Implementation Of Artificial Neural Networks Research Articles

ANALYSIS OF HISTORICAL DATA USING ARTIFICIAL INTELLIGENCE TECHNOLOGIES FOR PREDICTING THREATS IN COMPLEX DISTRIBUTED SYSTEMS

A threat prediction method based on the mining of historical data in complex distributed systems is proposed. The relevance of the selected research topic is substantiated from the point of view of considering floods as a physical process of water rise, the level of which is measured at stationary hydrological posts. The mathematical formulation of the problem is formulated, within the framework of which an artificial neural network is implemented based on the free software library “TensorFlow”. An analysis of the effectiveness of the implemented artificial neural network was carried out, according to the results of which the weighted mean square-law deviation of the predicted water level value from the actual one when forecasting for one day at stationary hydrological posts was 0.032. Thus, the neural network allows predicting the flood situation with acceptable accuracy, which gives time for special services to carry out measures to counter this threat.

Implementation of Back Propagation Artificial Neural Network for Heart Disease Abnormality Diagnosis

Heart is a very important human organ, where in normal people heart beats at 60-100 beats per second, but there are abnormalities in the heart rate that can occur due to certain causes so that it becomes slower (bradycardia) or faster (tachycardia). Electrical activity of the heart can be detected by an electrocardiogram (ECG), where the output of this device is a signal that describes the condition of a person’s heart. Artificial neural network (ANN) is one of the learning methods of artificial intelligence that can be used for pattern recognition or the other. One of the ANN learning paradigms is backpropagation where the computation goes through 2 stages, namely advanced calculation and backward calculation. This study aims to simulate backpropagation neural networks to recognize patterns from the output of the electrocardiogram using the MATLAB program. The input is form of printed electrocardiogram recording, and then it is normalized, next the data is processed by backpropagation computing with two phases (training phase and testing phase). The output of this ANN is a description of a patient’s condition whether normal, bradycardia or tachycardia.

Character Recognition By Implementing FPGA-Based Artificial Neural Network

A non-linear applied math knowledge modelling tool, Artificial Neural Networks (ANN) are predominantly used to model complicated interactions between inputs and outputs or to look for patterns within the data. Using VHDL coding, we developed a generic hardware-based ANN. This classifier has been trained to recognize letters on a 4x4 binary grid that a user fills out using 16 toggle switches. An LCD shows the most likely classification that the ANN proposed. The ANN was taught to recognize 20 English character patterns and 9 Arabic character patterns on a 4x4 grid to showcase the viability of the FPGA execution of ANN. The Altera DE2 development and education board use the generated design file to flash the Altera Cyclone II FPGA. A training supervisor is also included in the design, which is responsible for training the ANN to recognize a set of 29 standard characters drawn from the English and Arabic alphabets. Positive patterns were found, with the ANN successfully recognizing all training characters. Finally, this study demonstrates the adaptability and boundless possibility of hardware-based implementation of ANN and the infinite possibility of prototyping and building recent systems with VHDL coding on an FPGA platform.

Implementation of Artificial Neural Networks Using Magnetoresistive Random-Access Memory-Based Stochastic Computing Units

Hardware implementation of artificial neural networks (ANNs) using conventional binary arithmetic units requires large area and energy, due to the massive multiplication and addition operations in the inference process, limiting their use in edge computing and emerging Internet of Things (IoT) systems. Stochastic computing (SC), where the probability of 1s and 0s in a randomly generated bit-stream is used to represent a decimal number, has been devised as an alternative for compact and low-energy arithmetic hardware, due to its ability to implement basic arithmetic operations using far fewer logic gates than binary operations. To realize SC in hardware, however, tunable true random-number generators (TRNGs) are needed, which cannot be efficiently realized using existing complementary metal-oxide-semiconductor complementary metal-oxide-semiconductor (CMOS) technology. In this letter, we address this challenge by using magnetic tunnel junctions (MTJs) as TRNGs, the stochasticity of which can be tuned by an electric current via spin-transfer torque. We demonstrate the implementation of ANNs with SC units, using stochastic bit-streams experimentally generated by a series of 50 nm perpendicular MTJs. The numerical value (1 to 0 ratio) of the bit-streams is tuned by the current through the MTJs via spin-transfer torque with an ultralow current of <5 µA (= 0.25 MA⋅cm-2). The MTJ-based SC-ANN achieves 95% accuracy for handwritten digit recognition on the MNIST database. MRAM-based SC-ANNs provide a promising solution for ultra-low-power machine learning in edge, mobile, and IoT devices.

