جستجوی مقالات مرتبط با کلیدواژه « adaptive neuro » در نشریات گروه « فنی و مهندسی »

The purpose of this research is to perform experimental study and modeling of the plastic deformation of rectangular plates under the low rate impact loading by drop hammer system. In experimental section, some experiments were conducted on rectangular plates with different levels of energy in order to survey the mechanical behavior of steel and aluminum plates according to applied load. The modeling section consists of presenting an explicit function for experimental data by singular value decomposition (SVD) based on non-dimensional parameters and also multi-objective modeling and design of adaptive neuro-fuzzy inference system (ANFIS) by genetic algorithm. Generally, the aim of modeling is a reliable and satisfactory prediction of deflection – thickness ratio of plates under impact loads. A comparison between modeling results and experimental data is done in order to validate the results. The investigation of training and prediction data errors which has been based on root-mean-square error (RMSE) and coefficient of determination (R2) shows that the obtained results of the optimal design of ANFIS is closer to experimental results than mathematical modeling by SVD, with the exception that a mathematical function based on experimental data is presented by SVD method. Therefore, using these presented models for deflection-thickness ratio of plate under impact loading is desirable.

Groundwater is a major source of water supply for domestic, agricultural, and industrial uses; hence, its quality modeling is an important task in hydro-environmental studies. While many data-based models have been developed for this purpose, the performance of such data-based models can be drastically enhanced if they are based on temporal and spatial pre-processing. In this study, geostatistics tools (e.g., Co-Kriging), as spatial estimators, and self-organizing map (SOM), as a clustering technique, were employed in conjunction with Adaptive Neuro-Fuzzy Inference System (ANFIS) for the temporal forecasting of such quality parameters as electrical conductivity (EC) and total dissolved solids (TDS) of the groundwater in Ardabil Plain. Using the results thus obtained, the impact of spatial data clustering was also investigated on the same parameters. The results showed that, if propoer input data are selected, the proposed spatial clustering technique is capable of imporving groundwater quality forecasts made by ANFIS.

In this paper, an intelligent robust controller is proposed for a class of nonlinear systems in presence of uncertainties and bounded external disturbances. The proposed method is based on a combination of terminal sliding mode control and adaptive neuro-fuzzy inference system with bee’s algorithm training. For this purpose, a sliding surface is firstly designed based on terminal sliding control method. This sliding surface is considered as input for the intelligent controller which is an adaptive neuro-fuzzy inference system and using it, terminal sliding mode control law without the switching part is approximated. In the proposed method, an intelligent bee’s algorithm is also used for updating the weights of the adaptive neuro-fuzzy inference system. Compared with fast terminal sliding mode control, the proposed controller provides advantages such as robustness against uncertainty and disturbance, simplicity of controller structure, higher convergence speed compared with similar conventional methods and chattering-free control effort. The method is applied to an atomic force microscope for nano manipulation. The simulation results show the robustness and effectiveness of the proposed method.

Today, fast and accurate fault detection is one of the major concerns in the industry. Although many advanced algorithms have been implemented in the past decade for this purpose, they were very complicated or did not provide the desired results. Hence, in this paper, we have proposed an emerging method for deep groove ball bearing fault diagnosis and classification. In the first step, the vibration test signals, related to the normal and faulty bearings have been used for both of the drive-end and fan-end bearings of an electrical motor. After that, we have employed the one dimensional Meyer wavelet transform for signal processing in the frequency domain. Hence, the unique coefficients for each kind of fault were extracted and directed to the adaptive neuro-fuzzy system for fault classification. The intelligent adaptive neuro-fuzzy system was adopted to enhance the fault classification performance due to its flexibility and ability in dealing with uncertainty and robustness to noise. This system classifies the input data to the faults in the race or the balls of each of the fan-end and the drive-end bearings with specific fault diameters. In the final part of this study, the new experimental signals were processed in order to verify the results of the proposed method. The results reveal that this method has more accuracy and better classification performance in comparison with other methods, proposed in the literature.

Nowadays, powerful detection systems, which the learning procedure of them is black box and is not available, have been widely used to classify data. However, the understandability of the acquired knowledge from these detection systems can significantly help operator in carrying out classification performance with high accuracy and precision. Hence, knowledge acquisition in a form of fuzzy rule set is an important issue in the image processing that causes to comprehend the classification methods appropriately and to improve them subsequently. The purpose of this paper is proposing a method to extract fuzzy rules in IF-THEN form via an Adaptive Neuro-Fuzzy Inference System (ANFIS) for classification LiDAR data and digital aerial images. Detection of building and tree in urban areas needs to determine some features to perform the detection procedure; because classification algorithms decide about pixel entity based on its feature vector. These features can make the object separation possible by the textural, the spectral, and the structural characteristics. Nowadays, by increasing the number of the active and passive sensors, it is possible to record the textural, the spectral, and the structural characteristics of objects in different wavelengths by various approaches. In this paper, some potentially features were generated, and then optimal features were selected using the genetic algorithm. Using the selected optimum features, an ANFIS was used to recognize the objects accurately. In this regard, at first, the prepared training data was utilized as inputs of grid partitioning algorithm and a Sugeno fuzzy inference system with one output was generated by determining the type and the number of input membership functions, as well as theS type of output membership functions. Then, the grid partitioning algorithm figured out the best state of the membership functions after investigating the whole of the possible states. Afterwards, the training and checking data were entered into the generated ANFIS and during the training procedure, the final classifier was concluded to detect buildings and trees. Finally, by proposing a different fuzzy-based method and using the selected training data, as well as the output membership functions of the proposed ANFIS, a set of effective fuzzy rules were extracted. The proposed method has three main steps. In the first step, the tuned premise parameters (after training process) of inputs training data of ANFIS were extracted according to the mean values of the membership functions. In the second step, firstly, based on the number of membership functions of each feature, the total number of feasible fuzzy rules was determined. Then, for each training data, the fired values for all rules were computed. The rule that had the most effect in the process was chosen as the fired fuzzy rule of each training data related to the desired object class. In the third step, the fuzzy rules which has the importance more than a specified threshold in the classification procedure were extracted. The extracted fuzzy rules were considered and analyzed logically regarding the feature layers, and the results show the high capability of the fuzzy-based proposed method in extracting rules from the objects detection procedure.

