fuzzy inference system

The importance of quality in the industry to obtain and produce high quality products has been known for a long time. Quality in the production environment improves reliability, increase production and attracts customer satisfaction. Classical control diagrams, using precise and definite data, place production processes in two groups, «rejection» or «acceptance». Descriptive characteristics are in fuzzy conditions due to ambiguity in the number of defects in the product and decision making by the inspector, and fuzzy sets by defining continuous membership functions and using ambiguous and indefinite data by using triangular and trapezoidal fuzzy numbers in the form of control categories are classified and express the quality level of the product more realistically. This research is an applied and descriptive research, which was carried out with the aim of designing model of Statistical Process Control through a fuzzy inference system to control descriptive characteristics in the food industry. Sampling system has been used in the inspection station to collect information and according to sensory and physical characteristics, the quality level of the produced chocolates was determined. In the Classical Method, 28 cases were identified «under control» and only 2 cases were «out of control». But in the investigation with the fuzzy designed model, 28 samples were «under control», 1 sample was «relatively under control» and 1 sample was «out of control»; Based on the research result, practical suggestions were recommended to the relevant industry.

Since the increase in the depth of groundwater and its intensification can indicate serious limitations in the exploitation of these resources, predicting the changes of this parameter plays an important role in managing these resources and preventing possible damage to them. For this purpose, the use of smart methods has been strongly recommended by researchers. In this research, the methods of multilayer perceptron neural network (MLP), fuzzy inference system (fis), adaptive neural fuzzy inference system (ANFIS), and the combined method of fuzzy neural inference system and particle swarm optimization (ANFIS-PSO) were used for simulation of groundwater Fluctuations depth in Haji Abad area between March 1995 to October 2022 on a monthly scale. The training and testing phases were done with 75 and 25 percent of data, respectively. To measure the accuracy of the models, root mean square error (RMSE), mean absolute percentage error (MAPE) and mean absolute value of error (MAE) indices were used. The best results in the training phase are related to ANFIS-PSO, ANFIS, and MLP models, respectively. Simultaneously with the training of the mentioned models, the testing stage of said models was also implemented. Finally, the best results in this stage belonged to the neural network model with time delay [1 3 5], the ANFIS-PSO model with time delay [1 2 3], and the neural network model with time delay [1 2], respectively. The accuracy indices in the test stage for the best models are (0.1871, 0.1865, 0.1857) for RMSE, (0.7402, 0.6715, 0.6684) for MAPE, and (0.1326, 0.1238, 0.1198) for MAE, respectively. These values show that all three models have an error of less than 20 cm, an error percentage of less than 0.75%, and an absolute error of less than 14 cm, which indicates the acceptable accuracy of these models. Also, the coefficient of determination obtained from the regression relationship of the calculated and measured values of the groundwater depth in the test phase for all three models is around 0.82, which indicates a relatively high linear relationship between these two parameters.

Assessing water quality is an important step toward the optimal and appropriate use of drinking water resources. Therefore, the necessity of studying water quality characteristics in water resource management programs has been highly considered. Ambiguity and lack of inherent certainty governing water resources in the evaluation of goals, criteria, and decision-making units, as inconsistency and carelessness in the opinions and judgments of decision-makers have led to the tendency towards theories of fuzzy sets and, as a result, fuzzy logic as an efficient and useful tool for planning and making decisions. In the present underground water quality water was first classified by international standard methods (definitive evaluation method) for drinking purposes. Then classification was modeled and compared using Mamdani fuzzy inference. For this purpose, the four-year average of quality parameters of underground water sources related to 33 sources including 10 well rings, 22 spring mouths, and an aqueduct in operation in Sardasht Cityre used as inputs in two cases. In the deterministic evaluation method (Schoeller diagram), the characteristics and the water quality determination diagram were determined. In the four-year average fuzzy inference model, eight water quality parameters were classified into three groups, in the first group the parameters of Na+, Ca+2, and Mg+2, in the second group the parameters of HCO3-, SO4-2, and Cl- were placed in the third group of two TH and TDS parameters. After determining each group with two input parameters, each input parameter was considered including three membership functions, so that the rules considered for it were estimated as nine (3x3). The results based on the deterministic method showed that all the studied samples were in the good to acceptable group. But Mamdani's fuzzy findings showed that two samples with a confidence level of 50% were in the acceptable category and other samples with a confidence level of 83-87% were placed in the desirable category for drinking.

