particle swarm algorithm

Climate change causes changes in the flow of rivers by causing changes in temperature and precipitation. Therefore, river flow simulation is important as a prerequisite for some environmental and engineering issues. In the current research, the effect of climate change on the Mahabad’s river flow in the future periods (2045-2026) was predicted using machine learning models. First, two input scenarios were compiled, in which the first scenario included temperature and precipitation parameters and the second scenario included temperature, precipitation, and flow parameters one month ago. In the following, the performance of two ANN and ANN-PSO models in estimating the flow rate in the base period (1992-2014) was compared to select the best scenario and the best model for predicting the flow in the future period under the three scenarios SSP1.26, SSP2.45 and SSP5.85 of the CMIP6. The results of the error evaluation criteria showed that the ANN-PSO model makes the best estimation of the river flow using the second scenario and with the criteria (NSE=0.77, RMSE=6.4 MCM, MAE=3.4 MCM for the test data) and it was chosen to predict the flow in the future period (2026-2045). The results of investigating the effect of climate change on each of the meteorological parameters showed that climate change causes an increase in temperature and creates a fluctuating pattern in precipitation. The results of the climate change survey on flow showed that under the SSP1.26 scenario, there will not be much changes in flow in almost months, but in the SSP2.45 and SSP585 scenarios, there will be an increase in the discharge in December, and in May and April, the greatest decrease in discharge will be (16.50 MCM) and (13.33 MCM) respectively.

SWAT model is a suitable tool for simulating hydrological processes. This model requires many inputs that often cannot be measured directly and is considered one of the main sources of uncertainty in these models. The recalibration process can reduce the uncertainty in the model results by adjusting and adapting these inputs. The researches showed that calibrating a hydrological model by using the common automatic CV calibrating algorithms will not provide proper accuracy in the prediction of hydrological variables during the validation period, so PSO algorithm was used to calibrate the SWAT model. Since there is no mathematical and logical rule to determine the best combination of PSO algorithm parameters and these combinations are selected based on trial and error and among many different combinations, therefore trial and error based methods are very time-consuming and sometimes impossible. In this research, Taguchi method was used to determine the best combination of PSO algorithm parameters.

In this research, the ability to use the SWAT model to simulate monthly runoff in the Javanmardi Watershed, one of the main sub-basins of the Lordegan Watershed with an area of 380 square kilometers, was investigated. In this study, the PSO algorithm parameters, including the number of simulations (A), the number of repetitions (B), the speed calculation weight (C) and the movement parameter (D), were defined in four levels. Then, these parameters were designed and implemented according to the experiments in the L16 orthogonal array (using the Taguchi experiments design method). The performance scale used to evaluate the algorithms was RPD (Relative Percentage Deviation). Considering the variable nature of the response in this study, the S/N index "the lower the better" was used to analyze the Taguchi test results. The selection of arrays and calculations were done in Minitab 16 software.

In the sensitivity analysis stage, which was performed before the model recalibration, among the 28 parameters studied in this research, the model showed sensitivity to the changes of 22 parameters, and they were identified as variables influencing the simulation of runoff in Javanmardi Watershed. The results showed that the parameter of the runoff curve number (CN) is the most important factor and the parameters of soil apparent density in the wet state (SOL_BD) and average water usable by the plant (SOL_AWC) are among the most important factors controlling the flow rate in the study basin, respectively. Based on the results simulated by the PSO algorithm, it was found that the SWAT model has an acceptable accuracy for estimating the monthly runoff in the study area. So, in the recalibration phase, the r-factor and p-factor indices were 1.23 and 0.88, respectively, and the explanatory and Nash-Sutcliffe coefficients were 0.77 and 0.75, respectively. In the validation stage, the r-factor and p-factor indexes were 1.31 and 0.84, respectively and the explanatory and Nash-Sutcliffe coefficients were 0.72 and 0.73, respectively. In this study, the best combination resulting from the application of Taguchi method for the parameters of the number of simulations, the number of repetitions, the speed calculation weight and the appropriate parameters in the PSO algorithm were determined as 40, 100, 0.2 and 0.15 respectively (A4B4C4D3).

