Search

This paper employs machine learning algorithms to detect tax evasion and analyzes tax data. With the development of commercial businesses, traditional algorithms are not appropriate for solving the tax evasion detection problem. Hence, other algorithms with acceptable speed, precision, analysis, and data decisions must be used. In the case of assets and tax assessment, the integration of machine learning models with meta-heuristic algorithms increases accuracy due to optimal parameters. In this paper, intelligent machine learning algorithms are used to solve tax evasion detection. This research uses an improved particle swarm optimization (IPSO) algorithm to improve the multilayer perceptron neural network by finding the optimal weight and improving support vector machine (SVM) classifiers with optimal parameters. The IPSO-MLP and IPSO-SVM models using the IPSO algorithm are used as new models for tax evasion detection. Our proposed system applies the dataset collected from the general administration of tax affairs of West Azerbaijan province of Iran with 1500 samples for the tax evasion detection problem. The evaluations show that the IPSO-MLP model has a higher accuracy rate than the IPSO-SVM model and logistic regression. Moreover, the IPSO-MLP model has higher accuracy than SVM, Naive Bayes, k-nearest neighbor, C5.0 decision tree, and AdaBoost. The accuracy of IPSO-MLP and IPSO-SVM models is 93.68% and 92.24%, respectively. [ABSTRACT FROM AUTHOR]

Automation has attracted interest from the industry sector for its potential to improve energy efficiency, cost efficiency, and environmental performance. By elevating the LoA to the highest degree, associated costs will grow accordingly and its implementation will be far more complicated. This will also result in losing workers and decreasing environmental pollutants. On the other hand, increasing power consumption at high levels of automation leads to the production of greenhouse gases. This paper aims to increase the level of automation (LoA) considering the concept of sustainability. This study presents fuzzy multi-objective programming to determine the optimal LoA considering sustainability factors to achieve competitive advantages. To solve the model, the Zimmermann max-min approach was adopted and a cosmetics factory in Iran was chosen to optimize LoA according to this model. The results showed that it is possible to improve the LoA and also consider sustainability factors with the available resources without using the highest LoA. This study can help managers optimize the LoA in their organizations considering the current resources and sustainability issues, and control the company's return on investment and cost of overhead. They can run the model with every definition of LoA proposed till now. This research can benefit the environment and the workers' health in the production line by reducing environmental pollutants and prevent the dismissal of all personnel due to its negative social effects. It also reduces the risk of COVID-19 by minimizing the number of workers. So far, a mathematical model for selecting optimal LoA in the chemical industry considering sustainability has not been presented. [ABSTRACT FROM AUTHOR]

Increase of nonlinear loads in industries has resulted in high levels of harmonic currents and consequently harmonic voltages in power networks. Harmonics have several negative effects such as higher energy losses and equipment life reduction. To reduce the levels of harmonics in power networks, different methods of harmonic suppression have been employed. The basic idea in all of these methods is to prevent harmonics from flowing into a power network at customer sides and the point of common coupling (PCC). Due to the costs, none of the existing mitigating methods result in a harmonic-free power system. The remaining harmonic currents, which rotate in a power network according to the system impedances, may not necessarily result in an optimum harmonic power flow in terms of harmonic undesirable results such as harmonic losses or harmonic over voltage/current due to possible resonances. This paper proposes a new method to control the flow of the remaining harmonics at a power system level such that the effects of harmonics such as losses are optimized. This task is achieved by the use of series active power filters controlled in a different manner from their conventional methods. In this new application, the series active power injects a controlled harmonic voltage into a power system (e.g., a line, transformer, or other elements) to control the harmonic current (power) flow such that the overall performance of the power system is improved in terms of harmonic effects. The series active power filter in this new application is named as harmonic power flow controller (HPFC). In this paper, the theory, structure, applications, and the basics of the control method of a HPFC are described. To investigate the effects of the HPFC on the power network harmonics, an HPFC is used in the 14-busbar IEEE test system. To make the proposal more practical, an HPFC is designed for Iran north-west transmission system to alleviate the harmonic problem in this region. Simulations are carried out to show the effectiveness of the HPFC. Simulation results show that the HPFC can control the harmonic currents (power) and voltage consequently. In this paper, several control algorithms for an HPFC are also considered to achieve the desired harmonic voltages levels. [ABSTRACT FROM AUTHOR]