In addition to rich resources of fossil fuels, Iran has a lot of renewable energy potential. On the other hand, considering the climate diversity in the country and the natural conditions and potentials in different regions, instead of national planning, we should move towards regional energy planning and develop a regional renewable energy portfolio. In the present research, firstly, the potential measurement criteria of different types of renewable energy, including solar, wind, geothermal, hydroelectric and biomass, based on geographic information system maps and data received from the Meteorological Organization and SATBA, for 1361 latitudes and longitudes, has been scored. Then, using the Rapidminer software, the geographic points were divided into 5 clusters, each cluster includes areas of equal potential with the greatest similarity. Two of these 5 clusters are considered to be among the dry lands of the country. Then, based on the review of library resources and usage from the opinions of SATBA experts (Renewable Energy Research Group), a fuzzy inference model based on 5 sustainable development criteria including: access to technology, investment costs, capital productivity, employment rate, and environmental consequences along with design potential measurement criteria and based on the fuzzy rules defined on these criteria. The percentage share of each type of energy in the energy portfolio of each cluster was calculated. In the final step, based on demographic criteria including unemployment rate, population growth rate, acceptance culture (literacy rate), investment security, to prioritize clusters for strategic planning of the government and other influential institutions such as governorates, municipalities and chambers of commerce. For example, in cluster 4, includes some cities in the provinces of Isfahan, Khorasan, Yazd, Kermanshah, Fars, and Kohkiloyeh, which are classified as arid and semi-arid regions of the country according to the criteria of potential measurement and development criteria that has an energy portfolio with 25% share of wind energy, 39% share of solar energy, 10% share of hydroelectric energy and 26% share of biomass energy. Population growth and investment security are the first priority.

Nowadays, water resource management has been shifted from the construction of new water supply systems to the management and the optimal utilization of the existing ones. In this study, the reservoir operating rules of Doroodzan dam reservoir, located in Fars province, were determined using different methods and the most efficient model was selected. For this purpose, a monthly nonlinear multi-objective optimization model was designed using the monthly data of a fifteen-year period (2002-2017). Objective functions were considered as minimizing water scarcity index in municipal, industrial, environmental and agricultural sectors. In order to determine the operating rule curves of reservoir, in addition to the nonlinear multi-objective optimization model, the methods of ordinary least-squares regression (OLS), fuzzy inference system and adaptive network fuzzy inference system (ANFIS) were used. Also, the reliability, resiliency, vulnerability and sustainability criteria were used to compare the different methods of reservoir performance rules. The results showed that ANFIS model had the higher sustainability criterion (0.26) due to its greater reliability (0.7) and resilience (0.42), as well as its lower vulnerability (0.13), thereby showing the best performance. Therefore, ANFIS model could be effectively used for the creation of Doroodzan reservoir operation rules.

Monitoring and managing the wildlife populations and habitats required to model the species distribution and habitat suitability. So, Gazella subgutturosa potential distribution in Mishdagh Protected Area was modeled using fuzzy (based on ecological knowledge of local communities) and MaxEnt (based on species occurrence records) approaches; thus, in addition to model the species distribution using maximum entropy algorithm (MaxEnt approach) and fuzzy inference system (fuzzy approach), we can also assess and compare the performance of each approach. In addition, the accuracy of predictive models was tested using jackknife test. Also, we applied threshold of 10%. Based on results of fuzzy and MaxEnt approaches, the most important variables for species potential distribution modelling were land use, distance to farms and distance to water sources. Also, 47.45% and 14.08% of study area predicted as species potential presence area in fuzzy and MaxEnt approaches, respectively. According to results of jackknife test, success rates of fuzzy and MaxEnt approaches were 80.95% and 66.66%, respectively (p

Nowadays, greater recognition of drought and introducing its monitoring systems, particularly for the short-term periods, and adding predictability to these systems, could lead to presentation of more effective strategies for the management of water resources allocation. In this research, it is tried to present appropriate models to predict drought in city of Semnan, Iran, using time series, adaptive neuro-fuzzy inference system (ANFIS) and artificial neural networks (MLP and RBF). For these modeling processes, average monthly meteorological parameters of rainfall, temperature, minimum temperature, maximum temperature, relative humidity, minimum relative humidity, maximum relative humidity and SPI drought index were used during the period 1966 to 2013. The results showed that among the many developed models, the ANFIS model, with input data of average rainfall, maximum temperature, SPI and its last-month value, 10 rules and Gaussian membership function, showed appropriate performance at each stage of training and testing. The values of RMSE, MAE and R at training stage were 0.777, 0.593 and 0.4, respectively, and at testing stage were 0.837, 0.644 and 0.362, respectively. Then, the input parameters of this model were predicted for the next 12 months using ARIMA model, and SPI values were predicted for the next 12 months. The ANN and time series methods with low difference in error values were ranked next, respectively. The input parameters SPI and temperature had better performance and rainfall parameter had weaker performance.

