Search results for: data mining technique

Authors: Celso L. Fernando, Toshio Yoshii, Takahiro Tsubota

Abstract:

Understanding the causes of a road accident and predicting their occurrence is key to prevent deaths and serious injuries from road accident events. Traditional statistical methods such as the Poisson and the Logistics regressions have been used to find the association of the traffic environmental factors with the accident occurred; recently, an artificial neural network, ANN, a computational technique that learns from historical data to make a more accurate prediction, has emerged. Although the ability to make accurate predictions, the ANN has difficulty dealing with highly unbalanced attribute patterns distribution in the training dataset; in such circumstances, the ANN treats the minority group as noise. However, in the real world data, the minority group is often the group of interest; e.g., in the road traffic accident data, the events of the accident are the group of interest. This study proposes a combination of the k-means with the ANN to improve the predictive ability of the neural network model by alleviating the effect of the unbalanced distribution of the attribute patterns in the training dataset. The results show that the proposed method improves the ability of the neural network to make a prediction on a highly unbalanced distributed attribute patterns dataset; however, on an even distributed attribute patterns dataset, the proposed method performs almost like a standard neural network. 

Keywords: Accident risks estimation, artificial neural network, deep learning, K-mean, road safety.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 975

Authors: X. Mercilin Raajini, R. Raja Kumar, P. Indumathi, V. Praveen

Abstract:

Many cluster based routing protocols have been proposed in the field of wireless sensor networks, in which a group of nodes are formed as clusters. A cluster head is selected from one among those nodes based on residual energy, coverage area, number of hops and that cluster-head will perform data gathering from various sensor nodes and forwards aggregated data to the base station or to a relay node (another cluster-head), which will forward the packet along with its own data packet to the base station. Here a Game Theory based Diligent Energy Utilization Algorithm (GTDEA) for routing is proposed. In GTDEA, the cluster head selection is done with the help of game theory, a decision making process, that selects a cluster-head based on three parameters such as residual energy (RE), Received Signal Strength Index (RSSI) and Packet Reception Rate (PRR). Finding a feasible path to the destination with minimum utilization of available energy improves the network lifetime and is achieved by the proposed approach. In GTDEA, the packets are forwarded to the base station using inter-cluster routing technique, which will further forward it to the base station. Simulation results reveal that GTDEA improves the network performance in terms of throughput, lifetime, and power consumption.

Keywords: Cluster head, Energy utilization, Game Theory, LEACH, Sensor network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1903

Authors: S. Bennoud, M. Zergoug

Abstract:

The numerical simulation of electromagnetic interactions is still a challenging problem, especially in problems that result in fully three dimensional mathematical models.

The goal of this work is to use mathematical modeling to characterize the reliability and capacity of eddy current technique to detect and characterize defects embedded in aeronautical in-service pieces.

The finite element method is used for describing the eddy current technique in a mathematical model by the prediction of the eddy current interaction with defects. However, this model is an approximation of the full Maxwell equations.

In this study, the analysis of the problem is based on a three dimensional finite element model that computes directly the electromagnetic field distortions due to defects.

Keywords: Eddy current, Finite element method, Non destructive testing, Numerical simulations.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3142

Authors: Alejandro Paz Parra, Jose Luis Oslinger Gutierrez, Javier Olaya Ochoa

Abstract:

In the present paper, five electric current based methods to analyze electric faults on the stator of induction motors (IM) are used and compared. The analysis tries to extend the application of the multiple reference frames diagnosis technique. An eccentricity indicator is presented to improve the application of the Park’s Vector Approach technique. Most of the fault indicators are validated and some others revised, agree with the technical literatures and published results. A tri-phase 3hp squirrel cage IM, especially modified to establish different fault levels, is used for validation purposes.