Implementation of Artificial Neural Network to Predict the Permeability and Solubility Models of Gypseous Soil

The objective of the present study is to make a database that describes the leaching-permeability behavior of collapsible gypseous soil. The data will be implemented to develop ANN prediction models for predicting the saturated coefficient of permeability and percentage of solubility by weight. The complex soil behavior and tedious and time consume in soil testing have driven researchers to use Artificial Neural Network (ANN) as tool for prediction. The objectives of the study were to investigate leaching-permeability behavior of collapsible gypseous soils and to develop ANN models for estimating the saturated coefficient of permeability and solubility of the soils. The MATLAB R2015a software was used to predict the saturated coefficient of permeability and the solubility percentage by weight of gypseous soils. The dataset used in this work included (513) records of experimental measurements extracted from leaching-permeability tests conducted on gypseous soil samples taken from Baher Al-Najaf in Iraq. Four input variables were investigated to have the most important influence on the permeability and solubility percentage by weight. According to the achieved statistical analysis, the ANNs model have a reliable capability to find out the predictions with a high-level of accuracy. The gypseous soils exhibited a high rate of dissolution of soluble minerals content, which caused increase in the coefficient of permeability as the soil samples reach the state of long-term full saturation.

Implementation of Artificial Neural Network in Forecasting Sales Volume in Tokopedia Indonesia

Predicting sales is one way to get company profits. Tokopedia Indonesia is one of the marketplaces that is included in the type of e-commerce customer to customer (C2C). This research was conducted in order to help sellers in the Tokopedia Indonesia marketplace to predict the sales of their merchandise, so that sellers can prepare or stock items that are predicted to increase in sales by implementing Artificial Neural Networks. Artificial neural networks can help predict future sales values. The data is divided into training data and testing data. The results of the analysis of this study indicate that the network model obtained reaches an accuracy rate of 95%.

РАЗРАБОТКА МЕТОДОВ ОПРЕДЕЛЕНИЯ И ОБЕСПЕЧЕНИЯ НАДЕЖНОСТИ ИСКУССТВЕННЫХ НЕЙРОННЫХ СЕТЕЙ НА БАЗЕ МЕМРИСТИВНЫХ УСТРОЙСТВ

The article is devoted to the reliability of hardware implementation of artificial neural networks based on memristive devices (ANNM). On the basis of the system engineering methodology, the authors have developed a universal general approach to determining the reliability of ANNM “from the accuracy of functioning through fault tolerance”. The active and passive methods of ensuring the reliability of ANNM are described. An example of determining and ensuring the reliability of a specific version of the hardware implementation of ANNM is given.

Prediction of Water Quality Index in Drinking Water Distribution System Using Activation Functions Based Ann

Determination of the drastic changes in water quality is an urgent need in this polluted era and is more essential for the survival of the existing and growing water demand. It has been very difficult to analyze the water quality data. This study focused on the Water Quality Index (WQI) prediction of water samples collected from 1944 different wells surrounding the Vellore district. WQI prediction is carried out by ANN (i.e.) Artificial Neural Networks implementation which has used 15 groundwater variables that are collected in different parts of the Vellore district from 2008 to 2017. If 15 underground variable values meet the desired range then WQI is considered as better and appropriate for drinking. But if any one of the value doesn’t meet the desired range then it is not considered as better and hence not suitable for drinking. In this study the pre-processing of the collected data has been completed to reduce the computational time. Further feature extraction techniques are used to extract the required features. The extracted features are passed on to ANN classifiers that possess three activation functions like Tanh, Maxout, and rectifier. The novelty of this paper is that WQI is determined by combining the three activation functions like Tanh, Maxout, and rectifier. A comparative analysis has been performed for proposed work related with various methodologies.

Implementation of Radial Basis Function Artificial Neural Network into an Adaptive Equivalent Consumption Minimization Strategy for Optimized Control of a Hybrid Electric Vehicle

Continued increases in the emission of greenhouse gases by passenger vehicles have accelerated the production of hybrid electric vehicles. With this increase in production, there has been a parallel demand for continuously improving strategies of hybrid electric vehicle control. The goal of an ideal control strategy is to maximize fuel economy while minimizing emissions. Methods exist by which the globally optimal control strategy may be found. However, these methods are not applicable in real-world driving applications since these methods require a priori knowledge of the upcoming drive cycle. Real-time control strategies use the global optimal as a benchmark against which performance can be evaluated. The goal of this work is to use a previously defined strategy that has been shown to closely approximate the global optimal and implement a radial basis function (RBF) artificial neural network (ANN) that dynamically adapts the strategy based on past driving conditions. The strategy used is the Equivalent Consumption Minimization Strategy (ECMS), which uses an equivalence factor to define the control strategy and the power train component torque split. An equivalence factor that is optimal for a single drive cycle can be found offline with a priori knowledge of the drive cycle. The RBF-ANN is used to dynamically update the equivalence factor by examining a past time window of driving characteristics. A total of 30 sets of training data (drive cycles) are used to train the RBF-ANN. For the majority of drive cycles examined, the RBF-ANN implementation is shown to produce fuel economy values that are within ±2.5% of the fuel economy obtained with the optimal equivalence factor. The advantage of the RBF-ANN is that it does not require a priori drive cycle knowledge and is able to be implemented in real-time while meeting or exceeding the performance of the optimal ECMS. Recommendations are made on how the RBF-ANN could be improved to produce better results across a greater array of driving conditions.

