Search

This article is concerned with the study of reliability assessment of the urea plant. A non-linear autoregressive external input (NARX) neural network was developed to predict the reliability using the failure frequency and all possible combinations of the equipment in the plant. The model was fed using field data; ten hidden layers and three subsets for the training were used: 50%, 70% and 85%. The NARX was capable of simulating the system and demonstrates a good performance comparable with the traditional reliability method and real data.

During the last few years we have been working on a modernization plan for the Telescopio Nazionale Galileo (TNG) Control System1,2. On October 2019 we had the opportunity to execute the first step of this process. The telescope was going to be stopped for one month due to M1 mirror being aluminized, so we could change the azimuth control system, that had been thoroughly tested during the summer, with no additional observational time loss. In this paper we present the new control system based on the CompactRIO platform from National Instruments, the switching process between the old and the new control systems, and a performance comparison between them.

The article presents details of a reliability analysis for a Direct Methanol Fuel Cell system, focussing on both quantitative and qualitative methods. The system was analysed on an overall basis. The qualitative analyses presented include fault tree analyses, where four main causes of total system failure were outlined with their initiating failure modes. These included the Failure Mode, Event and Criticality analysis (FMECA) which focused on failures, causes, severity, frequency and the reaction or response to each issue. The results which included raw data for all combinations of failure were analysed in different ways and compared to the published study. It was found that for the system studied in this project, the most reliable configuration had a reliability figure of 0.98243 while the least reliable was 0.78358 compared to the published study which had a value of 0.9041. The main difference was that this study included redundancies (parallel pumps) and considered additional equipment (boost controller) compared to that of the published data. When altering the given system to match that of the published paper, the percentage error in reliability (2.1%) was very low, indicating that the differences were from configuration and analysis rather than the failure rate values.

GIANO-B is the high resolution near-infrared (NIR) spectrograph of the Telescopio Nazionale Galileo (TNG), which started its regular operations in October 2017. Here we present GIANO-B Online Data Reduction Software (DRS) operating at the Telescope. GIANO-B Online DRS is a complete end-to-end solution for the spectrograph real-time data handling. The Online DRS provides management, processing and archival of GIANO-B scientific and calibration data. Once the instrument control software acquires the exposure ramp segments from the detector, the DRS ensures the complete data flow until the final data products are ingested into the science archive. A part of the Online DRS is GOFIO software, which performs the reduction process from ramp-processed 2D spectra to extracted and calibrated 1D spectra. A User Interface (UI) developed as a part of the Online DRS provides basic information on the final reduced data, thus allowing the observer to take decisions in real-time during the night and adjust the observational strategy as needed.

The NIR echelle spectrograph GIANO-B at the Telescopio Nazionale Galileo is equipped with a fully automated online DRS: part of this pipeline is the GOFIO reduction software, that processes all the observed data, from the calibrations to the nodding or stare images. GOFIO reduction process includes bad pixel and cosmic removal, flat-field and blaze correction, optimal extraction, wavelength calibration, nodding or stare group processing. An offline version of GOFIO will allow the users to adapt the reduction to their needs, and to compute the radial velocity using telluric lines as a reference system. GIANO-B may be used simultaneously with HARPS-N in the GIARPS observing mode to obtain high-resolution spectra in a wide wavelength range (383-2450 nm) with a single acquisition. In this framework, GOFIO, as part of the online DRS, provides fast and reliable data reduction during the night, in order to compare the infrared and visible observations on the fly.

In this study, the use of fuzzy logic for signal validation is introduced. This tool uses the sensor readings at the current time step to predict system state at the next time step using observers. The deviation between the sensor reading and the model prediction rises, leading to fault detection and isolation. The fault is accommodated by replacing the reading of the faulty sensor by the model prediction. The results presented in this article demonstrate that a simple scheme can carry out sensor fault detection, isolation and accommodation using a fuzzy rule-based system based on experience of an operator into heuristics. Although the method lacks the mathematical basis that other analytical methods provide, this method is simple, effective and useful. PRACTICAL APPLICATIONS The coloring process has to be controlled and monitored carefully because if the color of the MALTA is not adequate, then the MALTA has to be recycled to another unit (dilution unit) where the colorant is eliminated. This recycling process uses delay around 3 working days, and the economic impact is high. The validation of signal and reduction of fault sensor contribute to the reduction of the recycling process. In this study, the incorporation of fuzzy logic in the treatment of sensor signal represents a significant factor in the improvement of MALTA production units.

This paper outlines the application of soft control strategies in the thermal treatment in the dairy industry. The milk pasteurizer is the most important process for guarantee the quality of the milk. The article describes with details the application of fuzzy logic and neural network in a plate heat exchanger (PHE). Recently the application of artificial intelligent has increased sharply, especially, fuzzy logic due to its easy implementation and low mathematical level. This article discusses technical features of three advanced control technologies: Fuzzy control, Model Reference Adaptive Fuzzy Control and neural networks, their implementation in food processes including a comparative study. Industrial relevance Existing control does not permit to introduce the experience of the operator for an adequate control of the milk pasteurization in industrial applications. Consequently, it was the aim of this very relevant to test the efficiency of soft techniques (fuzzy logic and neural networks) to control the temperature during milk pasteurization. The uses of soft techniques do not require complex mathematical models and complex tuning control methods. Since the controller can be realized by simple arithmetic operations, it can be improved continuously.

