جستجوی مقالات مرتبط با کلیدواژه « پرسپترون چند لایه » در نشریات گروه « کشاورزی »

This study was conducted to evaluate the relationship between leaf iron and some easily-available soil properties in citrus orchards in the southern region of Kerman province by artificial neural network modeling and stepwise regression. For this purpose, 40 orchards were selected from the study area and the physical and chemical properties of soil and iron in the plant leaves were measured. Using artificial neural network in different models with different data from soil properties as input and leaf iron as output, the ability of these models to predict leaf iron concentration was evaluated. The results showed artificial neural network with variables of organic carbon, pH, clay, phosphorus, TNV and electrical conductivity with explanation coefficient of 0.86 and 0.81 and root mean square error (RMSE) of 14.60 and 20.13 mg.kg-1 for data Training and testing were the best models in estimating leaf iron. Comparison of regression and neural network models in the test data showed that the neural network had a higher accuracy with an explanation coefficient of 0.81 than stepwise regression with an explanation coefficient of 0.2. The amount of RMSE in the neural network also improved and increased from 27.72 mg.kg-1 in the stepwise regression model to 20.13 mg.kg-1 in the neural network. Artificial neural networks have been able to predict the iron in plant leaves based on the easily-available properties of the soil, so that by choosing organic carbon as the input of the first model to the best model by selecting organic carbon, pH, clay, phosphorus, TNV and electrical conductivity, model accuracy increased.

Given the importance of measuring tree volume and the potential for selling standing trees, as well as the time-consuming and costly nature of measuring the actual volume of trees, this study aimed to find an alternative method that is more economical in terms of time and cost for predicting the standing volume of poplars (Populus deltoids) using an artificial neural network (ANN) method for harvesting and sales management. In this study, the diameters and heights of 416 randomly selected trees were measured. The tree diameters ranged from 10 to 15, 15 to 20, 20 to 25, 25 to 30, 30 to 35, 35 to 40, 40 to 45, and 45 to 50 centimeters. Their standing volumes were then predicted using both simple and multiple ANN models. In this study, diameter at breast height and height were considered predictor indices for the input data, while the standing volume of the tree was the output data. The number of input, hidden, and output layers was kept uniform at one layer. The number of hidden layer neurons was determined to be 10 using trial and error. The results showed that the simple ANN model using the diameter index yielded MAPE and R-squared values of 10.22 and 0.9785, respectively, while the model using the height index produced MAPE and R-squared values of 35.43 and 0.8004, respectively. Due to the simple model's ability to predict volume with an error of approximately 10% using the diameter predictor, the simple model with the diameter index was suggested as the best model overall, considering its ease of implementation and superior accuracy.

Different forms of land use development and human activities in protected areas are considered to be the main drivers of change, which have many effects on habitats, habitats, diversity and richness of species. The purpose of this research is to model the effect of human activities on the diversity of vegetation using the artificial neural network method and determine the impact of ecological and human variables on them. This research was done in the Central Alborz protected area under the management of Alborz Province. To achieve the mentioned purpose, firstly, 101 plots and 101 soil samples were collected and, soil and vegetation analysis were performed on the samples. Finally, using the multilayer perceptron neural network method and using 18 input variables including physical and chemical variables of the soil, physiographic variables, and human factors variables , the effect of human activities on the diversity of vegetation in the study area modeled. According to the results, the vegetation diversity model with the structure of 1-5-18 according to the highest value of the coefficients of determination in the three categories of training, validation, and test data is equal to 0.82. 0.81 and 0.68 show the best structure optimization performance, distance from roads, electrical conductivity, and percentage of organic matter in the soil show the greatest effect on the diversity of vegetation in the study area. The model presented in this research is used as a decision support system in evaluating the effects of human activities on the diversity of vegetation in protected areas and provides the possibility of predicting the extent of these effects on the diversity of vegetation in these areas.

