Farshid Keynia - Academia.edu (original) (raw)
Papers by Farshid Keynia
Scientia Iranica
The present study considers decreasing prediction error for the types of time series and the unce... more The present study considers decreasing prediction error for the types of time series and the uncertainty in estimation parameters, improving the structure of the deep neural network and increasing response speed in the proposed neural network method; besides, the competitive performance and the collaboration among the neurons of deep neural network are also increased. Selected data is related to Qeshm weather (suitable weather conditions to study our purpose) prediction during 2016 onwards. In this study, for the purpose of analyzing the prediction issue of power consumption of domestic expenses in the indefinite and severe fluctuation mode, we decided to combine two methods of Long Short-Term Memory and Convolutional Neural Network. For the training of the deep network, the BP algorithm is used. The results indicate that Gated Recurrent Unit networks compared other models (MLP, CNN and DNN) produce more realistic results, and also two-way networks obtained better results on test data compared Long Short-Term Memory networks. RMSE prediction are more realistic than the LSTM model on test and training data against the significant data. A GRU network has two gates of rt readjust and Zt forgetting, which helps to assure that long term dependencies of gradient fading will not occur.
The urban book series, 2023
Wind speed is one of the most vital, imperative meteorological parameters, thus the prediction of... more Wind speed is one of the most vital, imperative meteorological parameters, thus the prediction of which is of fundamental importance in the studies related to energy management, building construction, damages caused by strong winds, aquatic needs of power plants, the prevalence and spread of diseases, snowmelt, and air pollution. Due to the discrete and nonlinear structure of wind speed, wind speed forecasting at regular intervals is a crucial problem. In this regard, a wide variety of prediction methods have been applied. So far, many activities have been done in order to make optimal use of renewable energy sources such as wind, which have led to the present diverse types of wind speed and strength measuring methods in the various geographical locations. In this paper, a novel forecasting model based on hybrid neural networks (HNNs) and wavelet packet decomposition (WPD) processor has been proposed to predict wind speed. Considering this scenario, the accuracy of the proposed method is compared with other wind speed prediction methods to ensure performance improvement.
The urban book series, 2023
This work was conducted to design a combined cooling, heating, and power (CCHP) system with photo... more This work was conducted to design a combined cooling, heating, and power (CCHP) system with photovoltaic energy which provides simultaneous generation of electricity, heat, and cold for a high-rise office building (23 floors) in the city of Mashhad in Iran. Our strategy was to supply load electric, thermal, and refrigeration with the help of solar energy. In addition, its superiority over other systems was evaluated. Analysis and study of solar radiation and the maximum level of solar panels use, according to the architectural plan, were carried out at the project site. The analysis of shadow points, the use of inverters and electrical detectors to increase the maximum solar power, and its cost-effectiveness were carefully studied via PVSOL software. Additionally, the amount of heat, cold, and electricity consumption was accurately calculated according to international standards and utilizing HAP software. The criteria for saving on the initial cost reduction, carbon dioxide emission reduction, operating cost reduction, payback period, revenue, and the minimum life expectancy of the equipment compared to those in other methods were also evaluated. The results obtained from the designed system of simultaneous generation of electricity, heat, and refrigeration, which combines gas microturbines as the primary stimulus, a combination of absorption and compression chiller to provide refrigeration load, a boiler for auxiliary heat load, and a thermal photovoltaic system to produce both electric and thermal loads, were finally revealed. This is believed to be a cost-effective strategy for high-rise residential or commercial buildings with a geographical location like that of Mashhad. Based on the electricity sales to the grid, with the rate of increase in inflation in electricity tariffs, this design in the Mashhad project was estimated to have an annual income of 166.676 thousand
2023 IEEE IAS Global Conference on Emerging Technologies (GlobConET)
Optimization and Engineering
Reliability and accessibility of distribution systems are important goals that have significant i... more Reliability and accessibility of distribution systems are important goals that have significant impacts on the costs. The proper strategy of maintenance based on components arrangement and assets is the best way to reach these goals. This strategy is a kind of uses reliability-centered maintenance (RCM). Due to the limited maintenance budget, performing maintenance activities for all components of the system is neither possible nor logical. So most of the resources should be allocated to the most critical and important components. This paper presents a novel analytical method of prioritization of distribution systems’ components by introducing a new weighted cumulative Reliability-based diagnostic importance factor. This new factor includes different reliability indexes in form of diagnostic factors and will show that the order of components obtained by this method is better than another method in saving the budget and providing reliability of the system. The process of decision-mak...