In this paper, an intelligent integral terminal sliding mode control method with a new sliding surface is proposed based on adaptive neural-fuzzy inference. First, a terminal sliding mode controller using a novel sliding surface is designed based on Lyapunov’s stability theorem for controlling a class of chaotic systems in presence of uncertainty and disturbance. The proposed sliding surface is a combination of the conventional terminal sliding surface and integral of a nonlinear function of the states of the system. The purpose of choosing this surface includes achieving appropriate response speed, removing chattering and robustness against external disturbance. Then, we assume that a nonlinear part of the system is unknown and only input-output data is available. Therefore, an adaptive neuro-fuzzy inference system is used to approximate the unknown part of the system dynamics. Finally, in order to enhance the performance of the proposed method, the honey bee algorithm is utilizd for selecting the coefficients of integral terminal sliding mode controller. The simulation results show the effectiveness of the controller due to the improved speed, removed chattering, appropriate transient response and satisfactory performance in the presence of uncertainties in the system model.

Drought is the interaction between environment and water cycle in the world and affects natural environment of an area when it persists for a longer period. So، developing a suitable index to forecast the spatial and temporal distribution of drought plays an important role in the planning and management of natural resources and water resource systems. In this article، firstly، the drought concept and drought indexes were introduced and then the fuzzy neural networks and fuzzy C-mean clustering were applied to forecast drought via standardized precipitation index (SPI). The results of this research indicate that the SPI index is more capable than the other indexes such as PDSI (Palmer Drought Severity Index)، PAI (Palfai Aridity Index) and etc. in drought forecasting process. Moreover، application of adaptive nero-fuzzy network accomplished by C-mean clustering has high efficiency in the drought forecasting.

Estimation of the friction coefficient in pipes is very important in many water and wastewater engineering issues, such as distribution of velocity and shear stress, erosion, sediment transport and head loss. In analyzing these problems, knowing the friction coefficient, can obtain estimates that are more accurate. In this study in order to estimate the friction coefficient in pipes, using adaptive neuro-fuzzy inference systems (ANFIS), grid partition method was used. For training and testing of neuro-fuzzy model, the data derived from the Colebrook’s equation was used. In the neuro-fuzzy approach, pipe relative roughness and Reynolds number are considered as input variables and friction coefficient as output variable is considered. Performance of the proposed approach was evaluated by using of the data obtained from the Colebrook’s equation and based on statistical indicators such as coefficient determination (R2), root mean squared error (RMSE) and mean absolute error (MAE). The results showed that the adaptive nerou-fuzzy inference system with grid partition method and gauss model as an input membership function and linear as an output function could estimate friction coefficient more accurately than other conditions. The new proposed approach in this paper has capability of application in the practical design issues and can be combined with mathematical and numerical models of sediment transfer or real-time updating of these models.

Intelligent methods are widely used in fault detection and diagnosis in electrical machine, transformers and in general in all parts of industry. One of the most advanced methods in fault detection is fuzzy Inference system. Since, extraction of effective rules is so hard in complicated problems, using fuzzy Inference is impossible. In this paper, in order to determine the amount of inter-turn stator winding fault in Permanent Magnet Synchronous Motor (PMSM),the fuzzy logic set has been used. This research has been fulfilled in two steps. In first step, a mathematical model of PMSM under fault condition has been simulated in MATLAB/SIMULINK and required data; deferential current of each phase and speed of motor have been extracted, in second step, a fuzzy set has been created using Adaptive Neuro-Fuzzy Inference System (ANFIS) and this fuzzy set has been used in simulation for on-line test. Obtained results show that proposed method can follow fault in short time and can present amount of fault very precise.

Fracture energy of concrete is one of the basic parameters of fracture that present the concrete cracking resistance and also is one of the important characteristics in considering design of concrete engineering structures. In recent years، fracture parameters of concrete have been investigated using various experimental methods; and the role of these parameters in design of structures is an important issue. In this paper، a fracture model based adaptive neuro-fuzzy inference system (ANFIS) has been implicated to estimate the fracture parameter of concrete GF (specific fracture energy i. e. the area under the stress- displacement curve) using a three-point bending (3PB) specimen. The results showed that the adaptive neuro-fuzzy inference system and its proper training can be used in order to create an optimal model for each series of data and can be applied to evaluate the adaptive neuro-fuzzy inference system reliability as an effective tool to estimate the fracture energy of concrete.