جستجوی مقالات مرتبط با کلیدواژه "مدل anfis" در نشریات گروه "جغرافیا"

In geodesy, three levels are considered: the physical surface of the earth on which mapping measurements are made, the ellipsoidal reference surface (geometric datum) which is the basis of mathematical calculations, the geoid physical surface (physical datum) which is the basis for measuring heights. Satellite positioning systems measure the height of points relative to the ellipsoid surface. The geoid is one of the equipotential surfaces of the earth's gravity field, which approximates the mean sea level (MSL) by least squares. Geoid is very important in geodesy as a representative of the physical space or the space of observations made on the earth and also as the base level of elevations. The separation between the geoid and the geocentric reference ellipse is called geoid height (N). Although there is only one equipotential surface called geoid, various methods are used to determine it. These methods include: geometric method, geoid determination by satellite method, Gravimetric methods and geoid determination using GPS/leveling.

In this paper, the aim is to estimate the height of the local geoid using machine learning models. To do this, artificial neural network (ANN), adaptive neuro-fuzzy inference model (ANFIS), support vector regression (SVR) and general regression neural network (GRNN) models are used. The geodetic coordinates of 26 GPS stations in the north-west of Iran along with their orthometric height (H0) and normal height (h) were obtained from the national cartographic center of Iran. In all stations, the difference of orthometric height and normal height is considered as geoid height (N). Therefore, the geodetic longitude and latitude of the GPS stations are considered as the input of the machine learning models, and the corresponding geoid height was considered as the output. In order to test the results of machine learning models, two modes of 4 and 7 test stations are considered. Also, the output of the models is compared with the local geoid model IRG2016 presented by Saadat et al. for the Iranian region and also the global geoid model EGM2008.

Due to the availability of a complete set of observations of GPS stations along with orthometric height obtained from leveling in the north-west region of Iran, the study and evaluation of the models proposed in the paper has been carried out in this region. Observations of 26 GPS stations of North-west of Iran were prepared from the national cartographic center (https://www.ncc.gov.ir/). Two modes are considered for training and testing of ANN, ANFIS, SVR and GRNN models. In the first case, the number of training stations is 22 and the number of test stations is 4. But in the second case, by increasing the number of test stations to 7 stations, the error evaluation of the models has been done. It should be noted that the distribution of training and test stations is completely random. After the training step of machine learning models and choosing the optimal structure, the test step is performed in two different modes (4 and 7 stations). At this step, the value of the geoid height in the test stations is estimated and compared with the value obtained from the difference of orthometric height and normal height as a basis. Two statistical indices of relative error in percentage and RMSE in centimeters were calculated for all models and presented in Table (1) for the first case.Table 1. Relative error (%) of ANN, ANFIS, SVR, GRNN and IRG2016 models in the test stations considered for the first case According to the results of Table (1) and comparing the relative error values of all models in the test stations, it shows that the ANFIS model was more accurate than other models. After ANFIS model, IRG2016 model has higher accuracy than ANN, SVR and GRNN models. It should be noted that the IRG2016 local model uses the observations of all Iranian plateau stations to model the local geoid, and therefore it is expected that this model will be more accurate in the study area than other models.

The evaluations show that in the case of 22 training stations and 4 test stations, the RMSE of ANN, ANFIS, SVR, GRNN and IRG2016 models in the test step are 37.32, 19.83, 49.34, 53.82 and 29.65 cm, respectively. However, in the case of 19 training stations and 7 test stations, the error values of the models are 36.63, 58.31, 39.64, 41.29 and 24.68 cm, respectively. Comparison of RMSE shows that ANN model with less number of training stations provides higher accuracy than ANFIS, SVR and GRNN models. The results of this paper show that by using ANN and ANFIS models, geoid height can be estimated and used with high accuracy locally in civil and surveying applications.

