Search

This paper presents a new architecture for the Adaptive Neuro-Fuzzy Inference System (ANFIS) algorithm targeting FPGA implementation. This new architecture offers higher efficiency and scalability in comparison to the existing methods. The proposed architecture is modeled and simulated using VHDL and is targeted at a Xilinx FPGA. Existing implementation architectures are also modeled and comparisons are drawn between them in terms of both performance and logic utilization. The results show that the new architecture offers a reduction in calculation cycles of around 50% in comparison to the architecture from which it's derived. This increase in calculation speed comes with only a modest increase in logic utilization, specifically a 2.5% increase in look-up table (LUT) usage and a 1.5% increase in flip-flop usage. The new architecture also eliminates scalability issues which can arise in the existing architectures when extra input members are required.

A study on type and tolerance of biofilm bacteria from two seawater reverse osmosis (SWRO) plants sourced from same lagoon but with different types of membrane and pretreatment regimens was carried out. One system uses hollow fine fiber seawater RO membranes of cellulose triacetate with membranes arranged in one element per vessel and the other system uses spiral wound seawater RO membranes of polyamide (PA/thin film composite) with membranes arranged in two elements per vessel. The first system (system-1) uses chlorine with acidification, coagulation, and media filtration followed by micron cartridge filtration. The second system (system-2) uses only acidification and micron cartridge filtration without chlorination. Three different bacterial isolates were found predominantly in membranes belonging to system-1 while nine bacteria species were isolated from membranes in system-2, two species from the feed side and seven species from the brine side elements, respectively. All bacterial isolates we...

The hardware implementation of an Artificial Neural Network (ANN) using field-programmable gate arrays (FPGA) is a research field that has attracted much interest and attention. With the developments made, the programmer is now forced to face various challenges, such as the need to master various complex hardware-software development platforms, hardware description languages and advanced ANN knowledge. Moreover, such an implementation is very time consuming. To address these challenges, the paper presents a novel neural design methodology using a holistic modelling approach. Based on the end user programming concept, the presented solution empowers end users by means of abstracting the low-level hardware functionalities, streamlining the FPGA design process and supporting rapid ANN prototyping. A case study of an ANN as a pattern recognition module of an artificial olfaction system trained to identify four coffee brands is presented. The recognition rate versus training data features and data representation was analysed extensively.

Penelitian ini bertujuan untuk mendeskripsikan bagaimana kemampuan mendongeng dengan menggunakan media wayang kertas. Dengan dua rumusan masalah sebagai berikut yaitu bagaimana efektivitas dan peningkatan media wayang kertas terhadap kemampuan mendongeng mahasiswa dharmasiswa. Penelitian ini menggunakan jenis pendekatan kuantitatif. Karena penelitian ini bertujuan untuk memaparkan hasil data dengan menggunakan angka-angka sebagai representasi dari efektivitas media wayang dan Sasaran dalam penelitian ini adalah seluruh mahasiswa dharmasiswa Universitas Muhammadiyah Surabaya Tahun Ajaran 2017/2018 di kelas BIPA. Teknik pengumpulan data dengan cara tes berbicara mendongeng dengan menggunakan media wayang secara berkelompok dengan penilaian secara individu, setelah memperoleh data, data dianalisis menggunakan statistik deskriptif. Simpulan hasil penelitian menunjukan efektivitas media memperoleh nilai rata-rata sebesar 67,46 dengan kategori nilai BC (Cukup Baik). Tes kedua nilai rata-rata yang diperoleh mahasiwa darmasiswa sebesar 70,6 dengan kategori nilai B (Baik). Sedangkan, pada peningkatan penggunaan media pemerolehan nilai yang dihasilkan selalu melebihi angka yang telah ditentukan. Nilai pada pertemuan pertama mencapai 84,33 dibulatkan menjadi 85 persen. Sedangkan pertemuan kedua mencapai 88,25 dibulatkan menjadi 89 persen. artinya nilai yang diperoleh dari tes kedua diketahui lebih meningkat meskipun tes pertama nilai yang dihasilkan sudah lebih dari target.