Evapotranspiration (ET0), a major component of the hydrologic cycle, is important in water resources development and irrigation planning. The ET0 for west and northwest of Iran was estimated using the FAO Penman-Montieth method (FAO-56) as the reference. The performance of four different data-driven methods, namely the Artificial Neural Networks (ANN), Adaptive Neuro Fuzzy Inference System (ANFIS) with grid partition (GP), ANFIS with subtractive clustering (SC), and Gene Expression Programming (GEP) were investigated on the ET0 estimation. Latitude, longitude and altitude of stations, and the periodicity component were used as inputs to the applied models to predict the long-term monthly ET0 using the data from 12 stations in the west and northwest of Iran. The maximum coefficients of determination (R2) were found to be 0.994, 0.998 and 0.997 for the ANN, ANFIS-GP and ANFIS-SC models at the Sanandaj Station, respectively. The highest R2 (0.982) of the GEP model was calculated for the Khoy Station. The root mean squared error ranged 0.311-1.551, 0.368-1.319, 0.450-1.80 and 0.435-0.833 mm/day for the ANN, ANFIS-GP, ANFISSC and GEP models, respectively. The results revealed the high accuracy of the GEP in the prediction of ET0. Therefore, the GEP model can be applied as a simple method in future studies as an alternative to the use of empirical relationships for the ET0 estimation.

In examination of hydrologic issues and water resources, analysis of rainfall information has special importance. Due to various restriction, lack of harvest and visit reading rate of rainfall, limited us to access these information. So apply the methods of estimating water level in specific points is important in the various studies, on the base information of contiguous points. In this research, the common methods of interpolation, Kriging ground statistics and adaptive neuro-fuzzy ablation system were evaluated in Fars province. In this study, 20 synoptic stations of Fars province has been used during 29 years statistical period from 1981-1982 until 2009-2010. Through the investigation was done, December in years of 1992-1993 and 2004-2005 as the best pattern of wetness and April in years of 2008-2009 and 2009-2010 as the best pattern of drought period and also April and November in the 2006-2007 was chose as the annual normal pattern. In Adaptive neuro-fuzzy inference system (ANFIS) for each of the above years, the number of membership, Gaus 2mf, Gsussmf and Gbell were evaluated separately. It’s noticeable that at first consider 15 stations as a training in this system and 5station Tongab, Shourjeh, Baba Arab, Shiraz and Chamriz were evaluated. In this project rating of RMSE, R2 and EF evaluated and compared by two methods of Kriging and Adaptive neuro-fuzzy inference system. According to the obtained results it became clear that in the wetness periods Adaptive neuro-fuzzy inference system, provided more acceptable results. Also during the drought period for predict the rainfall, Kriging method is suggested. The most accurate results are obtained in normal periods in April by Kriging method and in November by Adaptive neuro-fuzzy inference system method.

Several modelling techniques have been developed for habitat suitability modelling. In the meantime, the Fuzzy Inference System (FIS) with ability to model uncertainty of input variables is an effective method to model wildlife species habitat suitability. So, Persian Leopard habitat suitability was predicted in Shimbar Protected Area using FIS. Therefore, the effective environmental variables were determined. We also defined and determined the linguistic variables, linguistic values, and range of them. Then, we designed the membership functions of the fuzzy sets of the input and output variables. Also, the definition of the fuzzy rules in the system was performed. Finally, the defuzzification of output was carried out. The accuracy of the predictive model was tested using AUC. Also, 11 FISs were developed to determine sensitivity of the models and important variables in modelling. The results showed that the predictive model was more efficient than the random model (AUC=0.960). In addition, the ‘distance to capra’ was the most important predictor. According to the success of FIS in Persian Leopard habitat suitability modelling, we suggest this method to improve and complete the existing spatial information of wildlife habitats in Iran, especially about regions and species that have been less studied.

