Gustavo Almeida

• PhD
• Researcher at Lappeenranta-Lahti University of Technology (LUT)

Boiler bank fouling reduces heat transfer efficiency in kraft recovery boilers. Here, we model the relationships between boiler parameters and boiler bank pressure drop, an indicator of fouling, based on recovery boiler operating data. We compared two models: an autoregressive integrated exogenous (ARIX) model and a feedforward neural network. The...

Detection and classification of anomalies in industrial applications has long been a focus of interest in the research community. The integration of computational and physical systems has increased the complexity of interactions between processes, leading to vulnerabilities in both the physical and cyber layers. This work presents a model structure...

This paper reviews the key information and communication technologies that are necessary to build an effective industrial energy management system considering the intermittence of renewable sources like wind and solar †. In particular, we first introduce the concept of software-defined energy networks in the context of industrial cyber-physical sys...

The two parts papers, Big Data Analytics in Chemical Engineering and Visual Analytics–Seeking the Unknown describe the importance of these analytics for chemical engineering data-driven knowledge discovery. This is one of the first papers, in 2017, that address Analytics for Chemical Engineering.

Fouling is still a major challenge for the operation of kraft recovery boilers. This problem is caused by accumulation of ash deposits on the surfaces of heat exchangers in the upper part of the boiler over time. The first consequence is the reduction of steam production due to loss of heat transfer and, finally, the shutdown of the boiler due to c...

Recent studies show that decision making in Business Process Management (BPM) and incorporating sustainability in business is vital for service innovation within a company. Likewise, it is also possible to save time and money in an automated, intelligent and sustainable way. Robotic Process Automation (RPA) is one solution that can help businesses...

RESUMO O presente trabalho refere-se ao estudo de análise de cenários operacionais em uma planta de evaporação de uma fábrica de celulose kraft no Brasil. Este tipo de estudo consiste em obter o comportamento do processo frente a perturbações em variáveis e parâmetros. Primeiro, com base em dados de projeto e operação, construiu-se um modelo comput...

Performance metrics are usually evaluated only after the neural network learning process using an error cost function. This procedure can result in suboptimal model selection, particularly for imbalanced classification problems. This work proposes the direct use of these metrics as cost functions, which are often derived from the confusion matrix....

Model degradation is still a challenge in real-time applications such as chatbot systems. This work refers to a webchat service of a Brazilian energy utility company, whose central part is composed of a supervised model and a question-and-answer list. User queries not met by it go to an NLP-based clustering model, responsible for identifying unknow...

This work presents a new decision-making strategy for multi-objective learning problem of artificial neural networks (ANN). The proposed decision-maker searches for the solution that minimizes a margin-based validation error amongst Pareto set solutions. The proposal is based on a geometric approximation to find the large margin (distance) of separ...

Data acquisition in process industries usually takes place at each sampling. The disadvantage is that a considerable amount of data without new information about the state of the process is continuously transmitted and processed. This negatively affects the communication system and computational power, which is more critical nowadays given the numb...

The goal of this study is to investigate stochastic optimal solutions for a boiler process in a pulp mill. The objective function is a steam generation while two pollutant emissions should be complied with their regulations. Support Vector Regression (SVR) is employed to build empirical models for representing a boiler process and air temperatures...

As indústrias em geral buscam cada vez mais, além do aumento de produção com qualidade, redução de custos e operações mais seguras, uma produção mais limpa. De outro lado, o uso de modelos obtidos diretamente a partir de dados históricos sobre as operações se fortaleceu a partir da geração massiva de dados pelos processos industriais. Outro fator q...

Customer’s satisfaction is crucial for companies worldwide. An integrated strategy composes omnichannel communication systems, in which chabot is widely used. This system is supervised, and the key point is that the required training data are originally unlabelled. Labelling data manually is unfeasible mainly nowadays due to the considerable volume...

To increase the understanding of hydrothermal carbonization (HTC) of lignocellulosic biomass residues, four feedstocks: giant bamboo, coffee wood, eucalyptus, and coffee parchment, were studied. The effect of operating conditions on the products in terms of yield, composition and energy densification were quantified. Each feedstock was treated for...

Coffee production in Brazil creates significant amounts of residues. The goals of this study are to evaluate the characteristics of these residues (parchment, husk, pulp, spent grounds, silverskin and defective beans); to discuss their potential for conversion to improved biofuels via thermochemical methods; and to develop mass and energy balances...

O monitoramento da qualidade da água é cada vez mais essencial para a preservação dos recursos hídricos. Os dados coletados das estações de monitoramento são uma fonte rica de informações sobre a qualidade da água, importantes para a proteção do recurso hídrico, a manutenção da capacidade ambiental e o controle da poluição. Incrementar as análises...

Steel tubes produced in steelmaking plants are generally subjected to severe in-service conditions. Hence, quality control plays a key role in this process. The bottleneck is that this information is made available only after tube production from laboratory analysis. Given process complexity and current data availability, this work employs a series...