The results show that the SWAT model has an acceptable accuracy for estimating the monthly runoff in the Jawanmardi Watershed, and the PSO method is an effective algorithm in calibrating and determining the uncertainty of the model in this basin, and the use of the Taguchi test design method is a suitable way to determine the best combination of PSO algorithm parameters is for researchers who use this method to optimize the SWAT model.

 Ultrafiltration is one of the most common membrane processes in the dairy industry, especially for condensing and separating milk components. Using this process, several products can be produced, including milk concentrate used for cheese production, low-lactose dairy products, milk protein concentrate, and serum proteins for dietary supplements. The efficiency and cost of a membrane process depend on the percentage of rejection of the soluble components. Therefore, the use of concentrated milk made by ultrafiltration in the production of various dairy products depends on the efficiency of the membrane process and the changes in milk components during this process. On the one hand, the physicochemical properties of camel milk are different from those of cow milk, especially in terms of type and amount of protein. Because significant differences exist between the physicochemical properties of camel and cow milk, likely, the membrane processing conditions and the physicochemical properties of their products will be different completely. Although many studies have been conducted on the efficacy of the ultrafiltration processing of cow milk, there is no information about the efficacy of camel milk ultrafiltration, and most of the research done regarding optimizing is based on classical algorithms, Therefore, in this study, the effects of transmembrane pressure and temperature on the solutes rejection (protein, lactose, ash, and total solids) during camel milk ultrafiltration process were investigated, Then, these properties were optimized using particle swarm algorithm. Also, because the performance of the particle swarm algorithm is highly dependent on related parameters such as the number of iterations, the number of particles, accelerate constant, inertia weight, and velocity of the particles, so before optimization, the effect of these parameters on optimal responses were examined by partial least squares regression (PLS).

 In this study, a pilot crossflow ultrafiltration system was used. A UF membrane (Model 3838 HFK-131, Koch membrane systems, Inc., USA) made of polysulfone amid (PSA) with MWCO of 20 kDa was applied. Camel milk was purchased from a local market in Mashhad and for camel skim milk production, its fat was separated by a pilot plant milk fat separator in the Food Research Complex, Ferdowsi University of Mashhad. The weight percentages of protein, fat, lactose, ash, and total solids of UF permeate samples were measured by ISO 8968-1:2014, ISO 1211: 2010, ISO 26462/IDF 214:2010, ISO 5544:2008, and ISO 6731:2010 at two replications, respectively. the process treatments were performed in the form of a central composite design (CCD) (5 replications at the central point) for two independent variables at three levels so that the total number of 13 treatments was obtained. The data were modeled using the statistical software of Design Expert (version 11) based on the response surface methodology and each of the response variables in the form of a regression model was presented as a function of independent variables.

 The rejection of total solids and protein of the tested samples varied in the range of 45.4-51.03% and 94.09-97.51%, respectively. It means that in each TMP and T, more than 45% of the total solids and 94% of the protein of camel milk were kept by the membrane. The results also showed that none of the linear, quadratic and interactive effects of TMP and T on the total solids and protein rejections were not significant. According to the results, the RL reduced with increasing T. Increasing the TMP also led to a reduction at high T and an increase in RL rate of the samples at lover T. Also, the effect of TMP on RA showed a non-linear trend, so that TMP at high T led to an increase, and at low T, it led to a reduction in the RA of the samples.

 The optimization results with the particle swarm algorithm showed that this algorithm has a high convergence speed and by recognizing and analyzing its parameters, the optimal conditions can be easily found. The optimum ultrafiltration conditions in this study with the lowest RL and RA were determined as 80 kPa TMP and 29.85 ͦ C T.