DOAJ (DOAJ: Directory of Open Access Journals), Oct 24, 2021
Short-term load forecast (STLF) is an important operational function in both regulated power syst... more Short-term load forecast (STLF) is an important operational function in both regulated power systems and deregulated open electricity markets. However, STLF is not easy to handle due to the nonlinear and random-like behaviors of system loads, weather conditions, and social and economic environment variations. Despite the research work performed in the area, more accurate and robust STLF methods are still needed due to the importance and complexity of STLF. In this paper, a new neural network approach for STLF is proposed. The proposed neural network has a novel learning algorithm based on a new modified harmony search technique. This learning algorithm can widely search the solution space in various directions, and it can also avoid the overfitting problem, trapping in local minima and dead bands. Based on this learning algorithm, the suggested neural network can efficiently extract the input/output mapping function of the forecast process leading to high STLF accuracy. The proposed approach is tested on two practical power systems and the results obtained are compared with the results of several other recently published STLF methods. These comparisons confirm the validity of the developed approach.
Artificial neural networks are mathematical models that have been inspired by the human nervous s... more Artificial neural networks are mathematical models that have been inspired by the human nervous system and brain. In this study, the purpose is establishing a method to predict the cost of tractor repair and maintenance more accurately. In this paper, the multi-layer neural network with Feed Forward Backpropagation training algorithm (FFBP) has been used to predict repair and maintenance costs of tractor. In addition, 60 real data from two-wheel drive tractors existing in Razavi agro-industry in Iran have been used to train and test the network. The appropriate parameters for network training have been selected through error test on data. In this study, the performance of Backpropagate Declining Learning Rate Factor algorithm (BDLRF) has been compared with Feed-Forward Backpropagation algorithm (FFBP), with error criteria (MAPE, MSE, and RMSE) and the result shows that training Feed Forward Backpropagation algorithm (FFBP) surpasses the (BDLRF) algorithm in predicting tractor repair...
Energy Engineering Management, 2015
International Energy Journal, 2016
In recent years, rapid advances in wind energy production in many countries have made the predict... more In recent years, rapid advances in wind energy production in many countries have made the prediction of wind power very important. In addition, wind power is a complicated signal for modeling and prediction. According to previous studies in this field, wind power prediction requires an efficient method. In the current survey, a method which is a combination of two intelligent methods of Elman neural network and Particle Swarm Algorithm is proposed to predict wind power. The efficiency of the proposed prediction method is shown for predicting of wind power output of wind farms. Results of El-PSO suggested method and El-GA method were compared and evaluated by analysis of variance method (ANOVA). All the results indicate efficient performance of the proposed method (El-PSO).
Journal of AI and Data Mining, 2020
Social networks are streaming, diverse and include a wide range of edges so that continuously evo... more Social networks are streaming, diverse and include a wide range of edges so that continuously evolves over time and formed by the activities among users (such as tweets, emails, etc.), where each activity among its users, adds an edge to the network graph. Despite their popularities, the dynamicity and large size of most social networks make it difficult or impossible to study the entire network. This paper proposes a sampling algorithm that equipped with an evaluator unit for analyzing the edges and a set of simple fixed structure learning automata. Evaluator unit evaluates each edge and then decides whether edge and corresponding node should be added to the sample set. In The proposed algorithm, each main activity graph node is equipped with a simple learning automaton. The proposed algorithm is compared with the best current sampling algorithm that was reported in the Kolmogorov-Smirnov test (KS) and normalized L1 and L2 distances in real networks and synthetic networks presented...