In this paper we describe an approach to Sensor Fault Detection, Isolation and Accommodation (SFDIA) that is mainly based on fuzzy processing of deviation from the model predictions. The predictive model is supported using a Decision Making Unit (DMU) and cased-based Reasoning principles (CRB) where pass cases are saved and filed in the Library for confirmed solution and repaired cases. This tool uses the sensor readings at the current time step to predict system state at the next time step using observers. Although the method lacks the mathematical basis that other analytical methods provide; this method is simple effective and useful.

The knowledge-based pattern recognition (KE) approach provides a basis for classification of state of fluid systems. The new method determines bubble size distribution at high pressure that is very important for the brewery industry and other alcoholic beverages (champagne). Conforming to the principle of “decreasing precision with increasing intelligence”, the KE approach has been applied to the determination of bubble size distribution. An important feature of these architectures is that they do not require global mathematical modeling of the system.

The paper outlines the successful empirical design and validation of a space–time PI controller based on study of the controlled variable output as function of time and space. The developed control was implemented on two heat exchanger systems (falling film evaporator and milk pasteurizer). The strategy required adding a new term over the classical PI controller, such that a new parameter should be tuned. Measurements made on commercial installations have confirmed the validity of the new controller.

In this paper is introduced a general formalism for the application of advanced techniques in adaptive control as Lyapunov-based MRAFC (Model Reference Adaptive Fuzzy Control) and the hybrid system ANFIS (Adaptive-Network-based Fuzzy Inference System) for the control of the freezing fruits (strawberries). It provides a description of fuzzification, inference rules, adaptive laws and defuzzification from the fuzzy set domain. In this work, we proposed the adaptive control as an alternative to the classical control methods and illustrated as the quality of the controller can be destroyed for the linguistic interpretability of the rules.

The objective of this paper is to integrate fuzzy logic into the fermentation process in a brewery. The brewery involved is a local commercial brewery, Beamish and Crawford Brewery plc in Cork, Ireland. Our approach consists of developing a control system for a fermentation process using fuzzy logic in two stages. In the first stage the software package fuzzyTech from Inform provided the fuzzy logic controller, and in the second stage InTouch from Wonderware provided the user front-end. The results of this new controller and the fault detection system show a clear alternative over the conventional controller (PID). Optimisation of the fermentation vessel is run simultaneously with this project. The main idea is based on examining the various input parameters that influence the fermentation process. These parameters include temperature, pressure, fermentation duration and yeast count. Next the quality or output parameters are identified. These include alcohol content, present gravity, pH, colours and bitters. A model based on fuzzy logic is developed to establish the inter-relationships between these factors and how they affect the output quality factors. The ultimate objective is to integrate this fuzzy logic model into the fermentation process control system.

This short paper presents a design technique based on passivity for single input single output (SISO) systems through a process of feedback passivisation. The controller is implemented in the control of a tank non-isothermal for the production of pineapple syrup. The design of a model and the implementation to laboratory scale allowed the validation of the controller.

This study is addressed to improve the quality of the signal of the Adaptive-Network based Fuzzy Inference System (ANFIS) reducing the level of fluctuations in the output due to periodical disturbances. The fuzzy filter computes the disturbances as periodic signals with two components, one at high frequency and another at low frequency. It was incorporated in the layer 0 of the network reducing iteratively effectively the heavy noise.

This short paper outlines the computer simulation using real data of a decoupled control system for a desalination unit. The control strategy incorporated a perfect decoupled controller for the control of the fresh water flow and conductivity. The model was estimated using real data and empirical tools instead of mass balances. The success is demonstrated in the reduction of wide fluctuations in the variables of the process and decreasing of the sensibility to the changes of pressure and/or pH and allows predicting problems of quality of water and waste of energy in the future.

The main objective of this note is to describe a real-time fault diagnosis and control for a cylindroconical fermenter to laboratory scale as an alternative to the classic systems. Development of a good controller for a fed-batch reactor is not enough without a fault diagnosis system. We are working to expand this idea. The failure detection system is based on the parity space approach joined with a fuzzy controller with more robustness and of superior performance in MIMO systems compared to conventional strategies. A 2-week period is reported.

In an undergraduate computer aided engineering course, the goal is to provide an unequivocal agreement between theory and practice, using rigorous modelling. This paper includes a practical lesson on Dynamic Model/Simulation of an Evaporator for the students. The model consisted of a 5 m³ single effect natural circulation evaporator. The result showed that for a 10 kmol/sec feed flow and energy input of approximately 5706 kJ/sec it was possible to concentrate the feed from 0.25 mol to 0.295 mol fraction. The paper discusses the material with details adapted for stimulating interest in the application of TKSolver in the simulation of chemical processes and thereby facilitating learning.