Robotic Process Automation (RPA) refers to process automation applications of traditional Information Technologies based on robot software with the ability to capture and interpret the specific processes of organizations. Studies show that RPAs are able to reduce resources and optimize processes effectively in relation to customers. Some of these c...

Os mapas de curvas residuais são muito úteis para interpretar o desempenho da destilação azeotrópica de sistemas homogêneos. Esses mapas são obtidos através de um sistema de equações diferenciais, comumente resolvido em duas etapas, a saber, um equilíbrio líquido-vapor e um sistema de equações diferenciais. Uma solução alternativa trata ambas as et...

A metodologia mais comum de obter os mapas de curvas residuais é através de um sistema de equações diferenciais, comumente resolvido em duas etapas, a saber, um equilíbrio líquido-vapor e um sistema de equações diferenciais. O objetivo deste trabalho é apresentar uma nova metodologia de elaboração dos mapas de curvas residuais. Nesta metodologia o...

O monitoramento contínuo de variáveis-chave em processos industriais é crucial para garantir produção com qualidade e segurança operacional. Ainda, a construção de modelos através de uma descrição puramente matemática não é geralmente satisfatória, dado a complexidade das operações. Assim, a disponibilidade de dados e um conjunto de técnicas basead...

A critical factor in steelworks concerns setting the steel release temperature from the ladle furnace. The challenge resides in estimating in advance the reduction the steel temperature will undergo during its non-processing time until the subsequent casting process. A poor estimation results in productivity and yield losses in casting and unnecess...

A bottleneck of laboratory analysis in process industries including steelmaking plants is the low sampling rate. Inference models using only variables measured online have then been used to made such information available in advance. This study develops predictive models for key mechanical properties of seamless steel tubes, by strength, ultimate t...

This paper introduces a general approach to design a tailored solution to detect rare events in different industrial applications based on Internet of Things (IoT) networks and machine learning algorithms. We propose a general framework based on three layers (physical, data and decision) that defines the possible designing options so that the rare...

Our Aim was to find the stochastic optimal solution for a boiler process in order to maximize steam generation while complying with pollutant emissions regulations. For a simulation base-model, support vector regression (SVR) is employed to present a general discrete-time dynamical system of a boiler, and a stochastic optimization was performed to...

O ensino de Engenharia tem sido bastante repensado, discutido e modificado, tendo como um dos principais focos o desenvolvimento de aprendizagens ativas, cooperativas e baseadas em desenvolvimento de projetos. Este trabalho propõe discutir metodologias alternativas de aprendizado, fundamentadas na implementação pedagógica das tecnologias digitais n...

This paper presents a novel approach to deal with the imbalanced data set problem in neural networks by incorporating prior probabilities into a cost-sensitive cross-entropy error function. Several classical benchmarks were tested for performance evaluation using different metrics, namely G-Mean, area under the ROC curve (AUC), adjusted G-Mean, Acc...

Hydrothermal carbonization (HTC) of biomass is a thermochemical conversion technology that improves fuel characteristics of biomass without preliminary drying. The treatment process involves hydrolysis, dehydration, decarboxylation, polymerization, and aromatization mechanisms when raw feedstock is under a suspension with pressurized water at tempe...

Resumo A carbonização hidrotérmica (HTC) é um método promissório de tratamento termoquímico que melhora as características combustíveis da biomassa sem secagem preliminar. Durante este processo a matéria prima é submetida ao aquecimento a temperaturas entre 180-250 ºC com adição de uma certa quantidade de líquido (geralmente água) em um sistema fec...

Governments and society are concerned about the climate change that has been increasing over the years. It is known that the main cause is the emission of greenhouse gases, due to the burning of non-renewable fuels that are mainly used in the generation of energy both in companies and in the transport industry. At this moment a great opportunity op...

Carbonization or slow pyrolysis is a promising method of thermochemical treatment, which allows biomass conversion into charcoal with enhanced after-treatment characteristics in heating value and energy density. For countries like Brazil, charcoal has a critical application on industrial scale, since it is the world’s largest coal and sole pig iron...

This paper discusses kappa number prediction models using Single Exponential Smoothing, Multiple Linear Regression Analysis, the Time Series Method of Box-Jenkins (ARIMA) and Artificial Neural Networks. Applying a database of an industrial eucalyptus Kraft pulp continuous digester, these four different methods were evaluated according to different...

Membrane bioreactor (MBR) has been widely employed for industrial effluent treatment, as its higher efficiency in removing pollutants makes effluent reuse more feasible. However, membrane fouling remains as a limiting factor for its greater diffusion. This work performed a sensitivity analysis study to investigate the effects of analytical and oper...

Multiple Instance Learning (MIL) is a recent paradigm of learning, which is based on the assignment of a single label to a set of instances called bag. A bag is positive if it contains at least one positive instance, and negative otherwise. This work proposes a new algorithm based on likelihood computation by means of Kernel Density Estimation (KDE...

Instantaneous measurements of process variables are usually not representative of the process effects as a whole when defining the condition of an output sample mainly in case of laboratory analysis. Moreover, process data have considerable dispersion. This leads to uncertainty in input–output time alignment and in variable relationship. This work...