Energy Engineering Management, 2015
2017 2nd International Conference on Computer and Communication Systems (ICCCS), 2017
In recent years due to increased competition between companies in the services sector, predict ch... more In recent years due to increased competition between companies in the services sector, predict churn customer in order to retain customers is so important. The impact of brand loyalty and customer churn in an organization as well as the difficulty of attracting a new customer per lost customer is very painful for organizations. Obtaining a predictive model customer behaviour to plan for and deal with such cases, can be very helpful. Employee churn or loss of staff will be close to the customer churn, but the impact of losing a major customer for organization certainly will be more painful (because organization do not have physical sense to losing their employees) while the consequences of finding well employees instead of missed employees, As well as the cost of in-service training that should be given to new employees could be one of the issues that each organization would be sensitive to losing its human resources.
IEEE Transactions on Power Systems, 2008
... [17] LJ Fogel, AJ Owens, and MJ Walsh, Artificial Intelligence Through Simulated Evolution. .... more ... [17] LJ Fogel, AJ Owens, and MJ Walsh, Artificial Intelligence Through Simulated Evolution. ... 2, MAY 2008 [36] CS Chang, DY Xu, and HB Quek, Pareto-optimal set based multiobjective tuning of ... [38] M. Salazar-Lechuga and JE Rowe, Particle swarm optimization and fitness ...
A Bayesian classifier is one of the most widely used classifiers which possess several properties... more A Bayesian classifier is one of the most widely used classifiers which possess several properties that make it surprisingly useful and accurate. It is illustrated that performance of Bayesian learning in some cases is comparable with neural networks and decision trees. Bayesian theorem suggests a straight forward process which is not based on search methods. This is the major point which satisfies the marvelous time complexity of Bayesian classifier. At the other hand, constructing phase of fuzzy intrusion detection systems suffer from time consuming processes which are based on search methods. In this paper we propose a novel method to accelerate such processes using Bayesian inference. Experimental results show meaningful time reduction.
Energy Conversion and Management
Scientia Iranica
The present study considers decreasing prediction error for the types of time series and the unce... more The present study considers decreasing prediction error for the types of time series and the uncertainty in estimation parameters, improving the structure of the deep neural network and increasing response speed in the proposed neural network method; besides, the competitive performance and the collaboration among the neurons of deep neural network are also increased. Selected data is related to Qeshm weather (suitable weather conditions to study our purpose) prediction during 2016 onwards. In this study, for the purpose of analyzing the prediction issue of power consumption of domestic expenses in the indefinite and severe fluctuation mode, we decided to combine two methods of Long Short-Term Memory and Convolutional Neural Network. For the training of the deep network, the BP algorithm is used. The results indicate that Gated Recurrent Unit networks compared other models (MLP, CNN and DNN) produce more realistic results, and also two-way networks obtained better results on test data compared Long Short-Term Memory networks. RMSE prediction are more realistic than the LSTM model on test and training data against the significant data. A GRU network has two gates of rt readjust and Zt forgetting, which helps to assure that long term dependencies of gradient fading will not occur.
The urban book series, 2023
Wind speed is one of the most vital, imperative meteorological parameters, thus the prediction of... more Wind speed is one of the most vital, imperative meteorological parameters, thus the prediction of which is of fundamental importance in the studies related to energy management, building construction, damages caused by strong winds, aquatic needs of power plants, the prevalence and spread of diseases, snowmelt, and air pollution. Due to the discrete and nonlinear structure of wind speed, wind speed forecasting at regular intervals is a crucial problem. In this regard, a wide variety of prediction methods have been applied. So far, many activities have been done in order to make optimal use of renewable energy sources such as wind, which have led to the present diverse types of wind speed and strength measuring methods in the various geographical locations. In this paper, a novel forecasting model based on hybrid neural networks (HNNs) and wavelet packet decomposition (WPD) processor has been proposed to predict wind speed. Considering this scenario, the accuracy of the proposed method is compared with other wind speed prediction methods to ensure performance improvement.