In this paper, the optimization algorithm based on the population as improved global particle swarm optimization is described and used for inverse modelling of two-dimensional magnetic field data. This algorithm is able to estimate the parameters of depth, shape factor, amplitude coefficient, magnetic inclination angle and origin point coordinates. To evaluate the efficiency of this method, the magnetic field of an artificial model was analysed, with and without added random noise. The results suggest that the proposed algorithm is capable of model parameter estimation with high accuracy. Accordingly, the improved global particle swarm optimization algorithm was used to analyse the magnetic field of the study area in the Ileh region in Iran located in Taybad city. The study area is very rich in terms of iron resources. The estimate for the study area is that the depth of the buried mass centre is about 114.9 m and its approximate shape is similar to a horizontal cylinder based on the calculated shape factor value which is 1.76. The calculated depth is an acceptable match with the average depth of drillings. [ABSTRACT FROM AUTHOR]

This paper develops and applies algorithms for optimally merging satellite precipitation products with rain-gauge precipitation for accurate rainfall estimation. The satellite-based precipitation products (SBPs) PERSIANN-CDR, TMPA-3B42, GPM-IMERG, and GSMaP MKV are combined and evaluated to generate accurate rainfall estimates. Daily satellite precipitation data are compared with corresponding station data to calculate the bias for the period 2014–2019. Three different algorithms are proposed whose adjustable parameters are optimally determined by solving constrained optimization algorithms to produce combinations of satellite-based precipitation products. The optimal combination is named optimally merged satellite-based precipitation (OMSBP). The root mean square error (RMSE), coefficient correlation (CC), and the Nash–Sutcliffe error (NSE) are employed to test the proposed method with precipitation data for the Tehran urban region, Iran. The spatially resolved results over the studied urban area establish that TMPA-3B42, with an average value MAE, MBE, and RMSE equal to 0.68 mm, − 0.31 mm, and 2.94 mm, leads to better estimates of precipitation than those of PERSIANN-CDR, IMERG, and GSMaP MKV. Moreover, algorithms alg7 and alg8 yielded better results with respect to the MAE and MBE, respectively. Lastly, algorithm alg3 produced better results than alg7 and alg8 based on the RMSE, NSE, and CC corresponding to precipitation predictions. [ABSTRACT FROM AUTHOR]

In this paper, exergy analysis of a steam power plant located in southern Iran named Zarand power plant has been studied. In order to optimize the performance of the Rankine cycle and achieve higher exergy efficiency, several parameters have been considered as decision variables. Knowing that there is the ability to change some of the parameters in the specific range in the process of electricity production in power plant, temperature and output pressure of the boiler and output pressure of four steps of extraction turbine have been selected as six decision variables. Also, exergy efficiency has been considered as the objective function. For this purpose, the exergy efficiency of the system is optimized using intelligent algorithms including bees, fireflies, and algorithm based on teaching and learning and they are compared with each other. The results show that in the case of suitable changes of decision variables and applying appropriate algorithms, exergy efficiency of the studied thermal power plant can be increased from 30.1% to 30.68047%, 30.70368%, and 30.70369%, respectively. It means using optimization algorithms of bees, fireflies, training-learning, exergy efficiency of the Rankine cycle of the studied power plant can be increased by 0.58047%, 0.60368%, and 0.60369%, respectively. [ABSTRACT FROM AUTHOR]

In this paper, a new thermodynamic cycle for power and fresh water production is presented, which includes MED-TVC desalinations and Allam cycles due to the important influence they have on the electric power industry in controlling and limiting greenhouse emissions. First, the Allam cycle analyzed and the appropriate location for heat-extraction identified to produce the required water vapor for the desalination system. It was determined that the heat site suitable for use in the desalination cycle is the heat from the outlet of the Compressor, which is estimated as 100 MW. The MED-TVC Desalination, one of the most suitable and most economical desalination, is used to combine the Allam cycle and desalination in the Kish district, located in the south of Iran. The proposed cycle is analyzed from the perspective of energy and exergy. The results show, the highest amount of exergy used to generate power in turbine and the amount of produced freshwater increase with increasing the capacity of Turbine. For this purpose genetic algorithm is used in two different scenarios to minimize and maximize the exergy destruction and produced fresh water respectively. The optimization of the system with genetic algorithm led to 18% decrease in total exergy destruction of the cycle in the first scenario and 7% decrease in the second optimization scenario. Furthermore, the efficiency of the cycle in first scenario and second scenario increased by 30% and 13% respectively. In the scenario optimization scheme, the amount of fresh water increased by 22%. [ABSTRACT FROM AUTHOR]