The development of automatic and reliable fault detection systems is still a challenge nowadays. Chemical processes are complex by nature by presenting non linear dynamics, multiple modes with constant interchanges, and spatial and serial correlations, to mention a few. To address these issues, this work explores the hidden Markov model (HMM) techn...

This paper presents the use of Support Vector Machines (SVM) methodology for fault detection and diagnosis. Two approaches are addressed: the SVM for classification (Support Vector Classification – SVC) and SVM for regression (Support Vector Regression – SVR). A comparison was made between the two techniques through the study of a reactor of cyclop...

Monitoring abnormal situations in chemical industries is a worldwide challenge. The occurrence of this kind of event is common, however, its detection is generally after its development into a faulty condition. The earlier it is detected, the greater the chance to guarantee safe, economical and clean operations. This study aims to develop a reliabl...

Monitoring abnormal situations in continuous chemical process industries is a worldwide challenge. The occurrence of this kind of event is common, however its detection is generally after its development into a faulty condition. The earlier it is detected, the greater the chance to guarantee safe, economical and clean operations. This study develop...

The development of automatic and reliable monitoring systems is an open issue in continuous industrial chemical processes. The challenges lay on simultaneously managing multiple normal modes of operation as well as the transitions among them with reasonable false alarm rates, and in reaching early fault detection. This work explores and attests the...

Data visualization techniques are powerful in the handling and analysis of multivariate systems. One such technique known as parallel coordinates was used to support the diagnosis of an event, detected by a neural network-based monitoring system, in a boiler at a Brazilian Kraft pulp mill. Its attractiveness is the possibility of the visualization...

The development of robust and reliable automatic fault detection systems is still a challenge nowadays. One problem is the spatial similarity regarding the behaviour of variables before and after the onset fault occurrences. This work proposes a sequential data modelling paradigm for addressing the fault detection problem, which takes the order of...

The extraction of information from tabular data is not a natural task for human beings, which is worse when dealing with high dimensional systems. On the other hand, graphical representations make the understanding easier by exploring the human capacity of processing visual information. Such representations can be used for many purposes, e.g., comp...

A reliable monitoring of abnormal situations is still a challenge in industrial chemical processes. The earlier its detection and diagnosis, the greater the possibility of mitigating losses. An alternative approach to carry out these tasks is to make use of signal processing tools. In this direction, the hidden Markov model (HMM) method is used to...

The visualization of relevant information from numerical data is not a natural task for human beings, mainly in case of multivariate systems. In compensation, graphical representations make the understanding easier since it explores the human capacity of processing visual information. Based on that, this study constructs a cause-effect map relating...

Early fault detection and diagnosis in chemical process monitoring represents a challenge to be overcome. Another one concerns the spatial overlapping problem among distinct fault classes, once some events may only be distinguished from the others by taking into account its order of occurrence. The hidden Markov model (HMM) technique is capable of...

One of the main problems brought by gasoline combustion in vehicles is the deposition of gum in the fuel system. Given the sort of refining used in Brazil, the content of naphtha resulting from catalytic cracking in gasoline coming from refineries is quite significant. The chains resulting from catalytic or thermal cracking processes have a strong...

The objective of this work is to analyze two recovery boiler performance variables, the steam generated and the reduction efficiency. These variables were related to black liquor flow rate, dry solids content, drum pressure and others ones. It was done by using the following techniques: data visualization, variable selection and neural networks. Th...

The recovery boiler plays a decisive role for the economic and environmental viability in the Kraft process. The residual black liquor from the wood chemical pulping is concentrated and burned in this equipment. The steam generated in the boiler is then used in heat transfer operations and as electrical energy by the mill. This study aims at predic...

The goal of this study is the neural network modeling of two performance variables of Kraft recovery boilers: production of superheated steam and ratio of the generated steam flow rate to the amount of dry solids feed to the boiler. Several physical and chemical phenomena occur simultaneously in these equipments, which play a fundamental role in th...

Early fault detection and diagnosis in the chemical process monitoring area is still a challenge. Other problem con-cerns the spatial overlapping among distinct fault classes, and hence some events may only be distinguished from the others by taking into account its order of occurrence. In this context, this work employs a signal processing tool so...

The detection of abnormal events in advance is still a challenge in chemical industries. The earlier it is done, the greater the possibility of at least mitigating losses. This study investigates the performance of a signal processing tool so-called hidden Markov model (HMM) in accomplishing detection tasks. The case study is based on a chemical re...

PI and PID controller design (tuning) employ in general classical techniques, such as Ziegler-Nichols and Cohen-Coon. One disadvantage in this case is the need for knowing the process dynamics. An alternative approach is the use of evolu-tionary optimization methods. In this work, one of them called Differential Evolution is used for tuning the PI...

The identification of abnormal events in advance is yet a challenge in chemical industries. The earlier it is, the greater the possibility of at least mitigating losses. Computer-based systems play an important role on it in order to have success. This study aims at analyzing the performance of a processing tool, so-called hidden Markov model (HMM)...