The aim of this research is to study two intelligent systems including artificial neural network and multi – layer adaptive neuro-fuzzy inference system (ANFIS) for modeling the output energy in production of greenhouse and open-field cucumber. In order to determine the pattern of consumed energy, the required information was collected directly from the 160 cucumber beneficiary. By analyzing the results, the average input and output energy was respectively equal to 1159901.1 MJ and 173985.256 MJ for greenhouse cucumber production and respectively 75648.11 MJ and 22694.433 MJ for open-field cucumber production . The maximum input energy in production of greenhouse and open-field cucumber was related to fuel and electricity respectively. The minimum input energy was related to seed energy. Using the input and output energy data, modeling of output energy was performed based the input energy and using the mentioned modeling methods. Refer to the results of ANFIS model, the values of coefficient of determination and root mean square error for greenhouse and open-field cucumber were equal to 0.9924, 9920 and 0.051 and 0.0130, respectively, Furthermore, these values were determined using the neural network modelwith optimum structure of (8-10-1) for greenhouse cucumber and structure of (8-12-1) for open-field cucumber. It was revealed that the coefficient of determination value was equal to 0.9492 and 0.9785, and RMSE was equal to 0.0121 and 0.0418, respectively. The result.

Due to the nature of the sediment data, selection of appropriate methods for processing the data before entering them to the artificial intelligence models can enhance the reliability of simulations results. In this study, the effects of sediment data processing procedures on ANN and ANFIS models outputs in 7 Dez Basin stations were evaluated. Accordingly, three scenarios were considered: In the first scenario, original data was used without exerting any processing technique; in the second scenario, the data was normalized; and in the third scenario, logarithm of data were used according to logarithmic distribution governing. The simulation results showed that using data logarithm leads to higher performance and lower error, especially in stations where the best fit probability distribution is one of the log family distributions. Finally, among applied models, ANFIS showed the best performance with coefficient efficiency of 0.95 and RMSE of 5.4, MSE of 1.4 and ME of 0.42 in Biatoon gauging station and using the third scenario.

Groundwater resources are the main source of fresh water in many parts of Iran. Groundwater resources are limited in quantity and recently due to increase of withdrawal, these resources are facing great stress. Considering groundwater resources scarcity, maintaining the quality of them are vital. Traditional methods to evaluate water quality insist on determining water quality parameter and comparison between them and available standards. The decisions in these methods rely on just specific parameters, in order to overcome this issue, water quality indices (WQIs) are developed. Water quality indexes include a range of water quality parameters and using mathematical operation represent an index to classify water quality. Applying the classic WQI will cause deterministic and inflexible classifications associated with uncertainties and inaccuracies in knowledge and data. To overcome this shortcoming, using the fuzzy logic in water resources problems under uncertainty is highly recommended. In this paper, two approaches are adopted to assess the water quality status of the groundwater resources of a case study. The first approach determined the classification of water samples, whilst the second one focused on uncertainty of classification analysis with the aid of fuzzy logic. In this regard, the paper emphasizes on possibility of water quality assessment by developing a fuzzy-based quality index even if required parameters are inadequate.
The case study is located in the northwest of Markazi province, Saveh Plain covers an area of 3245 km2 and lies between 34º45′-35º03′N latitude and 50º08′-50º50′E longitudes. The average height of the study area is 1108 meter above mean sea level. The average precipitation amount is 213 mm while the mean annual temperature is 18.2oC. To provide a composite influence from individual water quality parameters on total water quality, WQI is employed. In other words, WQI is a weighting average of multiple parameters. The present research used nine water quality parameters (Table 2). In this paper Fuzzy Water Quality Indices (FWQIs) have been developed, involving fuzzy inference system (FIS), based on Mamdani Implication. Firstly, five linguistic scales, namely: Excellent, Good, Poor, Very poor, and Uselessness were taken into account, and then, with respect to ‘If→then’ rules the FWQIs were developed. Later, the seven developed FIS-based indexes were compared with a deterministic water quality index. Indeed seven FWQIs based on different water quality available parameters have been developed. Then developed indices were used to evaluate the water quality of 17 wells of Saveh Plain, Iran.
The present study analysed groundwater quality status of 17 wells of Saveh Plain using FWQI and WQI. Based on the driven results from WQI and its developed fuzzy index, similar performance was observed in most of the cases. Both of them indicated that the water quality in six wells including NO.1, 2, 6, 12, 13, and 17 were suitable for drinking. Due to the fact that the values of both indexes were under 100, the mentioned wells could be considered as drinking water supplies. The indexes illustrated the very poor quality of wells NO.7, 9, 10, 11, 14, and 16. As a result, according to FWQI1 along with WQI, nearly 35% of wells have proper drinking water quality, while approximately 30% and 35% of them suffered from poor and very poor quality, respectively. The overall picture of water quality within the study area was not satisfying, hence, an accurate site selection for discovering water recourses with appropriate quality for drinking purpose must be responsible authorities’ priority. Analysis of FWQI2, FWQI3 and FWQI4 revealed that elimination of the parameters slightly changed the result of FWQI2; however, FWQI3 and FWQI4 did not vary considerably. Thus, Cl influenced the water quality slightly, but Ca and K did not affect the water quality of the plain. The results showed that inexistence of one of the mentioned parameters would not affect the computational process adversely. A glance at FWQI5, FWQI6 and FWQI7 revealed the improper performance of FWQI5 to show wells’ water quality status. Throughout the FWQI5 evaluation process, all the wells’ water quality stood in Excellent category. Due to the considerable values of TDS in the Plain, elimination of this parameter in FWQI5 caused inappropriate evaluation. Hence, whenever a case study deals with a high value of a specific quality parameter, elimination of that parameter would negatively demote validation of the analysis. Figures (3)-(6) represented the results of WQI along with seven FWQIs for 17 utilized wells’ water quality assessment in the study area during the proposed periods.
Throughout the present study, the capability of seven FIS-based indexing procedures in modelling the water quality analysis of 17 wells of Save Plain was discussed. The proposed FWQIs were developed on the basis of Mamdani approach by applying triangular and trapezoidal membership functions to determine the groundwater quality of the case study according to the nine parameters. The results revealed that FWQI1-4 outperformed others. On the other hand, FWQI5-7 which eliminated three out of the nine parameters, did not made a valid contribution to the computational context. This might be related to omitting the effective water quality parameters from the inputs of the model. The results also illustrated that, only six out of 17 wells of the region could be considered as suitable sources for the drinking purpose. The water quality status in five wells was not satisfying, and six wells were plagued by very poor quality of water.