The classical PID algorithm has proven to be a consistently popular solution across a range of different control applications. This popularity can be attributed to the algorithm offering both flexibility and robustness whilst being relatively simple to implement. The PID algorithm is given as: 1( 1) One important consideration for the optimal use of the PID controller is the selection of the parameters Kp, Tr and Td in Eq 1. Although a number of methods have been proposed for the selection of these parameters, such as the Ziegler-Nichols approach, this tuning operation continues to be a research theme. One modern approach which has been shown to give good results is the use of evolutionary algorithms such as in (1) and (2). The Particle Swarm Optimization (PSO) algorithm has proven to be particularly popular thanks to its simplicity, suitability as a function approximator and the favorable results it produces (3). This paper presents a new modeling technique as shown in Fig 2, featuring Co-Simulation between Matlab and PSIM, a specialist control and power electronics simulation environment. The tuned PID is applied to a buck converter, driven by a PV cell as shown in Fig 1. The novel solution is compared with and evaluated against a system which is tuned using a traditional method. The simulation results indicate that the controller developed using the PSO algorithm achieves a faster response time without adversely affecting peak errors. There are added benefits of this new approach, in terms of a shorter time scale for conceptual model development.

In this new approach to membrane and thermal seawater desalination processes developed by the Saline Water Conversion Corporation (SWCC), R&D Center, a nanofiltration (NF) membrane unit, which received non-coagulated filtered seawater feed, was placed ahead of the SWRO and the MSF pilot plant units to form, for the first time ever, fully integrated desalination systems of an NF-SWRO, NF-MSF, and NF-SWROreject-MSF. Preliminary results were presented at the IDA World Congress, Madrid '97. Further results obtained in this investigation at a pressure of 22 bars showed that the NF unit reduced turbidity and microorganisms, removed hardness ions of Ca++, Mg++, SO4=, HCO3−, and total hardness by 89.6%, 94.0%, 97.8%, 76.6% and 93.3%, respectively. The system also resulted in the reduction of the monovalent ions of Cl−, Na+, K+ each by 40.3% and the overall seawater TDS by 57.7%. The seawater (NF permeate water) produced by this process is considerably different in composition from (Gulf) seawater, and quality-wise is far superior to it as a feed to seawater desalination plants, and moreover without the problems normally associated with high concentration in seawater of scale forming ions, high TDS, high turbidity and microorganisms. This made it possible to operate both the SWRO and MSF pilot plants at high water recovery: 70% and 80%, respectively. It also allowed for the successful operation of the MSF unit at top brine temperature of 120°C without the addition to the make-up of antiscalant or acid or antifoam. The said desalination arrangements lead to significant improvement in the seawater desalination processes by lowering their energy consumption, by about 25–30%, and reducing chemical consumption thereby making the process more friendly to the marine environment. The observed increases in their product water output and recovery ratio by more than 70% resulted in the ultimate benefit of lowering the estimated cost of fresh water production by more than 27%. The SWRO permeate produced from the NF-SWRO arrangement has very low TDS, ≤200 ppm, making the requirement for a second-stage RO treatment of the SWRO permeate unnecessary. Moreover, the use of the desalination arrangement NF-SWROreject-MSF should allow for the conversion of up to 90% of the NF product into fresh water, where about 65% of the NF product is converted to fresh water by the SWRO unit and 25% of SWRO reject is converted by the MSF unit. Definitely, the achievement of the above results represents a milestone in seawater desalination technology. The paper describes the results obtained, the experimental approach used in this investigation along with description of the NF-SWRO, NF-MSF and NF-SWROreject-MSF pilot plants employed in this study. A simulated desalination model which compares the operation of existing SWRO plants with and without the NF modification is also described along with the technoeconomics of process analysis.

The paper presents a method for FPGA implementation of Self-Organizing Map (SOM) artificial neural networks with on-chip learning algorithm. The method aims to build up a specific neural network using generic blocks designed in the MathWorks Simulink environment. The main characteristics of this original solution are: on-chip learning algorithm implementation, high reconfiguration capability and operation under real time constraints. An extended analysis has been carried out on the hardware resources used to implement the whole SOM network, as well as each individual component block.

The aim of this paper is to review the state-of-the-art of Field Programmable Gate Array (FPGA) technologies and their contribution to industrial control applications. Authors start by addressing various research fields which can exploit the advantages of FPGAs. The features of these devices are then presented, followed by their corresponding design tools. To illustrate the benefits of using FPGAs in the case of complex control applications, a sensorless motor controller has been treated. This controller is based on the Extended Kalman Filter. Its development has been made according to a dedicated design methodology, which is also discussed. The use of FPGAs to implement artificial intelligence-based industrial controllers is then briefly reviewed. The final section presents two short case studies of Neural Network control systems designs targeting FPGAs.

