Sürdürülebilir Yeşil Kampüs Koleksiyonu / Sustainable Green Campus Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7755
Browse
3 results
Search Results
Master Thesis Data Driven Modeling Using Reinforcement Learning in Autonomous Agents(Izmir Institute of Technology, 2003) Karakurt, Murat; Özdemir, Serhan; Özdemir, Serhan; 03.10. Department of Mechanical Engineering; 03. Faculty of Engineering; 01. Izmir Institute of TechnologyThis research has aspired to build a system which is capable of solving problems by means of its past experience, especially an autonomous agent that can learn from trial and error sequences. To achieve this, connectionist neural network architectures are combined with the reinforcement learning methods. And the credit assignment problem in multi layer perceptron (MLP) architectures is altered. In classical credit assignment problems, actual output of the system and the previously known data in which the system tries to approximate are compared and the discrepancy between them is attempted to be minimized. However, temporal difference credit assignment depends on the temporary successive outputs. By this new method, it is more feasible to find the relation between each event rather than their consequences.Also in this thesis k-means algorithm is modified. Moreover MLP architectures is written in C++ environment, like Backpropagation, Radial Basis Function Networks, Radial Basis Function Link Net, Self-organized neural network, k-means algorithm.And with their combination for the Reinforcement learning, temporal difference learning, and Q-learning architectures were realized, all these algorithms are simulated, and these simulations are created in C++ environment.As a result, reinforcement learning methods used have two main disadvantages during the process of creating autonomous agent. Firstly its training time is too long, and too many input parameters are needed to train the system. Hence it is seen that hardware implementation is not feasible yet. Further research is considered necessary.Master Thesis Artificial Neural Networks and Fuzzy Logic Applications in Modeling the Compressive Strength of Portland Cement(Izmir Institute of Technology, 2004) Can, Sever; Akkurt, Sedat; Akkurt, Sedat; 03.09. Department of Materials Science and Engineering; 03. Faculty of Engineering; 01. Izmir Institute of TechnologyPortland cement production is a complex process that involves the effect of several processing parameters on the quality control of 28-day cement compressive strength (CCS). There are some chemical parameters like the C3S, C2S, C3A, C4AF, and SO3 contents in addition to the physical parameters like Blaine (surface area) and particle size distribution. These factors are all effective in producing a single quantity of 28-day CCS. The long duration of 28 day CCS test provided the motivation for research on predictive models. The purpose for these studies was to be able to predict the strength instead of waiting for 28 days for the test to be complete. In this thesis, artificial intelligence (AI) methods like artificial neural networks (ANNs) and fuzzy logic were used in the modeling of the 28-day CCS. The two models were compared for their quality of fit and for the ease of application.Quality control data from a local cement plant were used in the modeling studies. The data were separated randomly into two parts: the first one contained 100 data points to be used in training and the second part had 50 data points to be used in testing stages of the models. In this study, four different AI models were created and tested (3 ANN, 1 fuzzy logic). One of the ANN models (Model A) had 20 input parameters in 20x20x1 architecture with testing average absolute percentage error (AAPE) of 2.24%. The other ANN model (Model B) had four input parameters (SO3, C3S, Blaine and total alkali amount) in 4x4x1 architecture with AAPE of 2.41%. Both of the Model A and the Model B were created in the MatLAB® environment by writinga custom computer code. The last ANN model (Model C) actually refers to 72 differentANN models created in the MatLAB® neural networks toolbox. In order to obtain a model with the lowest error, different learning algorithms, training functions and architectures in combinations were tested. The lowest AAPE among these models appeared to be 2.31%. The fuzzy logic model (Model D) which had four input parameters (SO3, C3S, Blaine and total alkali amount) was created in the MatLAB fuzzy logic toolbox. In order to write the fuzzy rules, the sensitivity analysis of the Model B was utilized. The AAPE of the Model D was 2.69%. The model was compared with the ANN models for its error levels and ease of application. The results indicated that through the application of fuzzy logic algorithm, a more user friendly andmore explicit model than the ANNs could be produced within successfully low error margins.Master Thesis Artificial Neural Networks Model for Air Quality in the Region of Izmir(Izmir Institute of Technology, 2002) Birgili, Savaş; Tayfur, Gökmen; Tayfur, Gökmen; 03.03. Department of Civil Engineering; 03. Faculty of Engineering; 01. Izmir Institute of TechnologyIn this study, a systematic approach to the development of the artificial neural networks based forecasting model is presented. S02, and dust values are predicted with different topologies, inputs and transfer functions. Temperature and wind speed values are used as input parameters for the models. The back-propagation learning algorithm is used to train the networks. R 2 (correlation coefficient), and daily average errors are employed to investigate the accuracy of the networks. MATLAB 6 neural network toolbox is used for this study. The study results indicate that the neural networks are able to make accurate predictions even with the limited number of parameters. Results also show that increasing the topology of the network and number of the inputs, increases the accuracy of the network. Best results for the S02 forecasting are obtained with the network with two hidden layers, hyperbolic tangent function as transfer function and three input variables (R2 was found as 0,94 and daily average error was found as 3,6 j..lg/m3).The most accurate results for the dust forecasting are also obtained with the network with two hidden layer, hyperbolic tangent function as transfer function and three input variables (R2 was found as 0,92 and daily average error was found as 3,64 j..lg/m3).S02 and dust predictions using their last seven days values as an input are also studied, and R2 is calculated as 0,94 and daily average error is calculated as 4,03 Jlg/m3 for S02 prediction and R2 is calculated as 0,93 and daily average error is calculated as 4,32 Jlg/m3 for dust prediction and these results show that the neural network can make accurate predictions.