Investigation and evaluation of dust and microstrip phenomena is one of the important values ​​in the management of climate and environmental hazards in the Middle East, especially in the arid, western, southern and central parts of Iran. Methods and plans for studying this phenomenon and its management are of great importance and great value. According to studies on dust phenomena based on predictive methods with low error, contradictory and low, the evaluation of the characteristics of dust and its prediction will reduce the irreparable damage that results from it. To do this, in this research, dust monitoring and assessment of its prediction in Ardebil province was performed using the ANFIS model. The data used in this study is the amount of dust in the relevant statistical period to each station from its inception until 2016. The dust phenomenon was used in the observed and predicted time intervals to assess the dust and the ANFIS model for predicting dust phenomena. According to the findings of this study, in the monitoring and prediction of dust situation, the frequency of occurrence in observed years in the maximum amount of dust in Ardabil station with 74% and the lowest in Mashgin is 8%. In the years to come, the maximum amount of dust at Khalkhal Station was 61.67% and the lowest was 10% in Mashgin. In terms of amount of dust, the Ardebil station is more intense than the rest of the stations. In terms of the severity of drought that has been studied, each of the 5 stations studied has a dust concentration of more than 74%. For the 5 stations studied for the next 18 years using manually generated codes, the stations were divided in time series, with the highest average error of training at Pars-Abad Moghan Station with 0.091% and less The highest value was obtained at the Grammy station with a value of 0.001%.