In order to manage the forests and optimal and sustainable utilization of the forest, it seems necessary to know the information on the volume of the residual stand. In this study, a systematic randomized inventory was carried out in 186 circular 10-acre plots in the educational and research forest of Darabkola, Sari, Golestan, Iran and the volume of each plot was obtained. In the next step, the physiographic layers of the area were prepared using the topographic map and the vegetation characteristics were prepared using the LISS-III image of the IRS-P6 satellite with a resolution of 23.5. After preparing physiographic layers and vegetation characteristics, their value was calculated for all plots. Then, regarding these variables and using two methods of multilayer perceptron neural network and multiple regression model, modeling was done. The results showed that the multiple linear model could model the volume changes in the region with higher accuracy (R2=0.75 and RMSE=0.3). The results of this research can be used in management planning and as one of the effective factors in the design of logging routes and forest roads so that areas with larger volumes are covered more.

Saffron (Crocus sativus L.) is one of the most expensive crop which is grown in restricted areas of the world. Due to its economic values, some farmers, based on similarities of climatic conditions have cultivated it in some regions of country regardless of land capability and suitability, which sometimes the result was not satisfactory. Saffron yield prediction based on soil properties enables us to assess the land suitably for cultivation of this valuable plant. For this purpose, 100 soil samples were collected from Vamenan Saffron fields in Golestan province and the soil chemical and physical properties including the percentage of constituents of the mineral part of soil texture (Sand, Silt, Clay), Phosphorus, potassium, Nitrogen, pH, Electrical Conductivity (EC), Organic matter and Calcium Carbonate Equivalent were measured. In addition, the weight of Saffron wet flower (kg.Ha-1) was measured. In the present study, various combinations of soil properties as input were applied and nine models were developed using artificial neural networks and multiple linear regression models for predicting the saffron yield. Performance of the models was validated using Root Mean Square Error (RMSE), Correlation Coefficient (R) and Geometric Mean of Error Ratio (GMER) methods. The results of the correlation analyses showed phosphorus and organic matter are most effective factors in the production of Saffron. Results showed that performance of the models is much different where R2 value varies from 0.45 to 0.89. Comparing the performance of Saffron yield estimation models indicated the optimal model was obtained from the combination of phosphorous, organic matter, potassium and calcium carbonate equivalent as input and values of R2 and RMSE equal to 0.874 and 0.996 kg.ha-1, respectively.Evaluation of model results indicated that the coefficient varied was obtained from 0.45 to 0.89. The best model in saffron yield estimation was obtained when phosphorous, organic matter, potassium and electrical conductivity were as the input, so that values of R2 and root mean square error (RMSE) were obtained 0.891 and 0.89 kg.ha-1, respectively.

Short-term prediction of minimum temperature is important in mitigation chilling and frost injury in agriculture. In current study, the frequency of early autumn and late spring frosts in Sanandaj synoptic station, Iran were worked out. Then, using six variables, i.e. dry and wet bulb temperature, relative humidity, wind speed, wind direction, and cloud cover as the inputs, a multilayer perceptron artificial neural network model (MPL/ANN) based on the Levenberg-Marquardt training algorithm of MATLAB software package was applied for prediction of the minimum temperature for the next 3, 6, 9 and 12 hours ahead. The selected Transfer Functions were hardlims, logsig, polsin, radbas, satlins, softmax, tansig, and tribas. The statistical measures of MAD, MSD, RMSD, and R were used for comparisons. The results showed that in case of late spring frost, the poslin, logsig, tansig, and satlin functions in April with a correlation coefficient greater than 0.8 and error values of 1.17, 1.61, 1.88 and 2.00 (◦C) for the different times steps, respectively are the best options. Similarly, in October, the radbas, poslin, poslin, and tribas functions with a correlation more than 0.7 and error values of 1.60, 1.96, 1.99, and 1.36, were found to be the most suitable ones for prediction of the minimum temperature at 21:30, 00:30, 03:30 and 06:30 local time. Also, among the selected functions, the poslin with the highest frequency has the best performance in predicting nocturnal frosts in Sanandaj. The results confirmed the good performance of the ANN approach in short-term prediction of minimum temperature and frost occurrence in study region.

