فهرست مطالب ابراهیم شهبازبیگی

Estimation and forecasting of precipitation pattern in different parts of the world, especially in arid and semi-arid regions such as Iran is very important. In addition, various numerical methods such as artificial intelligence methods due to high accuracy and speed have the ability to simulate the phenomenon of precipitation and similar subjects. The use of these techniques plays an important role in saving time and costs in field and laboratory studies. Therefore, the application and popularity of various artificial intelligence techniques to estimate and simulate different issues such as rainfall is increasing day by day. The purpose of this study is to estimate the long-term rainfall in Rasht by a hybrid ANFIS and wavelet transform model.

In this study, long-term rainfall of Rasht city was simulated using an optimum artificial intelligence model for a 62 years period from 1956 to 2017. In other words, the wavelet transform was utilized to enhance the performance of the ANFIS model and hybrid WANFIS (Wavelet-ANFIS) model was defined. Firstly, the effective lags of time series data were detected through the autocorrelation function (ACF). Then, using the lags, eight models were developed for each ANFIS and WANFIS model. It should be noted that 42 years data was applied for training and 20 years data to test the artificial intelligence models. Next, the number of optimal membership functions of ANFIS model was selected equal to two.

results of ANFIS 1 to ANFIS 8 were evaluated. Additionally, different mother wavelets were examined to optimize the ANFIS model. This means that the demy was introduced as the best mother wavelet for increasing the performance of the ANFIS model. The comparison between ANFIS and WANFIS models signified that the wavelet transform enhanced the performance of the ANFIS model. Also, results of the hybrid WANFIS models were analyzed, indicating that WANFIS 8 was the superior model. The model estimated the rainfall with an acceptable accuracy. For instance, the R, MARE and RMSE for the superior model were computed 0.961, 0.855 and 24.510, respectively. Additionally, the values of VAF and NSC for this model were respectively estimated as 92.273 and 0.913.

Results showed that (t-1), (t-2), (t-3) and (t-12) were identified as the most influenced lags for estimation of long-term rainfall of Rasht city using the hybrid WANFIS model.

In this study, the scour pattern in the vicinity of cross-vane structures with I, U and J shapes in bending channels is simulated by a new artificial intelligence method called "generalized structures group method of data handling” (GSGMDH). Initially, all the parameters affecting the scour depth in the vicinity of cross-vane structures are identified and then using these parameters, six different models are defined for each of the GMDH and GSGMDH methods. By analyzing the results yielded by the artificial intelligence models, the superior models are introduced. The GMDH and GSGMDH superior models estimate the scour values in terms of all input parameters. In addition, the accuracy of the GSGMDH models is higher than that the GMDH ones. For example, for the GMDH and GSGMDH superior models, the values of "variance accounted for" in the test mode are calculated 73.075 and 86.408, respectively. Also, the superior model forecasts the objective function values with acceptable accuracy. For example, the correlation coefficient (R), the scatter index (SI), and the Nash-Sutcliffe model efficiency coefficient (NSC) for the GSGMDH superior model in the training mode are approximated 0.913, 0.214 and 0.800, respectively. Based to the results of the sensitivity analysis, the shape factor of cross-vane structures, the ratio of the difference between the upstream and downstream flow depths to the height of the structure and the densimetric Froude number (Fd) are introduced as the most effective input parameters. An uncertainty analysis exhibits that the GSGMDH superior model has an underestimated performance.

Rapid transition from a supercritical to subcritical flow is characterized by large-scale turbulence and energy dissipation. This transition is called hydraulic jump. The hydraulic jump is a type of rapid varied flows that used for water chlorination in treatment plants and energy dissipation of the flow and other hydraulic purposes. Due to the importance and complex structure of this phenomenon, many experimental, analytical and numerical studies have been carried out in this field. In general, hydraulic jumps occur after ogee spillways. Also, stilling basins are usually situated at downstream of ogee spillways. Therefore, for proper design of the length of stilling basins, accurate determination of the hydraulic jump length has a significant importance. In this study, a hybrid model (ANFIS-PSO) is introduced which uses Adaptive Neuro Fuzzy Inference Systems (ANFIS) and Particle Swarm Optimization (PSO) for predicting the hydraulic jump length on sloping rough beds. The Adaptive neuro fuzzy inference system is a kind of artificial neural network which is based on the Takagi-Sugeno fuzzy inference system. The inference system is a set of fuzzy IF–THEN rules that have learning capability to approximate nonlinear functions. In this model, PSO is applied to enhance the performance of ANFIS by adjusting the membership functions and subsequently minimizing the error. In fact, PSO is considered as an evolutionary computational method, optimizing continues and discontinues making decision functions. Additionally, PSO is considered as a population-based search method in which each potential solution, known as a swarm, represents a particle of a population. In this approach, the particle position changes continuously in a multi-dimensional search space, until reaching the optimal response and/or computational limitations. Also, in the current study, to evaluate the performance of ANFIS-PSO models, the Monte Carlo simulation (MCs) is applied. Monte Carlo simulation is a broad class of computational algorithms that rely on repeated random sampling to obtain numerical results. Their main idea is using randomness to solve problems that might be deterministic in principle. They are often used in physical and mathematical problems and are most useful when it is difficult or impossible to use other approaches. The Monte Carlo simulation is mainly used in different problems such as optimization and numerical integration from a probability distribution. Also, in this paper, the k-fold Cross Validation (k=4) is used for examination of the models ability. In k-fold Cross Validation, the original sample is randomly separated into k equal size sub-samples. In k sub-samples, a single sub-sample is retained as the validation data for testing the model, and the remaining k-1 sub-samples are used as training data. The cross-validation process is then repeated k times (the folds), with each of the k sub-samples used exactly once as the validation data. The k results from the folds can then be averaged to produce a single estimation. The advantage of this method over repeated random sub-sampling is that all observations are used for both training and validation, and each observation is used for validation exactly once. At first, five ANFIS-PSO models are defined using effective parameters on length of hydraulic jump. To validate the ANFIS-PSO models, the Kumar and Lodhi’s (2016) experimental measurements were used. The experimental model was conducted in a rectangular channel with a length of 8.0 m, 0.60 m width and 0.60 m depth. Three slopes of a flume, viz. 0.005, 0.010 and 0.016, were observed. Next, by analyzing the ANFIS-PSO models results, the superior model is presented. The superior model predicts the length of hydraulic jump in terms of Froude Number (F1), ratio of roughness bed (Ks/h1), ratio of sequent depth (h2/h1) and slope bed (S0). This model simulates the experimental measurement with suitable accuracy. For superior model, the Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE) and the correlation coefficient (R) were respectively computed equal to 3.750, 0.688 and 0.984. In addition, the scatter index (SI) for superior model was estimated equal to 0.055. Also, in order to more examine the ANFIS-PSO models results, the ratio of the predicted jump length to the observed jump length (λ=(Lr/h1)(Predicted)-(Lr/h1)(Observed)) was introduced. For the superior model, the average of this ratio was calculated equal to 1.003. According to numerical models results, the Froude number is identified as the most effective parameter for modeling the length of the hydraulic jump.