fuzzy clustering

Germplasm improvement and genetic diversity are fundamental solutions for reliable and sustainable production of field crops. To study the genetic diversity and the relationship between agronomic traits, 800 winter bread wheat genotypes along with five varieties of Falat, Sardari, Roshan, Zare, and Pishgam in the farm of the Faculty of Agriculture of Vali –e- Asr University in Rafsanjan were evaluated in an augmented design. Cluster analysis was conducted by two methods k-Means and fuzzy clustering. The studied genotypes were divided into 4 groups in both methods. The traits were including biological weight, seed yield, harvest index, height, spike length, and seed weight per spike were respectively the most important in grouping the genotypes in both clustering methods. In the regression analysis, the traits of days to flowering and days to ripening, spike length, biological yield, number of seeds and weight of seeds per spike, 1000 kernel weight, and harvest index accounted for 85.89% of the total variation in grain yield. The traits of Biological yield and harvest index had the most direct and positive effect on grain yield. Based on the results, some genotypes were superior to control cultivars based on a number of traits, especially seed yield and its components. These genotypes can be considered as candidate genotypes in breeding programs in Rafsanjan climate conditions.

Sustainable soil management with a correct understanding of soil properties helps to maintain soil fertility and prevent soil degradation. This research was conducted to evaluate the potentional use of soil management zones (MZs) as an efficient method to improve the productivity of wheat cultivation. According to this, 140 soil samples were taken randomly from wheat fields. Physical and chemical properties, along with topographical and spectral indices, were used to delineate MZs. Topographic attributes were extracted from digital elevation model (DEM) map preapared in SAGA GIS 7.8.2 software from the study area. Mapping the applied variables, selecting the optimal variable using principal component analysis, and delineating the study area based on soil properties, topographic attributes and spectral indices by using cluster algorithms in combination with geostatistics are major steps for delineating management zones. The fuzziness performance index (FPI) and normalized classification entropy (NCE) were investigated for the number of MZs. The semivariogram and kriging prediction maps showed different spatial distribution patterns, with spatial autocorrelation ranging varies from weak to strong for most of the applied variables. Finally, six principal components (PCs) with an eigenvalue of more than 1 and a total variance of 76.3% were chosen for further analysis. Based on the lowest values of FPI and NCE, six MZs were identified. The mean values demonstrate the difference between the applied properties in the MZs. This study showed that the delineation of MZs can be effectively used in soil management for cultivation crops to maximise crop production.

The proper estimation of the amount of suspended sediment in rivers has an important role in erosion and sediment studies, hydrology and management of watersheds. The simulation of suspended sediment in hydrological systems that has a lot of complexity and at the same time our understanding of the components and processes within them is always uncertain led to the use of many intelligent models, including artificial neural networks (ANNs). However, the use of these smart models also faces challenges. Determining the proper structure of the network requires optimization of the parameters used (such as the optimal number of neurons and layers, weight and bias, and the type of activation functions), which their proper calibration, using test and error, leads to a lot of time spent in low efficiency. In this study, a multilayer perceptron (MLP) was used to simulate the daily sediment load of the Nirchai River at the site of the Nair hydrometric station in Ardebil province. In order to train the models, in addition to the error back propagation (BP) algorithm, Particle Swarm Optimization (PSO) algorithm was used to optimize the weight and bias of ANNs. The fuzzy clustering method was also used to increase the power of generalization of the models. The results showed that training of ANN models with PSO algorithm with decreasing estimation error (decreasing the PBIAS of estimation and root mean square error up to 0.3% and 10.4 tons per day respectively) is more effective than ANN models that use only error BP techniques. Due to insufficient recorded sediment data in most hydrometric stations of the country on the one hand and the need to train ANNs with sufficient data on the other hand, the use of evolutionary algorithms (e.g. PSO algorithm) can be a good solution for improving the efficiency of intelligent models.

In many water resource projects such as dams، flood control، navigability، river aesthetics، environmental issues and the estimation of suspended load have great importance. The complexity of sediment behavior and mathematical and physical model inability in simulation of sedimentation processes have led to the development of new technologies such as fuzzy logic which has the ability to identify nonlinear relationship between input and output variables. In this study، the application of fuzzy clustering algorithm in estimating the annual amount of sediment was studied. So، the corresponding data of flow and sediment discharge of Valykben station in kasilian basin during 1349-1350 till 1353-1354 period was daily determined. The data was divided in two groups i. e. 75% as training data and 25% for test data. Then، the efficiency of model was obtained by using statistical parameters such as correlation coefficient، nash-satklyf coefficient، mean square error root and variance ratio. The result showed that the classification of data on the annual time scale and use of fuzzy clustering algorithm can estimate 0. 49 values of the measured annually suspended sediment transport. Furthermore، on the same scale of classification، i. e. annual scale، this value was obtained 0. 19. Thus، using fuzzy clustering algorithm can lead to higher accuracy and reliability than rating curve method، which is suggested for estimating suspended sediment transport.