Keywords: Motor fault diagnosis, induction motor, MCSA, ESA, Extended Park´s vector approach, multiparameter analysis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1688

Authors: Abeba Nigussie Turi, Xiangming Samuel Li

Abstract:

Data-driven business models are more typical for established businesses than early-stage startups that strive to penetrate a market. This paper provided an extensive discussion on the principles of data analytics for early-stage digital entrepreneurial businesses. Here, we developed data-driven decision-making (DDDM) framework that applies to startups prone to multifaceted barriers in the form of poor data access, technical and financial constraints, to state some. The startup DDDM framework proposed in this paper is novel in its form encompassing startup data analytics enablers and metrics aligning with startups' business models ranging from customer-centric product development to servitization which is the future of modern digital entrepreneurship.

Keywords: Startup data analytics, data-driven decision-making, data acquisition, data generation, digital entrepreneurship.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 827

Authors: Naveen Virmani, Rajeev Saha, Rajeshwar Sahai

Abstract:

Leagile is combination of both lean and agile system. Lean is concerned with less of everything i.e. less material, less time, less space, less manpower to produce a product, while agile is concerned with quick respond to customer demand and to reconfigure the system as soon as possible to meet the customer expectations well on time. The market is excessively competitive, so there is a dire need for the companies to adopt new and modern technologies with latest equipments. It has been seen that implementation of leagile system become tedious so the purpose of the paper is to find critical success factors (CSF) affecting leagile manufacturing system using literature review and rank them by using modified TOPSIS (Technique of order preference by similarity to ideal solution) technique.

Keywords: Agile manufacturing, lean manufacturing, leagile manufacturing, modified TOPSIS.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 994

Authors: E. Shahbazali, N. Afrasiabi, A. A. Safekordi

Abstract:

Nowadays, fluidized bed plays an important part in industry. The design of this kind of reactor requires knowing the interfacial area between two phases and this interfacial area leads to calculate the solid holdup in the bed. Consequently achieving interfacial area between gas and solid in the bed experimentally is so significant. On interfacial area measurement in fluidized bed with gas has been worked, but light transmission technique has been used less. Therefore, in the current research the possibility of using of this technique and its accuracy are investigated. Measuring, a fluidized bed was designed and the problems were averted as far as possible. By using fine solid with equal shape and diameter and installing an optical system, the absorption of light during the time of fluidization has been measured. Results indicate that this method that its validity has been proved in the gas-liquid system, by different reasons have less application in gas-solid system. One important reason could be non-uniformity in such systems.

Keywords: Fluidization, Holdup, Light Transmission, Two phase system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1503

Authors: Minsoo Lee, Yearn Jeong Kim, Yun-mi Kim, Sujeung Cheong, Sookyung Song

Abstract:

Bio-chips are used for experiments on genes and contain various information such as genes, samples and so on. The two-dimensional bio-chips, in which one axis represent genes and the other represent samples, are widely being used these days. Instead of experimenting with real genes which cost lots of money and much time to get the results, bio-chips are being used for biological experiments. And extracting data from the bio-chips with high accuracy and finding out the patterns or useful information from such data is very important. Bio-chip analysis systems extract data from various kinds of bio-chips and mine the data in order to get useful information. One of the commonly used methods to mine the data is classification. The algorithm that is used to classify the data can be various depending on the data types or number characteristics and so on. Considering that bio-chip data is extremely large, an algorithm that imitates the ecosystem such as the ant algorithm is suitable to use as an algorithm for classification. This paper focuses on finding the classification rules from the bio-chip data using the Ant Colony algorithm which imitates the ecosystem. The developed system takes in consideration the accuracy of the discovered rules when it applies it to the bio-chip data in order to predict the classes.

Keywords: Ant Colony System, DNA chip data, Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1468

Authors: Oscar Javier Herrera, Manuel Ángel Camacho

Abstract:

This paper addresses a cutting edge method of business demand forecasting, based on an empirical probability function when the historical behavior of the data is random. Additionally, it presents error determination based on the numerical method technique ‘propagation of errors.’ The methodology was conducted characterization and process diagnostics demand planning as part of the production management, then new ways to predict its value through techniques of probability and to calculate their mistake investigated, it was tools used numerical methods. All this based on the behavior of the data. This analysis was determined considering the specific business circumstances of a company in the sector of communications, located in the city of Bogota, Colombia. In conclusion, using this application it was possible to obtain the adequate stock of the products required by the company to provide its services, helping the company reduce its service time, increase the client satisfaction rate, reduce stock which has not been in rotation for a long time, code its inventory, and plan reorder points for the replenishment of stock.