Industrial and medicinal plants are one of the by-products of rangelands, the exploitation of which has a role in the economy of rangelands. The preservation and development of distribution of medicinal plants at the level of rangelands is important from various economic, social and technical aspects of rangelands. There are many industrial and medicinal plant species in the rangeland. Anchusa strigosa is a medicinal and indigenous species of Dezful. One of the important issues in the principled exploitation of these medicinal plants is to determine, its habitat areas. The purpose of this research is to investigate the relationship between the biophysical properties of the potential habitat of this species using different data mining techniques. Evaluating the effectiveness of data mining models, first 30 presence points and 30 absence points of the species were recorded using vegetation map and field studies. Then the parameters of height, slope, temperature, rainfall, texture and depth of soil were extracted from the obtained maps using geostatistical methods based on regional data. Determination of the influence coefficient of each of the mentioned parameters was done using the vector machine method. Then, the presence of the mentioned species was predicted based on biophysical parameters using data mining methods. The data mining algorithms used in this study include particle swarm, logistic regression, artificial neural networks and the nearest neighbor.The results showed that particle swarm algorithm and artificial neural network with accuracy rate of 0.95 and kappa coefficient of 0.9 have a higher accuracy in determining the species presence areas.The results of this study showed that using data mining methods, it is possible to predict the presence of the desired species, based on biophysical parameters with high accuracy.

Cropping pattern optimization is an important strategy in the optimal use of water in agriculture. The present study attempted to develop and evaluate four hybrid models based on maximizing economic return rate and minimizing risk based on risk indices including variance (MVar model), semi-variance (MSVar model), absolute value deviation from mean (MADev model) and conditional value at risk (MCOVaR model). The MCOVaR model is governed by observing the rules, which at a selective confidence level allows determining the best cropping pattern scenario with selective return rate and the lowest risk that can be measured in the decision-making stage. The data used are related to the downstream irrigation network of Ardabil Yamchi Dam and for the 1397-98 crop year. The development of the model in MATLAB is based on nonlinear programming and optimization problem solving using particle swarm algorithm in a multi-objective solution. The highest correlation coefficient in regression of the models with Fourier equation was R2=0.9997 with accuracy (RMSE=0.0011) for MADev model, The results showed that the highest yield of 0.31 belonged to potatoes with a maximum risk of 58% and the lowest yield of 0.13 belonged to the maize plant with a maximum risk of 63%. By calculating 50 scenarios of the best possible cropping patterns in terms of satisfying the target functions and governing conditions, the pareto front of each model was drawn and the table of system income values for risk levels of 20% and 30% was extracted and presented. The results obtained in all models indicate an increase in economic efficiency by increasing the risk level of the system in selecting the optimal cropping pattern and this increase in MADev model is more steep than other models and increasing risk-taking leads to increasing the area under cultivation of plants with more water consumption.

Monitoring network optimization is a decision making process for the best combination of available stations. Due to economic considerations and reduction of monitoring costs ، the optimization approach in this study is to reduce monitoring stations without reducing the amount and accuracy of the information obtained. In this study, an optimal design of groundwater quality monitoring network was carried out with the help of an optimization model in the Neishabour plain aquifer. The optimization of the wells network was accomplished by a Multi Objective Particle Swarm Optimization (MOPSO) algorithm. Two objectives containing of minimizing the root mean square error (RMSE) and the number of wells was applied in current research. Kriging interpolation was used for calculating groundwater chlorine concentration values and compared with observation values. As a result of this research was presented a Pareto front exctracted from MOPSO showing the number of wells against their corresponding RMSE, which could be a guide for the design of a groundwater quality monitoring network. The outcome showed that the sampling wells can be reduced to 58 percent with a minimum error increase (all 50 wells in base network with zero error may be reduced to 21 with chlorine concentration error of 13.57 mg/l) in the Neishabour aquifer. Also, the position of these wells was considered as the optimal position.