Threshold Voltage for Digital Residual Current Circuit Breaker Based on Tensor Flow

The article describes a new way to define the threshold voltage for Machine Learning-based Digital Residual Current Circuit Breaker (RCCB), enabling the right cut-off point. Using the described methods, the authors obtained a gap to common mid voltage being around 0.5 V. The proposed technique is illustrated with three different loads of 3W, 5W, and 9W as the scope of this work. The authors try to apply it in Residual Current Circuit Breaker (RCCB). It could be useful in a hospital with a limited number of technicians to maintain various machines quickly. This work tries to realize a machine that could find out the best condition to cut off the electricity when there is any leakage current but keep the supply if it is still under tolerance. This allows improving the mistake of the midpoint about 16.97% over its wide range. The effectiveness of Python libraries usage realized the Artificial Neural Network (ANN) implementation as one of machine-learning algorithms. The learning process is applied to the measured leakage current data set. It goes with input preprocessing, training, testing, and data analysis. From all of those steps, it is possible to determine the induction voltage threshold at 1.080 from 3.3V as its maximum value with a negligible loss value of 0.0006. By comparing the value with a reference, it can be concluded that this method could be used in a real situation.

Pipeline FPGA-Based Implementations of ANNs for the Prediction of up to 600-Steps-Ahead of Chaotic Time Series

Chaotic time series prediction can be performed by applying different architectures of artificial neural networks (ANNs) that can be implemented on field-programmable gate arrays (FPGAs). However, the main challenges are the reduction of hardware resources to develop faster ANNs and the prediction capabilities for large horizons. In this manner, the contribution is devoted to introduce pipeline architectures in which some registers are placed between combinational blocks to divide the logic into shorter stages that can run with a faster clock. The cases of study are the multilayer perceptron (MLP), nonlinear autoregressive with exogenous input (NARX), and echo state network (ESN). In addition, another contribution is devoted to introduce the application of the decimation technique to extend the prediction horizon of the ANNs from 12 to 600-steps-ahead. The prediction capabilities of the MLP, NARX and ESN are compared by using eight chaotic time series with different maximum Lyapunov exponents. The pipeline FPGA-based implementations show that the ESN with a reservoir of at least 30 neurons guarantees a large prediction horizon of 600-steps-ahead. Another important advantage of the ESN is that its FPGA-based implementation can be performed by reusing one neuron, thus requiring the lowest quantity of hardware resources.

Intelligent Data Analysis for Forecasting Threats in Complex Distributed Systems

A threat prediction method based on the intellectual analysis of historical data in complex distributed systems (CDS) is proposed. The relevance of the chosen research topic in terms of considering the flood as a physical process of raising the water level, which is measured at stationary and automatic hydrological posts, is substantiated. Based on this, a mathematical formulation of the problem is formulated, within the framework of which an artificial neural network based on the freely distributed TensorFlow software library is implemented. The analysis of the effectiveness of the implemented artificial neural network was carried out, according to which the average deviation of the predicted water level when forecasting for one day at a stationary hydrological post was 3.323%. For further research on forecasting water levels, an algorithm is proposed for evaluating historical data at automatic posts, which will allow using these data to predict water levels according to the proposed method and at automatic posts. Thus, the neural network allows predicting the flood situation with acceptable accuracy, which allows special services to take measures to counter this threat.

Development of a low-cost digital image processing system for oranges selection using hopfield networks

The orange processing industry has grown vertiginously around the world recently once that orange derived products are embedded in a profitable market that constitutes a substantial part of the economy in many countries. Normally, orange classification in industries still is performed manually or using expensive technologies. Recent researches aimed to develop systems capable of executing this task considering elements of artificial intelligence to find ways of automating this process. This work aims to present the development of a low-cost oranges classification system through image processing and artificial neural networks concepts for classification and prediction of their main characteristics. Therefore, a systematic photographic and methodological procedure was applied for image processing and implementation of Hopfield recurrent artificial neural networks creating a trustworthy selection system. The results obtained achieved an acceptable average percentage of 85% for correct answers considering both criteria of quality and size, which means that the implemented system reaches similar or better results when compared to methods proposed in similar works. Additionally, an economic analysis indicated a favorable payback between 3 and 5 months, attesting the feasibility of its implementation. Overall, this work ensures an effective orange selection system with minimal human contact and low-cost for the orange industry.