Keywords: Demand Forecasting, Empirical Distribution, Propagation of Error.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1844

Authors: M. Salehi, A. A. Motie Birjandi, F. Namdari

Abstract:

Weak damping of low frequency oscillations is a frequent phenomenon in electrical power systems. These frequencies can be damped by power system stabilizers. Unified power flow controller (UPFC), as one of the most important FACTS devices, can be applied to increase the damping of power system oscillations and the more effect of this controller on increasing the damping of oscillations depends on its proper placement in power systems. In this paper, a technique based on controllability is proposed to select proper location of UPFC and the best input control signal in order to enhance damping of power oscillations. The effectiveness of the proposed technique is demonstrated in IEEE 9 bus power system.

Keywords: Unified power flow controller (UPFC), controllability, small signal analysis, eigenvalues.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1904

Authors: Adel Sabry Eesa, Adnan Mohsin Abdulazeez Brifcani, Zeynep Orman

Abstract:

Feature selection has been used in many fields such as classification, data mining and object recognition and proven to be effective for removing irrelevant and redundant features from the original dataset. In this paper, a new design of distributed intrusion detection system using a combination feature selection model based on bees and decision tree. Bees algorithm is used as the search strategy to find the optimal subset of features, whereas decision tree is used as a judgment for the selected features. Both the produced features and the generated rules are used by Decision Making Mobile Agent to decide whether there is an attack or not in the networks. Decision Making Mobile Agent will migrate through the networks, moving from node to another, if it found that there is an attack on one of the nodes, it then alerts the user through User Interface Agent or takes some action through Action Mobile Agent. The KDD Cup 99 dataset is used to test the effectiveness of the proposed system. The results show that even if only four features are used, the proposed system gives a better performance when it is compared with the obtained results using all 41 features.

Keywords: Distributed intrusion detection system, mobile agent, feature selection, Bees Algorithm, decision tree.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1940

Authors: Masoud Sadeghian, Alireza Fatehi

Abstract:

In this paper, we use nonlinear system identification method to predict and detect process fault of a cement rotary kiln. After selecting proper inputs and output, an input-output model is identified for the plant. To identify the various operation points in the kiln, Locally Linear Neuro-Fuzzy (LLNF) model is used. This model is trained by LOLIMOT algorithm which is an incremental treestructure algorithm. Then, by using this method, we obtained 3 distinct models for the normal and faulty situations in the kiln. One of the models is for normal condition of the kiln with 15 minutes prediction horizon. The other two models are for the two faulty situations in the kiln with 7 minutes prediction horizon are presented. At the end, we detect these faults in validation data. The data collected from White Saveh Cement Company is used for in this study.

Keywords: Cement Rotary Kiln, Fault Detection, Delay Estimation Method, Locally Linear Neuro Fuzzy Model, LOLIMOT.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1673

Authors: Rashid A. Fayadh, F. Malek, Hilal A. Fadhil, Norshafinash Saudin

Abstract:

At receiving high data rate in ultra wideband (UWB) technology for many users, there are multiple user interference and inter-symbol interference as obstacles in the multi-path reception technique. Since the rake receivers were designed to collect many resolvable paths, even more than hundred of paths. Rake receiver implementation structures have been proposed towards increasing the complexity for getting better performances in indoor or outdoor multi-path receivers by reducing the bit error rate (BER). So several rake structures were proposed in the past to reduce the number of combining and estimating of resolvable paths. To this aim, we suggested two improved rake receivers based on signal sign separation in the maximal ratio combiner (MRC), called positive-negative MRC selective rake (P-N/MRC-S-rake) and positive-negative MRC partial rake (P-N/MRC-S-rake) receivers. These receivers were introduced to reduce the complexity with less number of fingers and improving the performance with low BER. Before decision circuit, there is a comparator to compare between positive quantity and negative quantity to decide whether the transmitted bit is 1 or 0. The BER was driven by MATLAB simulation with multi-path environments for impulse radio time-hopping binary phase shift keying (TH-BPSK) modulation and the results were compared with those of conventional rake receivers.