Due to the limited surface water resources and reservoir capacity of dams, scientific management and optimal utilization policy of reservoirs is essential and vital in meeting the water requirements. Considering the importance and utilization of water resources and the optimal allocation of this scarce resource among different uses in the Sistan region, it seems necessary to establish a plan to achieve this goal. In the studied area and in the country, so far, an investigation into the optimal allocation of water with the application of advanced PSO algorithm for optimization of semi-Sistan reservoirs has not been used. Therefore, in the present study, the optimal allocation of water resources in the Chah-Nimeh reservoirs under three management scenarios (scenario of stabilization of micro-organisms, agricultural development and transfer of second drinking water from reservoirs to the district of Zahedan) using the metamorphic technique of congestion Particles (PSO) have been investigated. particle swarm optimization method or abbreviated mass optimization (PSO) is a population based stochastic optimization algorithm that simulates the social behavior of birds Group is inspired. In the mass optimization algorithm of particles, there are organisms that they are called particle and are spilled in the search space of a function that is intended to minimize (or optimize) its value. Each particle calculates the value of the target function in the position of the space in which it is located. In this research, the algorithm was used to allocate optimal water resources. The objective function in this research is to optimize and maximize the amount of water supply. Also, the constraints are related to the objective function, the systematic constraints, the constraints and limitations of the algorithm and the constraints and limitations of the reservoirs in the region. According to the results, the optimum release rate in 1995 (first year) was 39.35 million cubic meters, with the demand of 98.31 million cubic meters, which did not meet the required amount of 72.91 million cubic meters. The comparison of the last four years shows that in the year 29, lack of supply was less than the next three years. The scenario of stabilization of the microstats in the study area was considered as a serious project. The results of this scenario showed that the application of the algorithm used can properly optimize the allocation of water resources. According to the results, it is suggested that ultra-modern modeling can achieve more optimal allocation with the minimum error in the target function.

Monitoring network optimization in water resources, a decision-making process is available for having the best combination in between stations. The study of particle swarm algorithm or PSO algorithm were used for determination of the location and number of network observation wells. The first using the least squares support vector machine model and input parameters coordinates, evaporation, precipitation last two months, one month before groundwater table with RBF kernel function was simulated groundwater table. Then the link LSSVM models and PSO model proper location well under two scenarios were determined. In the first scenario was to determine the location of fixed piezometers 42. In the second scenario was considered as a variable number of piezometers. The results of this study showed, Given that our objective function is to minimize the difference between the observed and the simulated, in the first scenario is the least amount of difference in repeat 280 with the objective function 0.9865. The results of the second scenario shows that the number was 28 piezometers Which represents a decrease of 55 percent compared to the initial state is the number of piezometers. In both scenarios, the distribution of points in the southern parts has increased due to the increase in the hydraulic slope and has decreased in the northern parts. In this scenario, the lowest error was repeated 338 with the aim of 0.9145. This optimization at various stages with several stages of trial and error, show the degree of importance and superiority of the latter scenario than the first scenario.

Exploitation of dam reservoirs is one of the major problems in the management of water resources. In this research, Crow Search Algorithm (CSA) was used for the first time to manage the operation of reservoirs. Also, the results related to the exploitation of the single-reservoir system of Shahid-Rajaei dam, located in Mazandaran province, northern Iran, which meets the downstream water demands, were compared to those obtained by applying the Particle Swarm and Genetic algorithms. Time reliability, volume reliability, vulnerability and reversibility indices, and a multi-criteria decision-making model were used to select the best algorithm. The results showed that the CSA obtained results close to the problem’s absolute optimal response, such that the average responses in the Crow, Particle Swarm and Genetic Algorithms were 99, 75 and 61 percent of the absolute optimal response, respectively. Besides, except for the time reliability index, the CSA had a better performance in the rest of the indices, as compared to Particle Swarm and Genetic Algorithms. The coefficient of variation of the obtained responses by CSA was 14 and 16 times smaller than the Genetic and Particle Swarm Algorithms, respectively. The multi-criteria decision-making model revealed that the CSA was ranked first, as compared to the other two algorithms, in the Shahid-Rajaei Reservoir's operation problem.

Groundwater and water resource management play key roles in water resource sustainability in arid and semi-arid areas. Forecasting groundwater level is very important for water resource management and planning. In this study, an artificial neural network and a particle swarm algorithm based on artificial neural network models have been used to estimate groundwater level in the Ardebil plain. Water table level data for the 1972 -2011 period was used as our data in this study. Model inputs were water table level of various months. Results of both models were evaluated by root-mean-square error, the correlation coefficient and Nash-Sutcliffe coefficient. Results showed the performance of the particle swarm algorithm based on artificial neural network models to be superior. Root-mean-square error results for the particle swarm algorithm model in spring, summer, autumn and winter were 0.476, 0.507, 0.309, and 0.386 respectively. These results show that the hybrid structure of the network in training leads to increased accuracy. Thus, the particle swarm algorithm based on artificial neural network models can be used to estimate groundwater level in the Ardebil plain with acceptable accuracy.