Over the past two decades, degradation damage has increased in natural ecosystems. This research focuses solely on the degradation of soil and vegetation cover in Sorkheh Eshpar National Park. For this purpose, some physical properties of the soil (specific gravity and true porosity, moisture content, a percentage of pebbles, acidity and conductivity and soil texture) were measured. Also, the diversity and density of the crop in two safe zones and other uses were calculated. Soil sampling was carried out in the form of strip transects at the beginning and the end of each joint unit and excavation of four soil profiles at a depth of 5-5 cm. The crown cover was taken with 2 × 2 squares throughout each transect and 30 plots per unit. A total of 600 plots and 480 soil profiles were taken from the study area. Shannonchr('39')s biodiversity index was selected as an output variable and soil characteristics, ecological factors, and human factors as inputs of the model. Normality of data was confirmed by Shapiro-Wilk test. SPSS software was used for soil degradation regression and vegetation cover. The combined, removable, and retrograde method was chosen with a value of = 422.0. With the MATLAB software, the artificial neural network was created with a multi-layered perceptron, with a secret layer, and 18 neurons, and a sigmoid tangent function, and a value of R = 0.5065. Considering the amount of R in the artificial neural network model, this model was selected as the optimum model for estimating and predicting soil degradation and vegetation cover in Sorkho-e Hesar National Park.

The growing demand for agricultural products and the reduction of access to conventional water resources have highlighted the need for proper management of water resources in agricultural sector. On the other hand, measuring and acquiring field data problems make it necessary to employ efficient models that can accurately predict yield and sensitivity analysis of yield over various parameters. Therefore, in this research, the effects of different parameters on the yield of forage maize in the country have been investigated and the effectiveness of the artificial neural network models to predict forage maize yield has been evaluated. The parameters of irrigation water salinity, soil salinity, plant growth period, water consumption and number of irrigation of 104 farms in 8 provinces that were collected during field studies were used to develop models in the artificial neural network. Also, by estimating the water requirements of selected farms, the expected yield was estimated using the optimal selected model and compared with the measured yield in the farms. The results showed that the artificial neural network model with two hidden layers and the final structure of 1-15-19-5 can accurately estimate forage maize yield (R2=0.85). The results of the sensitivity analysis of the selected optimal model showed that the parameters of irrigation events and plant growth period were the most effective and least effective parameters on the forage maize yield, respectively. Also, the results of the modeling showed that there is a gap between the measured yield in the farms and the expected yield, and if the full irrigation requirement is applied, the forage maize yield in the studied areas can be increased by an average of 12.5 tons per hectare.

Landslide susceptibility and its risk assessment is the main part of landslide risk mapping. In this study, landslide susceptibility of Oliya's Padena in Semirom is mapped using artificial neural network. A total of 23 factors in relation to landslide in the region were initially characterized. The spatial location of landslide events was then determined by field study as well as aerial photo analysis. AHP analysis tends to 14 out of 23 parameters as the important factors for further steps. A total of 72 (70%) and 31 (30%) out of 103 detected landslide events in the study area were selected as training and validation data for neural network analysis, respectively. A multilayer perceptron back propagation algorithm with sigmoid as activation function was developed. The best topology was determined by using conventional criteria including mean square error, root mean square error, maximum absolute error and correlation coefficient. Results show that a 14-4-1 array is the optimum topology for landslide susceptibility zoning in the region. The weight of each input layer was estimated by frequency ratio. In order to map landslide, ROC graph and area under curve indices were used and the accuracy of output map was computed. Results from validation shows that area under curve for the obtained model is about 0.938 (93.8%) that is considered as high resolution prediction group. According to this study, a total of 29.61 square kilometers (93.25%) of the landslide areas is categorized in very high and high susceptible groups.

Flood is a kind of natural disaster which causes financial damages and fatality for people. Every year, especially in areas like Maroon river basin which have changes in precipitation and temperatures, along with frequent and severe floods. This study aimed to identify the climatic parameters on flood area can be efficiently artificial neural network, better methods applied in anticipation of this event. The method used in this study to predict the process, multilayer perceptron neural network and radial basis that these two neural networks with multiple regression results were compared. Therefore, climatic daily data in 16 years cycle from four stations: Idnak, Dogonbadan, Dehdasht and yasouj (23 September 1994-22 September 2009) are used. By study of correlations between climatic parameters and discharge of Maroon river, effective parameters on flood are determined and multiple regression is used because of determination of effective entrance on flood and comparing the results with the neural network. Study of the results shows that multilayer Perceptron (MLP) along with training algorithm after flowing the error have 0.73 correlation in training process and in test process is 0.68 and also measure of NRMSE in training process is 0.57 and in test process is 0.66 that known as the best model for predicting storm water. Comparing the results of regression and neural network shows that neural network have a higher correlation than the regression, thus in neural network actual data and predicted data have more conformity than accomplished regression model.