Keywords: Selective and partial rake receivers, positive and negative signal separation, maximal ratio combiner, bit error rate performance.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1901

Authors: Klaus Stranacher, Vesna Krnjic, Thomas Zefferer

Abstract:

The public sector holds large amounts of data of various areas such as social affairs, economy, or tourism. Various initiatives such as Open Government Data or the EU Directive on public sector information aim to make these data available for public and private service providers. Requirements for the provision of public sector data are defined by legal and organizational frameworks. Surprisingly, the defined requirements hardly cover security aspects such as integrity or authenticity. In this paper we discuss the importance of these missing requirements and present a concept to assure the integrity and authenticity of provided data based on electronic signatures. We show that our concept is perfectly suitable for the provisioning of unaltered data. We also show that our concept can also be extended to data that needs to be anonymized before provisioning by incorporating redactable signatures. Our proposed concept enhances trust and reliability of provided public sector data.

Keywords: Trusted Public Sector Data, Integrity, Authenticity, Reliability, Redactable Signatures.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1758

Authors: K. Elleuch, A. Chaari

Abstract:

This paper deals with modeling and parameter identification of nonlinear systems described by Hammerstein model having Piecewise nonlinear characteristics such as Dead-zone nonlinearity characteristic. The simultaneous use of both an easy decomposition technique and the triangular basis functions leads to a particular form of Hammerstein model. The approximation by using Triangular basis functions for the description of the static nonlinear block conducts to a linear regressor model, so that least squares techniques can be used for the parameter estimation. Singular Values Decomposition (SVD) technique has been applied to separate the coupled parameters. The proposed approach has been efficiently tested on academic examples of simulation.

Keywords: Identification, Hammerstein model, Piecewisenonlinear characteristic, Dead-zone nonlinearity, Triangular basisfunctions, Singular Values Decomposition

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1920

Authors: K. M. Faraoun, A. Boukelif

Abstract:

In the present work, we propose a new technique to enhance the learning capabilities and reduce the computation intensity of a competitive learning multi-layered neural network using the K-means clustering algorithm. The proposed model use multi-layered network architecture with a back propagation learning mechanism. The K-means algorithm is first applied to the training dataset to reduce the amount of samples to be presented to the neural network, by automatically selecting an optimal set of samples. The obtained results demonstrate that the proposed technique performs exceptionally in terms of both accuracy and computation time when applied to the KDD99 dataset compared to a standard learning schema that use the full dataset.

Keywords: Neural networks, Intrusion detection, learningenhancement, K-means clustering

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3612

Authors: Gizem Intepe, Tufan Koc

Abstract:

S-Curves are commonly used in technology forecasting. They show the paths of product performance in relation to time or investment in R&D. It is a useful tool to describe the inflection points and the limit of improvement of a technology. Companies use this information to base their innovation strategies. However inadequate use and some limitations of this technique lead to problems in decision making. In this paper first technology forecasting and its importance for company level strategies will be discussed. Secondly the S-Curve and its place among other forecasting techniques will be introduced. Thirdly its use in technology forecasting will be discussed based on its advantages, disadvantages and limitations. Finally an application of S-curve on 3D TV technology using patent data will also be presented and the results will be discussed.

Keywords: Patent analysis, Technological forecasting. S curves, 3D TV

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7784

Authors: H. B. Kekre, Tanuja K. Sarode

Abstract:

This paper presents a very simple and efficient algorithm for codebook search, which reduces a great deal of computation as compared to the full codebook search. The algorithm is based on sorting and centroid technique for search. The results table shows the effectiveness of the proposed algorithm in terms of computational complexity. In this paper we also introduce a new performance parameter named as Average fractional change in pixel value as we feel that it gives better understanding of the closeness of the image since it is related to the perception. This new performance parameter takes into consideration the average fractional change in each pixel value.