The special features of some old residential complexes make reconstructing a better option than rebuilding. The complexity of redesign issues, the ability of computers to solve problems, and the capability of evolutionary optimization techniques in architecture make the use of these methods a priority over traditional processes. This paper, while introducing Particle Swarm Optimization (PSO) as one of the swarm intelligence based algorithms, answers the question of how this algorithm can be used to redesign residential complexes. To achieve this, after reviewing the application of PSO algorithms in architectural space planning, the redesign criteria and constraints was divided into four categories; including primary design specifications, contact-related criteria, designer-related criteria and redesign-related constraints. After the quantification process of these criteria, how to implement the algorithm to achieve optimal plans for each of the five plan types available in high-rise apartment complex in Mashhad, Iran, was shown as a case study. [ABSTRACT FROM AUTHOR]

In recent years, research has shown a growing interest in the use of hybrid wind photovoltaic (PV) systems that provide better performance compared to the use of a single component due to complementarity in meeting electricity demand. Over the past twenty-five years, hundreds of articles have addressed the topic of hybrid systems considering different configurations and final uses and, over the past decades, many reviews have made a comprehensive summary of various results obtained. However, some reviews deal with the research in a too general and qualitative way, without providing quantitative data, and other reviews are too focused on a specific topic aspect. To provide a qualitative-quantitative prospect of the research trend in the last twenty-five years, the present work is aimed at carrying out a literature review and statistical analysis starting from data extracted from the 550 most relevant and recent articles concerning hybrid systems, published between 1995 and 2020. The review aim was to produce an upgradable matrix literature database that schematizes the content of all articles in terms of different categories, such as the geographical distribution, their component configurations, operating mode and the auxiliary components used to support it, their intended uses and study methodologies (simulation, experimental, economic, energy, environmental and social analysis etc.) and software used. In addition, all the optimization algorithm, energy, economic, environmental and social indicators available in the literature were extracted and elaborated to identify the most used. The 550 articles were analysed, compared, and classified into several categories to provide an overall framework of the state of the art. The objective is to clearly and appropriately show important trends and findings in the development of hybrid wind and solar PV experimental, simulation and optimization projects. Data are elaborated to obtain a statistical analysis for each category or a combination of categories. In particular, the analysis highlighted that research is more focused on testing systems in warm or temperate localities, with the Köppen climate groups B and C prevalent over the others. From the geographical point of view, Asia is the continent most involved in world research (with China, India and Iran the first three countries for total publications produced). However, also in other parts of the world, a growing interest was noticed in this technology. The prevalent tested system configuration mode is the stand-alone hybrid systems, in a wide variety of climates and especially for residential uses. Simulations are mostly implemented in the analysed publications, mainly through HOMER and MATLAB software. Parametric analysis is widely used for optimal system design with a large variety of techniques. In particular, the system performance is examined mostly from an energy point of view. Economic analysis is also very common, alone or in combination with energy analysis. The most frequently used optimization algorithms are the particle swarm optimization (PSO) and genetic algorithm (GA), while the loss of power supply probability (LPSP) and renewable fraction (RE) for the energy analysis, the net present cost (NPC) and cost of energy (COE) for the economic analysis and the emissions (E) of CO 2 for the environmental analysis are most widespread indicators. Finally, an analysis on the size of the system components is performed to study which renewable source is more preferred at low and high installed power, for stand-alone, grid-connected systems and overall, considering different intended uses. The analysis highlights that PV systems are preferred at low installed powers, especially for residential use and stand-alone mode, while wind systems, in addition to being extensively used for low installed powers, demonstrates higher employment compared to PV systems as the power increases. The paper findings and upgradable matrix literature database are proposed as a valuable tool for engineers, experts and national and international policymakers. • 550 relevant papers were reviewed and used to create an upgradable matrix literature database. • Asia is the continent most involved in world research with Iran, India, and China. • The dry and temperate climate zones of Köppen climate group BWh are the most investigated. • Stand-alone hybrid systems and residential use are the most explored applications. • The use of photovoltaic system is preferred for low installed powers, while wind systems also for high installed powers. [ABSTRACT FROM AUTHOR]