Study of soil hydraulic properties such as saturated and unsaturated hydraulic conductivity is required in the environmental investigations. Despite numerous research, measuring saturated hydraulic conductivity using by direct methods are still costly, time consuming and professional. Therefore estimating saturated hydraulic conductivity using rapid and low cost methods such as pedo-transfer functions with acceptable accuracy was developed. The purpose of this research was to compare and evaluate 11 pedo- transferfunctions and Adaptive Neuro-Fuzzy Inference System (ANFIS) to estimate saturated hydraulic conductivity of soil. In this direct,saturated hydraulic conductivity and physical properties in 40 points of Urmia were calculated. The soil excavated was used in the lab to determine its easily accessible parameters. The results showed that amongexisting models, Aimrun et al model had the best estimation for soil saturated hydraulic conductivity. For mentioned model, the Root Mean Square Error and Mean Absolute Error parameterswere 0.174 and 0.028 m/day respectively. The results of the present research, emphasises the importance of effective porosity application as an important accessible parameter in accuracy of pedo-transfer functions. sand and silt percent, bulk density and soil particle density were selected to apply in 561 ANFIS models. In training phase of best ANFIS model, the R2 and RMSE were calculated 1 and 1.2×10-7 respectively. These amounts in the test phase were 0.98 and 0.0006 respectively. Comparison of regression and ANFIS models showed that the ANFIS model had better results than regression functions. Also Nuro Fuzzy Inference System had capability to estimatae with high accuracy in various soil textures.

Flood routing is one of the most important issues in river engineering because of prediction of the ascent and descent of flood hydrograph. As flood is a variable and unsteady flow، its routing requires accurate and detailed data collection at hydrometry stations. The Muskingum models very offer a useful procedure among the flood routing methods. Moreover، application of the artificial intelligence methods have grown substantially in the different water engineering and watershed modeling endeavors in recent decades. In the present study، the data collected by Wilson، Wu et al.، and veiss man Jr. and Lewis for three different rivers in the U. S. A were used for the flood routing processes using the Muskingum، artificial neural network، adaptive neuro-fuzzy inference system، and genetic programming. Simulation results of flood routing process using the mentioned methods were compared using the statistical indicators of R2، RMSE and MBE. The results indicated that the artificial intelligence methods were superior to the Muskingum method due to their lower RMSE. The RMSE value for the artificial intelligence techniques was 0. 00174 and for the Muskingum method it was 28. 727. The Muskingum method was not successful in flood hydrograph simulation with multi peaks. Despite the slight differences in accuracy estimation and error values ​​in the models، the artificial neural networks proved their superiority with the highest R2، and lowest RMSE and MBE. The adaptive neuro-fuzzy inference system and genetics programming were placed in next levels. Based on the ease of use and more accurate results، the use of artificial intelligence methods is recommended for further studies in this region.

