Intelligent systems in process engineering: a review

SNAPS: Sensor Analytics Point Solutions for Detection and Decision Support Systems

In this review, we discuss the role of sensor analytics point solutions (SNAPS), a reduced complexity machine-assisted decision support tool. We summarize the approaches used for mobile phone-based chemical/biological sensors, including general hardware and software requirements for signal transduction and acquisition. We introduce SNAPS, part of a platform approach to converge sensor data and analytics. The platform is designed to consist of a portfolio of modular tools which may lend itself to dynamic composability by enabling context-specific selection of relevant units, resulting in case-based working modules. SNAPS is an element of this platform where data analytics, statistical characterization and algorithms may be delivered to the data either via embedded systems in devices, or sourced, in near real-time, from mist, fog or cloud computing resources. Convergence of the physical systems with the cyber components paves the path for SNAPS to progress to higher levels of artificial reasoning tools (ART) and emerge as data-informed decision support, as a service for general societal needs. Proof of concept examples of SNAPS are demonstrated both for quantitative data and qualitative data, each operated using a mobile device (smartphone or tablet) for data acquisition and analytics. We discuss the challenges and opportunities for SNAPS, centered around the value to users/stakeholders and the key performance indicators users may find helpful, for these types of machine-assisted tools.

Modeling and Simulation of Energy Systems: A Review

Energy is a key driver of the modern economy, therefore modeling and simulation of energy systems has received significant research attention. We review the major developments in this area and propose two ways to categorize the diverse contributions. The first categorization is according to the modeling approach, namely into computational, mathematical, and physical models. With this categorization, we highlight certain novel hybrid approaches that combine aspects of the different groups proposed. The second categorization is according to field namely Process Systems Engineering (PSE) and Energy Economics (EE). We use the following criteria to illustrate the differences: the nature of variables, theoretical underpinnings, level of technological aggregation, spatial and temporal scales, and model purposes. Traditionally, the Process Systems Engineering approach models the technological characteristics of the energy system endogenously. However, the energy system is situated in a broader economic context that includes several stakeholders both within the energy sector and in other economic sectors. Complex relationships and feedback effects exist between these stakeholders, which may have a significant impact on strategic, tactical, and operational decision-making. Leveraging the expertise built in the Energy Economics field on modeling these complexities may be valuable to process systems engineers. With this categorization, we present the interactions between the two fields, and make the case for combining the two approaches. We point out three application areas: (1) optimal design and operation of flexible processes using demand and price forecasts, (2) sustainability analysis and process design using hybrid methods, and (3) accounting for the feedback effects of breakthrough technologies. These three examples highlight the value of combining Process Systems Engineering and Energy Economics models to get a holistic picture of the energy system in a wider economic and policy context.

Optimization of biogas generation using anaerobic digestion models and computational intelligence approaches

Anaerobic digestion (AD) technology has become popular and is widely used due to its ability to produce renewable energy from wastes. The bioenergy produced in anaerobic digesters could be directly used as fuel, thereby reducing the release of biogas to the atmosphere. Due to the limited knowledge on the different process disturbances and microbial composition that are vital for the efficient operation of AD systems, models and control strategies with respect to external influences are needed without wasting time and resources. Different simple and complex mechanistic and data-driven modeling approaches have been developed to describe the processes taking place in the AD system. Microbial activities have been incorporated in some of these models to serve as a predictive tool in biological processes. The flexibility and power of computational intelligence of evolutionary algorithms (EAs) as direct search algorithms to solve multiobjective problems and generate Pareto-optimal solutions have also been exploited. Thus, this paper reviews state-of-the-art models based on the computational optimization methods for renewable and sustainable energy optimization. This paper discusses the different types of model approaches to enhance AD processes for bioenergy generation. The optimization and control strategies using EAs for advanced reactor performance and biogas production are highlighted. This information would be of interest to a dynamic group of researchers, including microbiologists and process engineers, thereby offering the latest research advances and importance of AD technology in the production of renewable energy.

Early warning of changing drinking water quality by trend analysis

Monitoring and control of water treatment plants play an essential role in ensuring high quality drinking water and avoiding health-related problems or economic losses. The most common quality variables, which can be used also for assessing the efficiency of the water treatment process, are turbidity and residual levels of coagulation and disinfection chemicals. In the present study, the trend indices are developed from scaled measurements to detect warning signs of changes in the quality variables of drinking water and some operating condition variables that strongly affect water quality. The scaling is based on monotonically increasing nonlinear functions, which are generated with generalized norms and moments. Triangular episodes are classified with the trend index and its derivative. Deviation indices are used to assess the severity of situations. The study shows the potential of the described trend analysis as a predictive monitoring tool, as it provides an advantage over the traditional manual inspection of variables by detecting changes in water quality and giving early warnings.

Modelling the process of Al(OH)3 crystallization from industrial sodium aluminate solutions using artificial neural networks

This paper presents an attempt to define the non-linear correlation dependence between the degree of decomposition of the aluminate solution, the average diameter of the crystallized gibbsite, the total Na2O content in the obtained alumina and the specific utilization level of the process on the one hand and important input parameters of the process on the other. As input parameters having an influence on the process, the concentration of Na2O (caustic), the caustic ratio and the crystallization ratio, the starting and final temperature of the process, the average diameter of the crystallization seed and the duration of the decomposition process were considered. As the result of measurements of these process parameters and the acquisition of the resulting output parameters of the process, a database with 500 data lines was obtained. To define the correlation dependence, with the aim of predicting the process parameters of the decomposition process of the sodium aluminate solution, the artificial neural network (ANN) methodology was applied.