КОМБІНОВАНА МЕТРИКА ВІЗУАЛЬНОЇ ЯКОСТІ ЗОБРАЖЕНЬ ДИСТАНЦІЙНОГО ЗОНДУВАННЯ НА ОСНОВІ НЕЙРОННОЇ МЕРЕЖІ

The wide distribution of images of remote sensing (RS) of the Earth in various application areas makes it important to ensure the high quality of such images, which is important to identify necessary information. The complexity of the systems and the impact of various physical processes cause a significant number of distortions that lead to image corruption and possible loss of information. The use of processing methods that should reduce the impact of such factors requires control of their work, which uses quantitative indicators of visual quality. The article considers the task of creating a combined visual quality metric based on an artificial neural network (ANN), which provides high accuracy of visual quality assessment and stability of work on the noise characteristic of the RS. The problem of analysis of RS distortions is considered and the approach of using the database of test images TID2013 for verification on typical RS distortions is offered. The analysis of well-known visual quality metrics and their suitability for the estimation of such images is carried out. According to its results, it was determined that the best metrics provide the accuracy of image quality assessment for RS tasks at the level of 0.93 according to Spearman's rank-order correlation coefficient with subjective estimates of the TID2013 image database. The joint application of existing quality metrics allows eliminating the shortcomings of each of them and increasing the overall efficiency, so the article considers the problems and defines the requirements for creating a combined metric involving a neural network. A method of limiting the number of involved quality metrics with the involvement of Lasso regularization is proposed, which allows determining the most informative features (quality metrics) and simplifying the procedure of selection and reduction of their number. A study was conducted on the influence of the metric selection criterion and quantity on the accuracy of the combined metric. The influence of the structure of the neural network, the number of hidden layers, and the number of neurons in them are also analyzed. Based on the obtained results, the best implementation of ANN was selected, which with the involvement of 16 visual quality metrics allows achieving the accuracy of visual quality assessment at 0.97 according to Spearman's correlation with subjective estimates of the TID2013 database.

Development and implementation of artificial neural network in cooling system of cast rolls on 1950 LPK mill in order to increase stability of working rolls and improve quality of products

Automation and production management play an important role in ensuring product quality. Metallurgical production includes many types of technological equipment, which has its own technical feature and specificity. In recent years, there has been a breakthrough in the development and use of artificial neural networks in electronics. With their help, it is possible to fully automate production from the beginning to the very sale of products, to improve the quality of products, since such a system is capable of self-learning using the obtained databases in production and simulating the production process in software outside the process itself, improving production based on input and output parameters, partially without human participation, and to work autonomously in the further development of control technology and self-learning. A model of modernization of the existing CAPCS (Computer-Aided Process Control System) of the 1950 mill of LPK JSC in the Town of Vyksa is proposed in order to create the most favorable cooling mode for each standard size of the rolled strip.

Mathematical modeling of a self-learning neuromorphic network based on nanosized memristive elements with 1T1R crossbar architecture

Artificial neural networks play an important role in the modern world. Their main field of application is the tasks of recognition and processing of images, speech, as well as robotics and unmanned systems. The use of neural networks is associated with high computational costs. In part, it was this fact that held back their progress, and only with the advent of high-performance computing systems did the active development of this area begin. Nevertheless, the issue of speeding up the work of neural network algorithms is still relevant. One of the promising directions is the creation of analog implementations of artificial neural networks, since analog calculations are performed orders of magnitude faster than digital ones. The memristor acts as the basic element on which such systems are built. A memristor is a resistance, the conductivity of which depends on the total charge passed through it. Combining them into a matrix (crossbar) allows one layer of artificial synapses to be implemented at the hardware level. Traditionally, the STDP method based on Hebb’s rule has been used as an analog learning method. In this work, we are modeling a two-layer fully connected network with one layer of synapses. The memristive effect can manifest itself in different substances (mainly in different oxides), so it is important to understand how the characteristics of memristors will affect the parameters of the neural network. Two oxides are considered: titanium oxide (TiO2) and hafnium oxide (HfO2). For each oxide, a parametric identification of the corresponding mathematical model is performed to best fit the experimental data. The neural network is tuned depending on the oxide used and the process of training it to recognize five patterns is simulated.