Sugarcane fields are affected by different parameters and factors such as ground water table¡ salinity of saturated soil¡ depth of irrigation¡ variety and age of plants and etc. Evaluating effects of these parameters¡ it is possible to propose solutions to maximize sugarcane fields performance. In this paper Adaptive Neuro - Fuzzy Inference System (ANFIS) is used to model the performance of sugarcane fields. This study is performed based on three years data of «Mirza koochak khan cultivation and industry». Results showed that the proposed model has a correlation factor of 0. 978¡ RMSE of 1. 35 and error of 3. 2 The proposed model has a very high accuracy in performance forecasting of sugarcane fields.

The most important component of the hydrologic cycle, which plays a key role in water resource management, crop yield simulation, and irrigation scheduling is evapotranspiration. Therefore, developing a low cost and precise model is very essential for hourly ETo calculations. Although, there are numerous empirical formulas, due to the complicated nature of the hourly evapotranspiration event, the data availability, high cost, and data gathering error, their performances are not all satisfactory. Thereafter, this paper develops an hourly ETo estimation model based on fuzzy inference system (FIS) technique. After analyzing the different models and different combinations of hourly meteorological data, hourly reference evapotranspiration calculated with four fuzzy models. Penman-Montieth-FAO56 Model considered as the comparison basis for hourly estimating reference evapotranspiration models. Comparing models was done with mean root squared error, mean deviation error, coefficient of determination, Jacovides (t) and Sabagh, et al (R2 / t) criteria. The Required data gathered from the private weather station in Fariman city. With removing missing data, 9128 hourly data extracted from two-year statistical period, 2008-2009. Meanwhile, 70 percent of the data was used for model training, and 30 percent for model testing. The results showed that, fuzzy model output is acceptable in relation to Penman-Montieth-FAO56 and ASCE models output. The fuzzy model with four inputs has the highest correlation (0.99) to reference model. The fuzzy model with two inputs: solar radiation and relative humidity, presented proper values for evaluation criteria (RMSE=0.048, MBE=-0.018, R2=0.97, t=32, and R2/t=0.0295) in training phase. Under the testing phase, results were very similar to training phase. The comparison of Fuzzy model outputs with ASCE models also indicated that fuzzy model with three inputs of radiation, relative humidity, and temperature has the highest matching value (RMSE=0.05, MBE=-0.014, R2=0.95, t=13.9 and R2/t=0.068), in the training phase, which was justified with testing results. According to this study, fuzzy model can be a proper method for estimating hourly reference evapotranspiration. While, fuzzy model is simple, accurate, and does not have complex calculations like hybrid models.

The aim of this research is the assessment of the relation between rainfall and large scale synoptically patterns at Khorasan Razavi province. In this study, using adaptive neuro fuzzy inference system, the rainfall estimation has been done from April to June in the Area under study. Spring rainfall data including the information of 38 synoptic, Climatologic and rain gauge stations from 1970 to 2007 has been selected from Iranian Meteorological Organization and Ministry of Energy. In this paper, we are analyzed 38 years of rainfall data at Khorasan Razavi province located in northeastern part of Iran at latitude-longitude pairs (34°-38°N, 56°- 62°E). The Adaptive Neuro-Fuzzy Inference system based on synoptically patterns with 38 years of rainfall data was trained. For performance evaluation, network predicted outputs were compared with the actual rainfall data. In this Study, at the first step, the relationship Between synoptically pattern variations including Sea Level Pressure (SLP), Sea Surface Temperature (SST), Sea Surface Pressure Difference (?SLP), Sea Surface Temperature Difference (?SST), air temperature at 700 hpa, thickness between 500 and 1000 hpa level, relative humidity at 300 hpa and precipitable water were investigated. As the second step, the model was calibrated from 1970 to 1997. Finally, rainfall prediction is performed from 1998 to 2007. The model that used in this research has an input layer, one hidden layer and an output layer. The number of neuron for input layer, hidden layer and output layer was 13-28-1, respectively. The results of simulation reveal that adaptive neuro fuzzy inference systems are promising and efficient.