Keywords: Vector Quantization, Data Compression, Encoding, Searching.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1613

Authors: Sareh Rajabi, Salwa Bheiry

Abstract:

In general, Covid-19 created many financial and non-financial damages to the economy and community. Level and severity of covid-19 as pandemic case varies over the region and due to different types of the projects. Covid-19 virus emerged as one of the most imperative risk management factors word-wide recently. Therefore, as part of portfolio management assessment, it is essential to evaluate severity of such risk on the project and program in portfolio management level to avoid any risky portfolio. Covid-19 appeared very effectively in South America, part of Europe and Middle East. Such pandemic infection affected the whole universe, due to lock down, interruption in supply chain management, health and safety requirements, transportations and commercial impacts. Therefore, this research proposes Analytical Hierarchy Process (AHP) to analyze and assess such pandemic case like Covid-19 and its impacts on the construction projects. The AHP technique uses four sub-criteria: Health and safety, commercial risk, completion risk and contractual risk to evaluate the project and program. The result will provide the decision makers with information which project has higher or lower risk in case of Covid-19 and pandemic scenario. Therefore, the decision makers can have most feasible solution based on effective weighted criteria for project selection within their portfolio to match with the organization’s strategies.

Keywords: Portfolio management, risk management, COVID-19, analytical hierarchy process technique.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 832

Authors: Ahmed Z. Gabr, Ahmed A. Helal, Hussein E. Said

Abstract:

With the incessant increase of power systems capacity and voltage grade, the safety of grounding grid becomes more and more prominent. In this paper, the designing substation grounding grid is presented by means of genetic algorithm (GA). This approach purposes to control the grounding cost of the power system with the aid of controlling grounding rod number and conductor lengths under the same safety limitations. The proposed technique is used for the design of the substation grounding grid in Khalda Petroleum Company “El-Qasr” power plant and the design was simulated by using CYMGRD software for results verification. The result of the design is highly complying with IEEE 80-2000 standard requirements.

Keywords: Genetic algorithm, optimum grounding grid design, power system analysis, power system protection, single layer model, substation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2804

Authors: Yasaswi Palagummi, Sareh Rowlands

Abstract:

Generalised Zero-Shot Learning, often known as GZSL, is an advanced variant of zero-shot learning in which the samples in the unseen category may be either seen or unseen. GZSL methods typically have a bias towards the seen classes because they learn a model to perform recognition for both the seen and unseen classes using data samples from the seen classes. This frequently leads to the misclassification of data from the unseen classes into the seen classes, making the task of GZSL more challenging. In this work, we propose an approach leveraging the Shifted Window based Self-Attention in the Swin Transformer (Swin-GZSL) to work in the inductive GZSL problem setting. We run experiments on three popular benchmark datasets: CUB, SUN, and AWA2, which are specifically used for ZSL and its other variants. The results show that our model based on Swin Transformer has achieved state-of-the-art harmonic mean for two datasets - AWA2 and SUN and near-state-of-the-art for the other dataset - CUB. More importantly, this technique has a linear computational complexity, which reduces training time significantly. We have also observed less bias than most of the existing GZSL models.

Keywords: Generalised Zero-shot Learning, Inductive Learning, Shifted-Window Attention, Swin Transformer, Vision Transformer.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 222

Authors: Ekachai Phaisangittisagul, Rapeepol Chongprachawat

Abstract:

Obtaining labeled data in supervised learning is often difficult and expensive, and thus the trained learning algorithm tends to be overfitting due to small number of training data. As a result, some researchers have focused on using unlabeled data which may not necessary to follow the same generative distribution as the labeled data to construct a high-level feature for improving performance on supervised learning tasks. In this paper, we investigate the impact of the relationship between unlabeled and labeled data for classification performance. Specifically, we will apply difference unlabeled data which have different degrees of relation to the labeled data for handwritten digit classification task based on MNIST dataset. Our experimental results show that the higher the degree of relation between unlabeled and labeled data, the better the classification performance. Although the unlabeled data that is completely from different generative distribution to the labeled data provides the lowest classification performance, we still achieve high classification performance. This leads to expanding the applicability of the supervised learning algorithms using unsupervised learning.