In forest resource management, decision-making processes, such as qualitative factors, are not logged in mathematical equations so we need to new solutions than algorithmic methods. According to the capabilities of neural networks and recent application of them in forest resources, the purpose of this study was to compare the multi layer perceptron and the radial basis network to predict commercial and cordwood volume, in order to evaluate the performance of different networks to find the best type of network for achieving acceptable and valid results. In this purpose, 367 trees were marked of research and educational forest of kheyroud. Some factors such as diameter at breast height, diameter at stump, stump height, total height, topographic factors (slope, aspect and elevation), species, tree situation and minimum median diameter of last log were selected and then they were measured. They considered as input variables in network. Commercial and cordwood volume determined by traditional renewal volume and then they used as output network. Multi-layer perceptron (MLP) and radial basis function (RBF) were used for modeling. The hyperbolic tangent function and softmax function respectively used for network training in hidden layer of multi layer perceptron and radial basis function networks. As well as, the linear function used for network training in output layer. The data were divided into three sections for modeling: training, validation and test, each of which was 70%, 15% and 15%, respectively. Determination of the number of hidden layers and neurons of each layer was also performed by test and error and continued until the error value reached the minimum. Due to result, R2 value was respectively 0.94, 0.71 for commercial and cordwood volume in multi-layer perceptron network and 0.88, 0.65 for cordwood volume in radial basis function network. Also, RMSE value was respectively 1.297, 0.337 for commercial and cordwood volume in MLP network and 3.72, 0.397 for cordwood volume in RBF network. The result showed that multi-layer perceptron than radial basis network has acceptable accuracy to predict the commercial volume and cordwood volume. The only advantage of the radial basis function than multi-layer perceptron was less time required for training in modeling. Using a network and a model that has a higher accuracy with several variables among existing networks and models is prioritized. Thus, according to this new and powerful technique, the need for identifying a range of potential uses in the forest science community is felt as an alternative tool.

Selecting a model that simulate the runoff with high accuracy and less error, can be helpful in favorable management of water resources plans and increasing the performance of these plans. Also, increasing the accuracy of runoff simulation in the basins with no meteorological data, is of great significance in efficient management of water resources in these basins.

In this study, the artificial neural network methods were used to estimate the distribution of ascid family (Acari: Mesostigmata). For this aim, latitude, longitude and elevation from the sea level of 137 points were defined as inputs and output of method was number of species of this family on those points and Perceptron with propagation algorithm was evaluated in artificial neural network method. To evaluate the ability of neural networks used to predict dispersion, statistical comparison of parameters such as variance, statistical distribution and mean of spatial predicted values by neural network and their actual values were used. The results showed that there was no significant difference (p> 0.4) in the training and test phases between the values of the statistical characteristics of variance, the statistical distribution and the mean of real and predicted spatial data of this family by the neural network. It can be concluded that the artificial neural network method was able to predict the dispersion of this family with proper precision by integrating three factors of latitude and longitude and elevation from the sea level.