The urban book series, 2023
This work was conducted to design a combined cooling, heating, and power (CCHP) system with photo... more This work was conducted to design a combined cooling, heating, and power (CCHP) system with photovoltaic energy which provides simultaneous generation of electricity, heat, and cold for a high-rise office building (23 floors) in the city of Mashhad in Iran. Our strategy was to supply load electric, thermal, and refrigeration with the help of solar energy. In addition, its superiority over other systems was evaluated. Analysis and study of solar radiation and the maximum level of solar panels use, according to the architectural plan, were carried out at the project site. The analysis of shadow points, the use of inverters and electrical detectors to increase the maximum solar power, and its cost-effectiveness were carefully studied via PVSOL software. Additionally, the amount of heat, cold, and electricity consumption was accurately calculated according to international standards and utilizing HAP software. The criteria for saving on the initial cost reduction, carbon dioxide emission reduction, operating cost reduction, payback period, revenue, and the minimum life expectancy of the equipment compared to those in other methods were also evaluated. The results obtained from the designed system of simultaneous generation of electricity, heat, and refrigeration, which combines gas microturbines as the primary stimulus, a combination of absorption and compression chiller to provide refrigeration load, a boiler for auxiliary heat load, and a thermal photovoltaic system to produce both electric and thermal loads, were finally revealed. This is believed to be a cost-effective strategy for high-rise residential or commercial buildings with a geographical location like that of Mashhad. Based on the electricity sales to the grid, with the rate of increase in inflation in electricity tariffs, this design in the Mashhad project was estimated to have an annual income of 166.676 thousand
2023 IEEE IAS Global Conference on Emerging Technologies (GlobConET)
Optimization and Engineering
Reliability and accessibility of distribution systems are important goals that have significant i... more Reliability and accessibility of distribution systems are important goals that have significant impacts on the costs. The proper strategy of maintenance based on components arrangement and assets is the best way to reach these goals. This strategy is a kind of uses reliability-centered maintenance (RCM). Due to the limited maintenance budget, performing maintenance activities for all components of the system is neither possible nor logical. So most of the resources should be allocated to the most critical and important components. This paper presents a novel analytical method of prioritization of distribution systems’ components by introducing a new weighted cumulative Reliability-based diagnostic importance factor. This new factor includes different reliability indexes in form of diagnostic factors and will show that the order of components obtained by this method is better than another method in saving the budget and providing reliability of the system. The process of decision-mak...
DOAJ (DOAJ: Directory of Open Access Journals), Oct 24, 2021
Short-term load forecast (STLF) is an important operational function in both regulated power syst... more Short-term load forecast (STLF) is an important operational function in both regulated power systems and deregulated open electricity markets. However, STLF is not easy to handle due to the nonlinear and random-like behaviors of system loads, weather conditions, and social and economic environment variations. Despite the research work performed in the area, more accurate and robust STLF methods are still needed due to the importance and complexity of STLF. In this paper, a new neural network approach for STLF is proposed. The proposed neural network has a novel learning algorithm based on a new modified harmony search technique. This learning algorithm can widely search the solution space in various directions, and it can also avoid the overfitting problem, trapping in local minima and dead bands. Based on this learning algorithm, the suggested neural network can efficiently extract the input/output mapping function of the forecast process leading to high STLF accuracy. The proposed approach is tested on two practical power systems and the results obtained are compared with the results of several other recently published STLF methods. These comparisons confirm the validity of the developed approach.
Artificial neural networks are mathematical models that have been inspired by the human nervous s... more Artificial neural networks are mathematical models that have been inspired by the human nervous system and brain. In this study, the purpose is establishing a method to predict the cost of tractor repair and maintenance more accurately. In this paper, the multi-layer neural network with Feed Forward Backpropagation training algorithm (FFBP) has been used to predict repair and maintenance costs of tractor. In addition, 60 real data from two-wheel drive tractors existing in Razavi agro-industry in Iran have been used to train and test the network. The appropriate parameters for network training have been selected through error test on data. In this study, the performance of Backpropagate Declining Learning Rate Factor algorithm (BDLRF) has been compared with Feed-Forward Backpropagation algorithm (FFBP), with error criteria (MAPE, MSE, and RMSE) and the result shows that training Feed Forward Backpropagation algorithm (FFBP) surpasses the (BDLRF) algorithm in predicting tractor repair...