Keywords: Autoencoder, high-level feature, MNIST dataset, selftaught learning, supervised learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1832

Authors: A. Tashakori, M. Ektesabi

Abstract:

Electric vehicle (EV) is one of the effective solutions to control emission of greenhouses gases in the world. It is of interest for future transportation due to its sustainability and efficiency by automotive manufacturers. Various electrical motors have been used for propulsion system of electric vehicles in last decades. In this paper brushed DC motor, Induction motor (IM), switched reluctance motor (SRM) and brushless DC motor (BLDC) are simulated and compared. BLDC motor is recommended for high performance electric vehicles. PWM switching technique is implemented for speed control of BLDC motor. Behavior of different modes of PWM speed controller of BLDC motor are simulated in MATLAB/SIMULINK. BLDC motor characteristics are compared and discussed for various PWM switching modes under normal and inverter fault conditions. Comparisons and discussions are verified through simulation results.

Keywords: BLDC motor, PWM switching technique, in-wheel technology, electric vehicle.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4845

Authors: Samppa Suoniemi, Harri Terho, Rami Olkkonen

Abstract:

In recent methodological articles related to structural equation modeling (SEM), the question of how to measure endogenous formative variables has been raised as an urgent, unresolved issue. This research presents an empirical application from the CRM system development context to test a recently developed technique, which makes it possible to measure endogenous formative constructs in structural models. PLS path modeling is used to demonstrate the feasibility of measuring antecedent relationships at the formative indicator level, not the formative construct level. Empirical results show that this technique is a promising approach to measure antecedent relationships of formative constructs in SEM.

Keywords: CRM system development, formative measures, PLS path modeling, research methodology.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2262

Authors: M. Klimavicius

Abstract:

This paper proposes a modeling methodology for the development of data analysis solution. The Author introduce the approach to address data warehousing issues at the at enterprise level. The methodology covers the process of the requirements eliciting and analysis stage as well as initial design of data warehouse. The paper reviews extended business process model, which satisfy the needs of data warehouse development. The Author considers that the use of business process models is necessary, as it reflects both enterprise information systems and business functions, which are important for data analysis. The Described approach divides development into three steps with different detailed elaboration of models. The Described approach gives possibility to gather requirements and display them to business users in easy manner.

Keywords: Data warehouse, data analysis, business processmanagement.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1393

Authors: Domingo López-Rodríguez, Enrique Mérida-Casermeiro, Juan M. Ortiz-de-Lazcano-Lobato

Abstract:

Hopfield model of associative memory is studied in this work. In particular, two main problems that it possesses: the apparition of spurious patterns in the learning phase, implying the well-known effect of storing the opposite pattern, and the problem of its reduced capacity, meaning that it is not possible to store a great amount of patterns without increasing the error probability in the retrieving phase. In this paper, a method to avoid spurious patterns is presented and studied, and an explanation of the previously mentioned effect is given. Another technique to increase the capacity of a network is proposed here, based on the idea of using several reference points when storing patterns. It is studied in depth, and an explicit formula for the capacity of the network with this technique is provided.

Keywords: Associative memory, Hopfield network, network capacity, spurious patterns.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1108

Authors: Nuawi M. Z., Lamin F., Ismail A. R., Abdullah S., Wahid Z.