Drought is a concept that is generally understood on a basic level, but is difficult to quantify. Palmer defined a drought as a meteorological phenomenon that is characterized by ‘‘prolonged and abnormal moisture deficiency. A drought can alternatively be broadly defined as a temporary, recurring reduction in the precipitation in an area.
Aridity and drought are not synonymous. Aridity is a measure of long-term average climatic conditions. Both humid and arid regions experience droughts. However, the inter-year variation in precipitation is greater in arid regions and there is a greater probability of below average precipitation in any particular year. Arid regions are thus more prone to droughts and may experience more severe impacts from droughts.
In this research was used temperature and precipitation monthly data of Urmia, Tabriz, saghez, Maragheh, and Mahabad station in statistically period 1985-2014. Run test was used to study the homogeneity of data. Randomness and homogeneity of data was approved.at a confidence level of %95. SEPI Index and ANFIS model was used for determining and forecasting drought in Urmia lake basin. SEPI index is more complete than SPI. Results of SEPI were used in ANFIS model.
Fuzzy index SEPI[1]: Standardized precipitation index and evapotranspiration (SEPI) to address some of the disadvantages of SPI index is provided. Evapotranspiration and precipitation index SPI index and SEI standardized integration is achieved. The index is the result of drought monitoring phase of architectural models using fuzzy logic in a fuzzy inference system is designed. How to design this model and determine SEPI is described below.
Fuzzy architecture drought monitoring: for derivatization indices SPI and SEI using Fuzzy Inference System, Due to the structure of fuzzy models were considered.
SPI index[2]: Standardized Precipitation Index is an indicator widely used in Drought Monitoring. This index is one of the few indicators drought monitoring and could even say the only indicator that the time scale is considered. Depending on the time scale to determine the effect of different sources of agricultural drought, hydrological and so determined. Time scale can be determined from one month to several years. SPI index is used to calculate the only element rainy climate. Monthly precipitation amounts for each station in the desired time scale is calculated.
SEI index[3]: Since the index SPI Single Entry, rain, The SPI index values under the influence of changes in temperature and evapotranspiration parameter that is powerful factor in the drought, it will not be. So to enter the effect of temperature and evapotranspiration in SPI, SEI (evapotranspiration index Standard) To calculate this index, before any measures should reference evapotranspiration for the period to be estimated.
define the rules for combining indicators SPI and SEI: Different classes index SPI and SEI rules or the same combination of conditional statements in the form if, as a class of SEPI index in the lead, is defined. This rule only a combination of different modes SPI and SEI indices that lead to SEPI index shows. In this regard, the rules can be combined to fit different for successive written and stored in the knowledge base. Since the output of the resultant composition, indices SPI and SEI are involved in determining the status of SEPI, Weight each of the indicators with regard to the effect of precipitation and temperature parameters on the severity of the drought was considered As a result, SPI indices and weights 0.667 and 0.333, respectively SEI were included in the calculations.
According to the results, according to the research, education Anfis model with 75 percent of the data series is well done SEPI and much has been done to ensure education is nearly 100 percent. So that the graphic maximum of 0.26 percent error in saghez station on a scale of 6 months and the lowest average error of 0.10 percent in Urmia station is on a scale of 6 months. In modeling, validation data, the average error modeling is naturally higher than the average training error. Most average forecast error saghez on a scale of 6 months at the station 0.34 percent and 0.10 percent, the lowest on a scale of Urmia station is 6 months. But the coding maximum of 0.65 percent error in saghez station on a scale of 6 months and the lowest average error of 0.32 percent in Tabriz station is on a scale of 6 months. SEPI index in the time scale of 6 and 12 months is used for investigate the characteristics of adaptive neuro-fuzzy inference system in order to drought and drought forecasting model. According to the findings in this study, the frequency of drought in the stations of Urmia and Saghez and Maragheh on a scale of 6 months is more than the scale of 12 months in the basin of Lake Urmia but in Tabriz and Mahabad Stations situation is the vice versa. The drought in Urmia Lake basin is increasing trend but temperature has increasing trend with more intensity. The highest and lowest percentage of drought was seen in Urmia and Mahabad station respectively. The results of the forecasting of index by ANFIS model showed that the most training error is in Tabriz station (0.51) and the lowest training error is in Maragheh station (0.36) in a scale of 12 months in coding. In validation data modeling the average of modeling error is higher than the average training error naturally. According to the definition of drought SEPI was presented based on amounts of 0.73 or higher or mild drought to higher floors as dry conditions arise The scale of 6 months in Urmia station with 13.14 percent to 10.89 percent saghez station, Tabriz stations with 5.58 percent, with a 5.1% Mahabad station and Maragheh with the amount of 4.82 percent, the drought has occurred. The time scale of 12 months in Tabriz station by 9%, saghez station with 7.26 percent, with 6.11 percent of Urmia station, Maragheh with 5.5% and the amount of Mahabad stations with a 3.44 percent, from months of study in the series, drought has occurred.
Results of SPEI are:1.Drought trend is increasing in urmia lake basin. Temperature has increasing trend extremely.
2.The highest percentage of drought is in Urmia station and its lowest is in Mahabad station.
3.Percent of frequency of drought in Urmia station, Saghez and Maragheh on a scale of 6 months is more than to 12 months, but the scale of Tabriz and Mahabad stations with the photos. Stations Tabriz and Mahabad is in the opposite situation.
Results of ANFIS Model are:In study area and in ANFIS model whatever forecasting coming years is shorter; confidence of forecasting will be more.
Due to the errors amount obtained in model validation, in study area forecasting of drought by ANFIS model was done with confidence 94%.

Dust phenomenon is one of the most harmful natural disasters that causes major environmental impacts all over the world. In Iran، Zabol region is hardly affected by this kind of environmental disaster. Current study was made with the aim of identifying time characteristics and evaluation of dust prediction possibility in Zabol Station as the most dusty station in the country. In this regard، firstly، the statistical characteristic of the data related to frequency of monthly، seasonal and annual dusty days in Zabol station with statistics data of 41 years were studied and analyzed. Time series process analysis method has been used for definition of time fluctuations of the study element and monthly classification of the dusty days was made by using statistical multivariable cluster analysis method. Dust prediction has been done by the use of Adaptive Neuro Fuzzy Inference System (ANFIS) through allocating 70 percent of data to education and 30 percent of it to validity determination of the model. The results showed that August and July months are the dustiest months of the year during the statistical period. Based on the made cluster analysis، the months of July and August with the most dusty days have been placed in a separate cluster. The monthly، seasonal and yearly trend in this station is increasing. The prediction results of dust by ANFIS Method shows its high capability in dust prediction. Fuzzy Inference System (FIS) structure determined by four functions in arc form by hybrid training، method، predicts of dust with 93 percent reliability.