Energy Engineering Management, 2015
International Energy Journal, 2016
In recent years, rapid advances in wind energy production in many countries have made the predict... more In recent years, rapid advances in wind energy production in many countries have made the prediction of wind power very important. In addition, wind power is a complicated signal for modeling and prediction. According to previous studies in this field, wind power prediction requires an efficient method. In the current survey, a method which is a combination of two intelligent methods of Elman neural network and Particle Swarm Algorithm is proposed to predict wind power. The efficiency of the proposed prediction method is shown for predicting of wind power output of wind farms. Results of El-PSO suggested method and El-GA method were compared and evaluated by analysis of variance method (ANOVA). All the results indicate efficient performance of the proposed method (El-PSO).
Journal of AI and Data Mining, 2020
Social networks are streaming, diverse and include a wide range of edges so that continuously evo... more Social networks are streaming, diverse and include a wide range of edges so that continuously evolves over time and formed by the activities among users (such as tweets, emails, etc.), where each activity among its users, adds an edge to the network graph. Despite their popularities, the dynamicity and large size of most social networks make it difficult or impossible to study the entire network. This paper proposes a sampling algorithm that equipped with an evaluator unit for analyzing the edges and a set of simple fixed structure learning automata. Evaluator unit evaluates each edge and then decides whether edge and corresponding node should be added to the sample set. In The proposed algorithm, each main activity graph node is equipped with a simple learning automaton. The proposed algorithm is compared with the best current sampling algorithm that was reported in the Kolmogorov-Smirnov test (KS) and normalized L1 and L2 distances in real networks and synthetic networks presented...
Energy Engineering Management, 2015
2017 2nd International Conference on Computer and Communication Systems (ICCCS), 2017
In recent years due to increased competition between companies in the services sector, predict ch... more In recent years due to increased competition between companies in the services sector, predict churn customer in order to retain customers is so important. The impact of brand loyalty and customer churn in an organization as well as the difficulty of attracting a new customer per lost customer is very painful for organizations. Obtaining a predictive model customer behaviour to plan for and deal with such cases, can be very helpful. Employee churn or loss of staff will be close to the customer churn, but the impact of losing a major customer for organization certainly will be more painful (because organization do not have physical sense to losing their employees) while the consequences of finding well employees instead of missed employees, As well as the cost of in-service training that should be given to new employees could be one of the issues that each organization would be sensitive to losing its human resources.
IEEE Transactions on Power Systems, 2008
... [17] LJ Fogel, AJ Owens, and MJ Walsh, Artificial Intelligence Through Simulated Evolution. .... more ... [17] LJ Fogel, AJ Owens, and MJ Walsh, Artificial Intelligence Through Simulated Evolution. ... 2, MAY 2008 [36] CS Chang, DY Xu, and HB Quek, Pareto-optimal set based multiobjective tuning of ... [38] M. Salazar-Lechuga and JE Rowe, Particle swarm optimization and fitness ...
A Bayesian classifier is one of the most widely used classifiers which possess several properties... more A Bayesian classifier is one of the most widely used classifiers which possess several properties that make it surprisingly useful and accurate. It is illustrated that performance of Bayesian learning in some cases is comparable with neural networks and decision trees. Bayesian theorem suggests a straight forward process which is not based on search methods. This is the major point which satisfies the marvelous time complexity of Bayesian classifier. At the other hand, constructing phase of fuzzy intrusion detection systems suffer from time consuming processes which are based on search methods. In this paper we propose a novel method to accelerate such processes using Bayesian inference. Experimental results show meaningful time reduction.
Energy Conversion and Management