Abstract:

A filter is used to remove undesirable frequency information from a dynamic signal. This paper shows that the Znotch filter filtering technique can be applied to remove the noise nuisance from a machining signal. In machining, the noise components were identified from the sound produced by the operation of machine components itself such as hydraulic system, motor, machine environment and etc. By correlating the noise components with the measured machining signal, the interested components of the measured machining signal which was less interfered by the noise, can be extracted. Thus, the filtered signal is more reliable to be analysed in terms of noise content compared to the unfiltered signal. Significantly, the I-kaz method i.e. comprises of three dimensional graphical representation and I-kaz coefficient, Z∞ could differentiate between the filtered and the unfiltered signal. The bigger space of scattering and the higher value of Z∞ demonstrated that the signal was highly interrupted by noise. This method can be utilised as a proactive tool in evaluating the noise content in a signal. The evaluation of noise content is very important as well as the elimination especially for machining operation fault diagnosis purpose. The Z-notch filtering technique was reliable in extracting noise component from the measured machining signal with high efficiency. Even though the measured signal was exposed to high noise disruption, the signal generated from the interaction between cutting tool and work piece still can be acquired. Therefore, the interruption of noise that could change the original signal feature and consequently can deteriorate the useful sensory information can be eliminated.

Keywords: Digital signal filtering, I-kaz method, Machiningmonitoring, Noise Cancelling, Sound

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1884

Authors: Débora N. Zambrano, Marina O. Gosatti, Leandro M. Dufou, Daniel A. Serrano, M. Mónica Guraya, Soledad Perez-Catán

Abstract:

Nanoporous g-Al₂O₃ samples were synthesized via a sol-gel technique, introducing changes in the Yoldas´ method. The aim of the work was to achieve an effective control of the nanostructure properties and morphology of the final g-Al₂O₃. The influence of the reagent temperature during the hydrolysis was evaluated in case of water at 5 ºC and 98 ºC, and alkoxide at -18 ºC and room temperature. Sol-gel transitions were performed at 120 ºC and room temperature. All g-Al₂O₃ samples were characterized by X-ray diffraction, nitrogen adsorption and thermal analysis. Our results showed that temperature of both water and alkoxide has not much influence on the nanostructure of the final g-Al₂O₃, thus giving a structure very similar to that of samples obtained by the reference method as long as the reaction temperature above 75 ºC is reached soon enough. XRD characterization showed diffraction patterns corresponding to g-Al₂O₃ for all samples. Also BET specific area values (253-280 m²/g) were similar to those obtained by Yoldas’s original method. The temperature of the sol-gel transition does not affect the resulting sample structure, and crystalline boehmite particles were identified in all dried gels. We analyzed the reproducibility of the samples’ structure by preparing different samples under identical conditions; we found that performing the sol-gel transition at 120 ºC favors the production of more reproducible samples and also reduces significantly the time of the sol-gel reaction.

Keywords: Nanostructure alumina, boehmite, sol-gel technique, N2 adsorption/desorption isotherm, pore size distribution, BET area.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1345

Authors: Olga V. Kharchenko

Abstract:

Applicability of a KTA crystal-based laser system with optical parametric oscillators (OPO) generation to lidar sounding of the atmosphere in the spectral range 3–4 µm is studied in this work. A technique based on differential absorption lidar (DIAL) method and differential optical absorption spectroscopy (DOAS) is developed for lidar sounding of trace atmospheric gases (TAG). The DIAL-DOAS technique is tested to estimate its efficiency for lidar sounding of atmospheric trace gases.

Keywords: Atmosphere, lidar sounding, DIAL, DOAS, trace gases, nonlinear crystal.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2178

Authors: Abiodun M. Aibinu, Momoh J. E. Salami, Amir A. Shafie, Athaur Rahman Najeeb

Abstract:

Autoregressive Moving average (ARMA) is a parametric based method of signal representation. It is suitable for problems in which the signal can be modeled by explicit known source functions with a few adjustable parameters. Various methods have been suggested for the coefficients determination among which are Prony, Pade, Autocorrelation, Covariance and most recently, the use of Artificial Neural Network technique. In this paper, the method of using Artificial Neural network (ANN) technique is compared with some known and widely acceptable techniques. The comparisons is entirely based on the value of the coefficients obtained. Result obtained shows that the use of ANN also gives accurate in computing the coefficients of an ARMA system.

Keywords: Autoregressive moving average, coefficients, back propagation, model parameters, neural network, weight.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2290