Unfortunately the groundwater balance in all plains of Chaharmahal va Bakhtiari province is negative. The area of this province is only 1% of Iran Country, whereas it has 10% of water resources in this country. So, flood spreading will be helpful both in increasing the water table and flood control. Hence this research carried out to determine the most suitable areas for flood spreading in Kiar, Mizdej, Sefiddasht and Borujen plains. Slope, infiltration rate, alluvial thickness, alluvial quality and land use factors were explored as affective parameters in flood spreading. These parameters were explored, classified, weighted and overlaid using GIS system. Boolean logic, overlay Index and Fuzzy clustering methods were used to overlay these layers and the study areas were divided into two suitable and unsuitable classes for flood spreading. Results showed that based on overlay Index, Boolean logic and Fuzzy clustering 10.85%, 10% and 14.4 % of study area was suitable for flood spreading, respectively.  Infiltration rate was recognized as most limiting parameter among explored parameters that it due clay texture of soil in most parts of province. Alluvial thickness was recognized as another limiting factor for water spreading in Chaharmahal va Bakhtiari province.  

This study was conducted to know the intensity and spatial pattern of soil heavy metals pollution around the national Iranian lead and zinc complex factories in Dizajabad area, Zanjan province. A sum of 272 surface samples (0-10 cm) were collected based on a grid methods with nodes of 250 and 500 meters and the samples were extracted with HNO3 and the concentrations of Pb, Zn and Cd were determined. Average concentrations of Pb, Zn, and Cd in soil were 181.3, 182.28 and 1.48 mgkg-1, respectively. To study spatial pattern of variables, experimental variograms were calculated and best model was fitted in each case, which was exponential for all. Influence ranges for variograms of Pb, Zn, and Cd were 1131, 1073 and 1026 m, respectively. Using ordinary kriging method, interpolation was carried out, which performed well. To classify the data in taxonomic space, a fuzzy c-means with extragrades clustering algorithm was run on the data using several fuzziness exponents (Φ). Plausible fuzzy clusters were obtained at Φ=1.2. To find 7 as the optimal number of classes from a 2-10 assemblage, fuzziness validation functions F, H and S were used. Most polluted classes A, B, E and G are located at center and around the factory, while C, D and F are covered northern, northeast and southeast parts of the study area. Major parts of extragrade class was the location of factory and landfill compliance.

Increasing soil pollution due to heavy metals is a major concern of present day soil research. This study conducted to know intensity and spatial pattern of soil heavy metals pollution in a 10،000 ha area of Anguran region near Zanjan. A number of 315 surface (0-10 cm) samples collected at nodes of a 500 meter equilateral grid. Beside HNO3-extracted Pb، Cu، Ni، Cd and Zn content، EC، pH، OC and CaCO3 of the samples were determined. To classify data in taxonomic space، a fuzzy c-means with extragrades clustering algorithm ran on the data using several fuzziness exponents (Φ). Plausible fuzzy clusters obtained at Φ=1. 3. To find eight as the optimal number of classes from a 2-10 assemblage، fuzziness validation functions F، H، and S were used. Scrutiny of class centroids and membership values revealed that though number of variables was not numerous، the algorithm clustered data sensitively. Spatial distribution of classes mapped through geostatistical analysis of membership values. Though extragrade class embraced extreme values، still all centroids of regular classes showed severe pollution. Most polluted classes C، E، F and H located at center to southeast، while A، B، D and G covered northern and western parts of the study area. Extragrade class widely spread in the area that confirmed interspersed outliers among all others. Major part of extragrade class lied across southeastern part of the area. Results of the study showed that numerical classification of soil pollution is rather realistic، thus provides a pragmatic approach to the problem.

In this paper by using fuzzy logic and c-mean clustering, different combinations of daily climatic variables including air temperature, relative humidity, wind speed and sunshine hours as input data for predicting pan evaporation were investigated. Therefore, meteorological parameters in years of 1386 and 87 at three synoptic stations (Ahvaz, Abadan and Izeh) in Khuzestan province were used. The results showed that the Takagi-Sugeno fuzzy model and c-mean clustering were able to predict pan evaporation, so that fuzzy models with cluster number of 2 and 3 estimated pan evaporation parameter by correlation coefficients varying from 0.86 to 0.91.