 13 wt.% of SIS. The CNT content was 10 wt.%, which had the most pronounced changes in electrical resistivity upon sorption of the oil component. The sensors were prepared by deposition of the SIS/CNT layer onto gold electrodes through electrospinning and applied for sensing of oil dispersed in water for 50, 100, and 1000 ppm.

Relay protection equipment is important to ensure the safe and stable operation of power systems. The risks should be evaluated, which are caused by the failure of relay protection. At present, the fault data and the fault status monitoring information are used to evaluate the failure risks of relay protection. However, there is a lack of attention to the information value of monitoring information in the normal operation condition. In order to comprehensively improve monitoring information accuracy and reduce, a generalized proportional hazard model (GPHM) is established to fully exploit the whole monitoring condition information during the whole operation process, not just the monitoring fault condition data, with the maximum likelihood estimation (MLE) used to estimate the parameters of the GPHM. For solving the nonlinear equation in the process of parameter estimations, the adaptive homotopy algorithm is adopted, which could ensure the reversibility of the Jacobi matrix. Three test... [more]

Aiming at addressing the problem that the faults in axial piston pumps are complex and difficult to effectively diagnose, an axial piston pump fault diagnosis method that is based on the combination of Mel-frequency cepstrum coefficients (MFCC) and the extreme learning machine (ELM) is proposed. Firstly, a sound sensor is used to realize contactless sound signal acquisition of the axial piston pump. The wavelet packet default threshold denoises the original acquired sound signals. Afterwards, windowing and framing are added to the de-noised sound signals. The MFCC voiceprint characteristics of the processed sound signals are extracted. The voiceprint characteristics are divided into a training sample set and test sample set. ELM models with different numbers of neurons in the hidden layers are established for training and testing. The relationship between the number of neurons in the hidden layer and the recognition accuracy rate is obtained. The ELM model with the optimal number of hi... [more]

There are many interference components in Fourier amplitude spectrum of a contaminated fault signal, and thus the segment obtained based on the spectrum can lead to serious over-decomposition of empirical wavelet transform (EWT). Aiming to resolve the above problems, a novel method named improved empirical wavelet transform (IEWT) is proposed. Because the power spectrum is less sensitive to the contaminated interference and manifests the presence of fault feature information, IEWT replaces the Fourier amplitude spectrum of EWT with power spectrum in segment acquirement, and threshold processing is also introduced to eliminate the bad influence on the acquirement, and thus the best decomposition result of IEWT can be obtained based on feature energy ratio (FER). The loose slipper fault signal of hydraulic pump is tested and verified. The result demonstrates that the proposed method is superior and can extract the fault feature information accurately.

Gas turbine converts chemical energy into mechanical energy and provide energy for aircraft, ships, etc. The performance diagnosis of rotating components of gas turbine are essential in terms of the high failure rate of these parts. A problem that the sudden changing of operation state of turbines may lead to the misdiagnosis due to the defect of gas turbine’s model. This paper constructs the strong tracking filter based on the unscented Kalman filter to achieve accurate estimation of gas turbine’s measured parameters when the state changes suddenly. In the strong tracking filter, a parameter optimization method based on the residual similarity of measured parameters is proposed. Next, adopt the measured parameters filtered by the strong tracking filter to construct the health parameters estimation algorithm based on the particle filter. The particle weight is optimized by the mean adjustment method. Performance diagnosis is realized by checking the changes of health parameters output... [more]