This experiment was designed for easy and accurate estimation of corn plant leaf area with multilayer perceptron (MLP) neural network and conducted at Faculty of Agricultural Sciences, University of Guilan, Rasht, Iran in 2013. Three plant densities (75, 85 and 95 thousand plants/ha) and five genotypes (Persia 454, 484, 565, 626 and 647) were considered as treatments. Samplings were conducted at various times (66, 80, 94 & 108 DAP). At each sampling, number of leaves per plant, number of green leaves, leaf length and width, plant leaf area, leaf and stem dry weight, plant height, stem diameter and biological yield were measured. Correlations analysis indicated that measured characteristics had positive significant correlation with plant leaf area (r≥0.859**) and they can be used as inputs for estimation of leaf area. Among these variables, the highest sensitivity was associated to leaf width, number of green leaves, leaf length, number of leaves per plant and stem diameter, respectively. However the model with a lower number of variable, i.e. including leaf width, number of green leaves and leaf length was more appropriate for quick estimation of leaf area. When a single input had been used for estimation of leaf area, leaf dry weight offered a better simulation than other variables (d = 0.989), so that 95.69% of leaf area changes was described through leaf dry weight (R2 = 0.9569) and it can estimate leaf area well (RMSE (%) = 15.67). In both methods of estimation for leaf area (by using single input and sensitivity analysis), the best fitted models were not affected by cultivar, plant density and interaction of these two factors. Therefore, a general model can be used for rapid and accurate leaf area estimation of genotypes and plant densities used in the experiment.

Different factors influence on the properties of wood composite panels. Evaluating all of these factors not only increases waste of time and energy but also decreases accuracy in estimation of influence value of selected factors in manufacturing panels in order to obtain optimum point of different properties of wood composite panels. Hence, application of a new statistical method is necessary for determination of model estimating production´s optimum point. This study was aimed to evaluate the artificial neural networks performance to model the physical properties of the particleboard made of canola stalks particles. The physical properties of the particleboard were modeled and estimated using different weight ratios of melamine formaldehyde to urea formaldehyde, canola stalks to poplar particles and mat moisture content through MLP, RBFN and ANFIS artificial neural networks. The results showed that MLP neural network has better performance than RBFN and ANFIS neural networks to estimate the physical properties of the particleboard. The results also showed that the artificial intelligence models have a proper precision and ability to predict the particleboard's physical properties. The results of the sensitivity analysis also showed that for estimating and , the most important parameter was mat moisture content with a positive effect on the modeling, and melamine formaldehyde to urea formaldehyde ratio was also the most effective parameter for estimating and .

This study aims at comparing the performance of logistic regression, maximum entropy, and multilayer perceptron techniques in preparing the predictive habitat distribution map of Amygdalus scoparia in rangelands of Qom province. For this purpose, vegetation sampling was done using random systematic methods after identifying pure habitats of this species. For soil sampling, eight profiles were excavated and two samples were taken from 0-30 and 30-80 cm depth. After analyzing the soil characteristics in the laboratory and generating the layers of physiographic characteristics (slope, aspect, and elevation), geology and physical and chemical characteristic of soil, predictive habitat distribution modeling was performed. Then the accuracy of generated map was evaluated by kappa. Based on the calculated kappa coefficient, logistic regression model was able to predict the habitat distribution of studied species at the excellent level (kappa= 0.91). Meanwhile, predictive maps derived from maximum entropy and multi-layer perceptron had very good aggrement with actual map (kappa 0.85 and 0.8, respectively). These results indicate that the logistic regression model is more accurate than other methods for predicting the distribution of this species due to specific circumstances. Based on logistic regression model, geological formation (Igneous formations) and soil gravel amount are the most influential factors effecting the the presence of this species in this habitat. These results indicate that in order to select the optimal modeling approach, it should be given special attention to the ecological niche of the species studied in addition to the capabilities of each method.

Gully erosion is a type of water erosion that causes significant sedimentation in watersheds and ýdamages in agricultural lands, rangelands, and infrastructures. This study was conducted to ýdetermine the potential of gully erosion by artificial neural network. The Levenberg-Marquardt ýý(LM) algorithm and Multi-Layer Perceptron were used employing soil, geology, land use, ýdistance to fault, slope, aspect, distance from roads, distance from drainage, and elevation data ýas its variables. Results showed that the structure of 1-13-9 with sigmoid activation function in ýthe hidden layer is more suitable for gully erosion potential assessment. Zonation of gully ýerosion revealed that the watershed area was divided into different classes of different extent, ýincluding 65.94% in very low, 4.07% in low, 3.78% in medium, 4.75% in high, and 21.94% in ývery high potential class. Furthermore, results indicated that slope less than 10%, 50 m distance ýfrom the stream, rangeland area, and lithological units of EM and M2 had the greatest impact ýon the occurrence of gully erosion.