Power flow management of grid connected HRES system using a CFANN technique

ABSTRACT This manuscript presents a hybrid technique-based power flow management of grid-connected Hybrid Renewable energy sources (HRES). HRES has Photovoltaic (PV), Wind Turbine (WT) as well as Diesel Generator (DG). The proposed hybrid technique is the combined implementation of Cuttlefish Algorithm (CFA) and artificial neural network (ANN) and commonly called CFANN technique. Atproposed technique, CFA is used to create the precise control signals for system and builds the control signal database off-line based on power range among source side and load side. Dataset made to operate ANN online and lead the control procedure at less implementation time. The behaviours of system are examined depend on objective function. To analyse power flow management, equality and inequality restrictions are distinct that specifies the renewable energy source availability and power demand. By this appropriate control, HRES can considerably improve the dynamic security of power system. The result shows that the energy sources of renewable energy system and its power converters are controlled and the error is reduced by 92.42%. At that time, proposed model is implemented on MATLAB/Simulink work platform and implementation is evaluated to existing methods.

СПОСОБ ИСПОЛЬЗОВАНИЯ СРЕДСТВ БЫСТРОГО ПРОТОТИПИРОВАНИЯ ДЛЯ РЕАЛИЗАЦИИ СВЕРТОЧНОЙ НЕЙРОННОЙ СЕТИ НА ПЛИС

Research in the field of artificial intelligence is carried out with increasing interest everyyear. The fields of application of artificial intelligence are quite extensive: automation, analysis ofa large amount of data, smart home technology, machine vision, etc. Artificial intelligence technologiesare based on the use of artificial neural networks, which are based on the principles ofthe animal nervous system. In this case, the actual issue is the implementation of artificial neuralnetworks on various software and hardware platforms: programmable logic integrated circuits ofthe FPGA type (Field Programmable Gate Array), on special purpose integrated circuits (Application-Specific Integrated Circuit, ASIC), GPU, CPU etc. FPGA performs best in low-power mobilesystems. ASIC demonstrates the highest performance at a fairly high development cost.The problem of rapid prototyping of projects based on the use of artificial neural networks forFPGAs using conventional methods (using HDL languages, HDL encoders, graphic programming)is that either such a project is complex and time-consuming to debug (HDL languages), orthe resulting code is not optimal (HDL encoders), or the duration of the project development andthe complexity of reconfiguring the neural network (graphical programming) are high. Therefore,in the framework of this work, an effective method for designing fully connected and convolutionalneural networks for their implementation on FPGAs using the Xilinx System Generator for DSPand Matlab / Simulink package is considered. Artificial neural networks generated in this way areeasily reconfigurable and allow solving the following problems: image recognition, optimal filtering(for example, for problems of subsurface radar).

Hybrid CMOS/memristor crossbar structure for implementing hopfield neural network

Competent hardware implementation of artificial neural networks is still an important contest. In recent literature, memristor has been introduced as a promising candidate for synapses implementation. However, integrating a memristive circuit with neuron hardware is auspicious research with challenging issues. In this paper, a scalable circuit-level hybrid CMOS/memristor hardware is introduced for implementing a Hopfield neural network. The proposed circuit is fully compatible with the crossbar structure. The performance of the proposed hardware is evaluated for different scales and compared with its software-based counterpart. Moreover, the accuracy of large-scale hardware for Hopfield neural network with 45 neurons and 4320 memristors is evaluated. It is demonstrated that the performance of the circuit is in the line of the software simulation. In comparison with similar works, the proposed circuit consumes 2000 times less energy and retrieves patterns 130 times faster. The implemented circuit is a step toward a general and feasible memristive hardware implementation for recurrent neural networks.

Second Generation Current Conveyor Based Floating Fractional Order Memristance Simulator and a New Dynamical System

Abstract In recent years, due to its non-volatile memory, non-locality, and weak singularity features, fractional calculations have begun to take place frequently in artificial neural network implementations and learning algorithms. Therefore, there is a need for circuit element implementations providing fractional function behaviors for the physical realization of these neural networks. In this study, a previously defined integer order memristor element equation is changed and a fractional order memristor is given in a similar structure. By using the obtained mathematical equation, frequency-dependent pinched hysteresis loops are obtained. A memristance simulator circuit that provides the proposed mathematical relationship is proposed. Spice simulations of the circuit are run and it is seen that they are in good agreement with the theory. Also, the non-volatility feature has been demonstrated with Spice simulations. The proposed circuit can be realized by using the integrated circuit elements available on the market. With a small connection change, the proposed structure can be used to produce both positive and negative memristance values.