This paper deals with the holistic modeling approach of a combined Photovoltaic, Wind and Fuel Cell power system. The method is using the DK5 modeling / design environment from Mentor Graphics and is based on the new Handel-C programming language. The goal of the work carried out was to achieve an optimized holistic digital control system design, followed by its rapid prototyping into a single Field Programmable Gate Array (FPGA). This would also enable easy connection (by adapting the model/controller) to the grid. The holistic functional simulation of the system is performed in the same environment as its controller hardware implementation and timing analysis. The controller design was then downloaded in hardware onto a RC10 development board containing a Xilinx Spartan FPGA and was successfully tested experimentally. This approach enables the design and fast hardware implementation of efficient controllers for Distributed Energy Resource (DER) hybrid systems.

This paper presents a hardware implementation of a multilayer feed-forward neural network based on back-propagation. The implementation is assumed to design and implement modules that emulate FF-BP functions with computing blocks of the predefined System Generator library and user defined blocks integrated in the System Generator library. The main application of the developed structure is an artificial olfactory system used to recognize the type of coffee presented in a test chamber. Data acquisition was achieved through the PC-MIO-16E-1 acquisition card and a virtual instrument, developed in Labview, for signal pre-processing and data logging into text files. The patterns presented (the type of coffee) have been recognized through neural networks. In order to select the RNA with the highest accuracy in recognising the coffee type, several different RNAs were simulated.

This paper presents the results obtained in the implementation of Feed-Forward Artificial Neural Networks (FF-ANN) with one or several layers, used in the development of smart devices that needs learning capability and adaptive behavior. The networks were implemented using ANN specific blocks created by the authors using the System Generator software. The training and the testing of the networks was conducted using sets of 150 training and test vectors with 7 elements.

In this paper we propose a method to implement SOM neural network in FPGA circuits: a self organized map neural network with on-chip learning algorithm. The method implies the building of a neural network by generic blocks designed in Mathworks' Simulink environment. The main characteristics of this solution are onchip learning algorithm implementation and high reconfiguration capability and operation under real time constraints.

When measuring level in wells or vessels using ultrasonic pulse-echo techniques, carefully positioned, narrow beam angle transducers are usually employed. When it comes to small cross sections, a sufficiently narrow beam transducer is hard to find. That is why the possibility of using large beam angle transducer had been investigated. Experiments were carried out, showing that in certain conditions, neither the beam angle, nor the transducer position is critical.

This paper presents a synthesizable programmable logic blocks architectures, describes the associated formula that makes the blocks to be generic for a backpropagation neural network (NN) with on-chip delta rule learning. The architecture proposed herein takes advantage of distinct datapaths for the forward and backward propagation stages to significantly improve the performance of the learning phase. The architecture is easily scalable and able to cope with arbitrary network sizes with the same hardware.

In this paper an ultrasonic data acquisition system and a digital signal processing algorithm in order to improve signal to noise ratio and ensure the envelope detection are presented. The data acquisition system is based on an ADuC812 micro-converter and a pair of MA400ST/R ultrasonic transducers. The received data are transmitted to a host computer via an RS232 serial interface. The digital signal processing is performed in Matlab, and, makes extensive use of the well known moving average filter (MAF) algorithm. MAF are extensively used as low-pass filters due to their good smoothing performance and ease of implementation in software and hardware as well. In this paper we will demonstrate that using the same MAF algorithm, one can address all the filtering tasks (low-pass, high-pass, notch), implied by the process of envelope detection of ultrasonic echoes.

In this paper we propose a method to implement in FPGA a feedforward neural network with on-chip delta rule learning algorithm. For this, we have develop a generic blocks designed in Mathworks' Simulink environment, capable to generate the signals for controlling the neurons from a neural network. The main characteristics of those blocks is its high reconfigurability that's makes it suitable for developing of a generic controlling block capable to manage calculus function of neurons from different layers. The properties of the block are set according to the numbers of total layers, number of the neurons from the layers and the number of layer from whom and in different function block parameters windows. The novelty of the proposed method resides in the possibility to design neural networks with on-chip learning only with predefined block systems created in system generator environment. The major benefit of this design methodology result from the possibility to create a higher level design tools used to implement neural networks in logical circuits.

Gesture recognition is a domain of grate interest of our days due to the multiple application possibilities, from the spatial or subaquatic robots manipulation, to the sign language used for communication by the peoples with hearing or speech disabilities. This paper shows an application of the artificial neural networks (ANN) implemented in field programmable gate arrays (FPGA) for the hand static gestures (postures) recognition. The adopted recognition method uses an ANN structured on two levels. The first level, a feedforward ANN trained using supervised Hebbian algorithm, is used for input data preprocessing. The second one, used for data classification is a competitive ANN. Using an ANN for input data preprocessing offers flexibility regarding the implementation of a preprocessing method. This combination of two ANN leads to 100% recognition rate for the training set and two other sets of test.