This paper addresses a possible decrease in flood damages by using structural methods and determines proper locations (layout) to construct protective levees and height of levees (design of levees) in high-risk areas. Because of discontinuity of the objective function and a large number of decision variables, the optimization uses the genetic algorithm (GA). The simulation of the region is carried out with the Hydrologic Engineering Center's River Analysis System (HEC-RAS) model. The GA is employed to maximize the benefit of flood control and also to minimize the cost of protective levees' construction. Thus, the fitness function of the algorithm maximizes the net benefit of the project. After a sensitivity analysis, mutation and crossover probability are assumed to be 0.03 and 0.7, respectively. Thus, 1,000 generations and 98 chromosomes for each run are considered. The optimization model is run 5 times for floods with return periods of 50 and 100 years. Furthermore, the algorithm is used to evaluate flood damages in the Sarm and Khoor Abad River, located in the Qom province of Iran. Project net benefits resulting from the optimization model for rivers using 100-year flood exceed those for a 50-year flood. Results indicate that construction of protective levees reduces the rate of damages up to 99% in comparison with a nonconstruction of levees scenario. By taking into account the construction cost of levees, net benefits decrease by 0.97% and 1.1% for floods with 50-year and 100-year return periods, respectively. [ABSTRACT FROM AUTHOR]

This paper brings together the benefits of a membrane-thermal desalination plant with an integrated system containing parabolic trough collectors and wind turbines in collaboration with the electrical network to produce power and freshwater. This proposed cogeneration system was employed for providing power and freshwater in Chabahar in Iran by three types of desalination system consist of the Reverse Osmosis (RO), Multi-effect distillation (MED), and Thermal Vapor Compression (TVC). Through this novel system, the heat generated in the parabolic trough collector is transferred into the organic Rankine cycle (ORC) to produce power. So, the generated power of the organic Rankine cycle and an auxiliary wind turbine can employ by the Reverse Osmosis unit, and the surplus amount of the electricity can sell to the electrical network. This configuration is enabled to increase the stability of the system, and it ultimately reduces the cost of water production by 23%. Besides, the integration of Multi-effect distillation (MED) and Thermal Vapor Compression (TVC) systems enjoys the rejected waste heat from the condenser to increase the produced freshwater. The obtained results from exergy analysis demonstrated that the exergy destruction of the solar collectors and wind turbines contributed by 39.5% and 22.2%, respectively. The results of multi-objective particle swarm optimization reveal that the exergy efficiency and the cost of freshwater production reach to 26.2% and 3.08 US$/m3, each. The environmental assessments of this system demonstrate that this hybrid system prevents the amount of 52164 tons of CO 2 emission per year. [ABSTRACT FROM AUTHOR]

Regular optimization techniques have been widely used in landslide-related problems. This paper outlines two novel optimizations of artificial neural network (ANN) using grey wolf optimization (GWO) and biogeography-based optimization (BBO) metaheuristic algorithms in the Ardabil province, Iran. To this end, these algorithms are synthesized with a multi-layer perceptron (MLP) neural network for optimizing its computational parameters. The used spatial database consists of fourteen landslide conditioning factors, namely elevation, slope aspect, land use, plan curvature, profile curvature, soil type, distance to river, distance to road, distance to fault, rainfall, slope degree, stream power index (SPI), topographic wetness index (TWI) and lithology. 70% of the identified landslides are randomly selected to train the proposed models and the remaining 30% is used to evaluate the accuracy of them. Also, the frequency ratio theory is used to analyze the spatial interaction between the landslide and conditioning factors. Obtained values of area under the receiver operating characteristic curve, as well as mean square error and mean absolute error showed that both GWO and BBO hybrid algorithms could efficiently improve the learning capability of the MLP. Besides, the BBO-based ensemble surpasses other implemented models. [ABSTRACT FROM AUTHOR]