Uma revisão sobre os processos de polimerização em suspensão

Inúmeros processos podem ser utilizados para produção de materiais poliméricos. Cada processo apresenta características peculiares, que permitem produzir resinas com as mais variadas propriedades, visando a diferentes aplicações do material polimérico final. Os processos de polimerização em suspensão são bastante empregados para produção de resinas poliméricas por apresentarem muitas vantagens, como a facilidade de separação, fácil remoção de calor e controle de temperatura e, principalmente, pelos baixos níveis de impureza e de aditivação no produto final. Por isso, processos de polimerização em suspensão são apropriados para obtenção de produtos para aplicações biotecnológicas e médicas, dentre outras. O objetivo principal deste trabalho é apresentar uma discussão dos aspectos fundamentais inerentes aos processos de polimerização em suspensão, visando a uma compreensão do efeito das principais variáveis de processo sobre o desempenho de polimerizações em suspensão. Mostra-se também que a espectroscopia do infravermelho próximo (NIRS) pode ser muito útil para permitir o monitoramento e o controle em tempo real de processos de polimerização em suspensão, para a obtenção de materiais poliméricos com características especiais.

Multiple objective evolutionary algorithm for temporal linguistic rule extraction

Autonomous temporal linguistic rule extraction is an application of growing interest for its relevance to both decision support systems and fuzzy controllers. In the presented work, rules are evaluated using three qualitative metrics based on their representation on the truth space diagram. Performance metrics are then treated as competing objectives and the multiple objective evolutionary algorithm is used to search for an optimal set of nondominant rules. Novel techniques for data pre-processing and rule set post-processing are designed that deal directly with the delays involved in dynamic systems. Data collected from a simulated hot and cold water mixer are used to validate the proposed procedure.

A knowledge-based approach to the deflocculation problem: integrating on-line, off-line, and heuristic information

A knowledge-based approach for the supervision of the deflocculation problem in activated sludge processes was considered and successfully applied to a full-scale plant. To do that, a methodology that integrates on-line, off-line and heuristic information has been proposed. This methodology consists of three steps: (i). development of a decision tree (which involves knowledge acquisition and representation); (ii). implementation into a rule-based system; and (iii). validation. The set of symptoms most useful in diagnosing the deflocculation problem has been identified, the different branches to diagnose pin-point floc and dispersed growth have been built (using generic and specific knowledge), and all this knowledge has been codified into an object-oriented shell. The results obtained in the application of this knowledge-based approach to the Granollers WWTP (which treats about 130000 inhabitants-equivalents) showed that the system was able to identify correctly the problem with reasonable accuracy. Our positive experience building this system suggests that this approach is a practical and valuable element to include in an intelligent supervisory system combining numerical and reasoning techniques.

Effect of fluid dispersion on cybernetic control of microbial growth on substitutable substrates

Many fermentation media contain two or more substrates, which a microorganism utilizes for similar purposes. Depending on the conditions prior to and during a fermentation, the substrates may be utilized in succession or simultaneously. Since it is difficult to portray this behavior through mechanistic models, a cybernetic method was proposed earlier. Here the microorganism chooses the mode of substrate utilization that maximizes its own survival, usually expressed by the growth rate. In a fully dispersed bioreactor, simultaneous utilization generates higher growth rates but leads to low biomass concentrations since this utilization pattern is preferred at low concentrations of the substrates. In this study it has been shown that by allowing less than complete dispersion in the broth it is possible to shift from sequential to simultaneous utilization at high concentrations, thereby enabling both high growth rates and large biomass concentrations. This strategy thus allows the natural incomplete dispersion in large bioreactors to be gainfully exploited.

Computer applications in bioprocessing

Biotechnologists have stayed at the forefront for practical applications for computing. As hardware and software for computing have evolved, the latest advances have found eager users in the area of bioprocessing. Accomplishments and their significance can be appreciated by tracing the history and the interplay between the computing tools and the problems that have been solved in bioprocessing.

Applications of neural networks to recovery of biological products

Artificial neural networks (ANN) are being applied to recovery of products from fermentation broths. Recovery methods for which mathematical models are complex or non-existent are particularly suitable for control and analysis by ANNs. Use and potential of artificial neural networks for product recovery applications are reviewed.

Fermentation seed quality analysis with self-organising neural networks

Industrial fermentation processes operate under well defined operating conditions to attempt to minimise production variability. Variability occurs for many reasons but a long held belief is that variation in the state of the seed is highly influential. In this paper a seed stage (a batch process) of an industrial antibiotic fermentation is considered and the performance of the main production fermentations is correlated with the quality of the seed using an unsupervised Kohonen self-organising feature map (SOM). It is shown that using only seed information poor performance in the final stage fermentations can be predicted. Data from industrial penicillin G fermenters is used to demonstrate the procedure. Copyright 1999 John Wiley & Sons, Inc.