This paper presents theoretical and practical aspects in conjunction with hardware implementation of wavelet based denoising filters. The wavelets can tackle the denoising problem optimally from the point of view that the wavelet based denoising attempts to remove whatever noise is present and retain whatever signal is present regardless to the frequency content of the signal. In the field of designing signal processing systems, "time to market" represents a key factor. Hardware implementation using field programmable gate arrays (FPGA) can reduce time to market for signal processing systems. The paper analyzes and compares different solution for hardware implementation of wavelet based denoising filters using FPGAs.

Hysterektomin eliminerar reproduktionsförmågan hos kvinnan som genomgår operationen och samtidigt förändras hennes kvinnliga identitet och sexuella funktion. Allt detta leder till psykologiska konsekvenser för kvinnan, men påverkar också hennes partner. Då det inte finns mycket forskning om kvinnans upplevelser kring hysterektomi, och ännu mindre forskning kring mannens, är syftet med den här studien att belysa kvinnors och mäns upplevelse av hysterektomi. Metoden som används är en litteraturstudie där 17 vetenskapliga artiklar granskades och användes genom att sammanställa resultaten. Resultatet visar att kvinnan kan vara orolig och osäker inför sitt beslut att genomföra hysterektomi. Efter att beslutet har tagits kommer tankar och oro om möjliga konsekvenser från operationen. Kvinnan är därför i stort behov av stöd både från familj och från omvårdnadsansvarig sjuksköterska. Ett år efter hysterektomi återgår de flesta kvinnor till ett bättre liv och känner större välbefinnande. Mannen saknar information och kunskap om hysterektomi, vilket påverkar hans förmåga att stötta kvinnan. Efter hysterektomi upplever de flesta kvinnor en symtomlättnad vilket leder till en förbättrad livskvalité. Resultatet visar även att männen upplever kvinnans symtomlättnad som befriande. Det sexuella samlivet påverkas ofta till det bättre. Mannens upplevelse av hysterektomi påverkar hur han kommunicerar och stöttar sin kvinna. Därför bör också mannen sättas i fokus av omvårdnadsansvarig sjuksköterska. Undervisning som riktar sig till sjuksköterskan angående detta behövs. Forskning angående mannens upplevelse är närmast obefintlig. Kvinnans erfarenhet av hysterektomi påverkas av hur mannen uppfattar hysterektomi. Därför bör sambandet mellan hur mannens upplevelse påverkar kvinnans upplevelse utforskas mer. Konkreta exempel på en avdelning kan vara att sätta sig in i mannens förförståelse, ge honom information utifrån hans villkor och försöka bidra till kommunikation mellan parterna.

When an under-lain thrust fault slips, especially triggered by earthquakes, the overburden soil may deform and fail so that a fault zone also develops inwardly. The research about the deformation and the failure of the overburden soil is an essential issue to evaluating the safety of ground or underground structures near the potential faulted zone. In this study, a MRT tunnel, closed to a thrust fault and fault dip 60 degree, is considered, both of the sandbox experiment and the numerical analyses are adopted to discuss the damage degree of a tunnel submerged in an overburden soil under the thrust faulting. In the numerical analyses, a small-scale model, simulation of the sandbox, is justified according to the experimental results and used to discuss the base behavior of the overburden soil. In addition, a full-scale model is used to evaluate the damage degree of tunnel segments by defining a dangerous factor. Moreover, considering the real behaviors of sand particles, the distinct element method is adopted as well. In the sandbox experiment, the results indicate that the development of the shear zone was apparently hindered by the existence of a model tunnel near the fault tip, and induced significant deformation of the tunnel. In addition, the results from numerical analyses, the finite element method and the distinct element method, are similar to the experimental results. The numerical analysis results of the full-scale model indicate that the damage degree is increased while the tunnel is close to the fault tip, and the footing wall is more dangerous than the hanging wall. The defined dangerous factor is able to reflect the damage degree of the tunnel. In the distinct element analysis, the full-scale model, it can be observed that a fault zone extends from the fault tip to the ground surface, but discussions on the distribution of stress and moment in the tunnel are not included and will be considered in the future study.Copyright © 2006 by ASME