Kadir Erkan

Comparison of Kalman Filter and Wavelet Filter for Denoising

This paper presents denoising the signal using wavelet filter and Kalman filter. The noise is zero mean and the variance value is 0.001. Kalman filter removes disturbances or faults from the signal by using initialization and propagation of error covariance statistics. Implementation of Kalman filter is impractical in large scale models as shown for the oscillator system. As an alternative wavelet filter has been used for the same system. Coiflet 2 which is orthogonal wavelet has been used. Soft thresholding has been applied. Decomposition is performed at level 9. The results of wavelet filter and Kalman filter are shown. Response of wavelet filter is better when compared with Kalman filter result

Using Current Mode Fuzzy Gain Scheduling Of PI Controller for UPS Inverter

This paper presents control of UPS inverter using fuzzy gain scheduling of PI controller. This control scheme has double loop current mode control scheme in core. This scheme includes two control loops as inner and outer control loops. Inner control loop uses inductor current of inverter filter as feedback while outer control loop uses inverter output voltage as feedback. Fuzzy logic controller (FLC) is used to adjust the voltage loop PI controller parameters. The voltage error and its derivative are used as input variables of the FLC. In this study, current mode fuzzy gain scheduling of PI controller are simulated digitally using PSIM and C++ under linear, nonlinear and fluorescent loads. And the results show that current mode fuzzy gain scheduling of PI controller can provide low THD and good regulation quality especially under nonlinear and fluorescent loads

Experimental Autotuning PID Control of Temperature Using Microcontroller

Experimental autotuning proportional-integral-derivative (PID) temperature control of oven using microcontroller is presented in this paper. Different types of autotuning PID controller methods have been applied to the oven designed as an experiment set. This experiment set is an FODPT (first order plus dead time) system. Relay (PI, PID) and integral square time error criterion (ISTE) set point (PI, PID) are used as autotuning PID method. To be able to control this system a digital signal processing card is designed using PIC 17C44 as microcontroller and ADS1212 as A/D converter. And the results are discussed to define which controller is the best for the oven

Soft computing and signal processing based active fault tolerant control for benchmark process

Ideally, when faults happen, the closed-loop system should be capable of maintaining its present operation. This leads to the recently studied area of fault-tolerant control (FTC). This paper addresses soft computing and signal processing based active FTC for benchmark process. Design of FTC has three levels: Level 1 comprises a traditional control loop with sensor and actuator interface and the controller. Level 2 comprises the functions of online fault detection and identification. Level 3 comprises the supervisor functionality. Online fault detection and identification has signal processing module, feature extraction module, feature cluster module and fault decision module. Wavelet analysis has been used for signal processing module. In the feature extraction module, feature vector of the sensor faults has been constructed using wavelet analysis, sliding window, absolute maximum value changing ratio and variance changing ratio as a statistical analysis. For the feature cluster module, the self-organizing map (SOM), which is a subtype of artificial neural network has been applied as a classifier of the feature vector. As a benchmark process three-tank system has been used. Control of the three-tank system is provided by fuzzy logic controller. Faults are applied to three level sensors. Sensor faults represent incorrect reading from the sensors that the system is equipped with. When a particular fault occurs in the system, a suitable control scheme has been selected on-line by supervisor functionality to maintain the closed-loop performance of the system. Active FTC has been achieved by switch mode control using fuzzy logic controller. Simulation results show that benchmark process has maintained acceptable performance with FTC for the sensor faults. As a result, when the system has sensor faults soft computing and signal processing based FTC helps for the best performance of the system.

Microcontroller based temperature control of oven using different kinds of autotuning PID methods

This paper presents microcontroller based autotuning proportional-integral-derivative (PID) controller for an oven designed as an experiment set. Different types of autotuning PID controller methods have been examined. Proportional, P, control method has been applied first. Relay and integral square time error criterion (ISTE) tuning methods are used as autotuning PID method. For relay tuning method, proportional (P), proportional-integral (PI) and proportional-integral-derivative (PID) and for ISTE disturbance (PI, PID) have been used. These methods have been applied to the experiment set which is an FODPT (First Order Plus Dead Time) system. To be able to control this system a digital signal processing card is designed. PIC17C44 is used as microcontroller and ADS1212 is used as A/D converter. And the results are discussed to define which controller is the best for this experiment set.

Implementation of intelligent active fault tolerant control system

This paper addresses implementation of intelligent active fault tolerant control for experiment set which is a FODPT (First Order Plus Dead Time) system. Temperature control has been done using fuzzy logic controller (FLC). Faults often cause undesired reactions, so to keep the system stable and acceptable control performance is an important problem for control system design. In this paper, multiplicative, additive types of sensor faults have been examined and disturbance has been applied for temperature sensor as a fault. Feature vectors of the sensor faults have been constructed using wavelet analysis, sliding window and a statistical analysis. Classifier of the feature vectors has been done using The Self Organizing Map (SOM). Switch mode control with fuzzy logic controller chosen by supervisor has been used for reconfiguration. When a fault occurs in the system, a suitable controller has been selected online to maintain the closed-loop performance of the system.

Comparison of wavenet and neuralnet for system modeling

This paper presents nonlinear static and dynamic system modeling using wavenet and neuralnet. Wavenet combines wavelet theory and feed-forward neuralnet, so learning approach is similar to neuralnet. The selection of transfer function is crucial for the approximation property and the convergence of the network. The purelin and the tansig functions are used as the transfer functions for neuralnet and the first derivative of a gaussian function is used as the transfer function for wavenet. Wavenet and neuralnet parameters are optimized during learning phase. Selecting all initial values random, but for wavenet, it may be unsuitable for process modeling because wavelets have localization feature. For this reason heuristic procedure has been used for wavenet. In this study gradient methods have been applied for parameters updating with momentum. Error minimization is computed by quadratic cost function for wavenet and neuralnet. Nonlinear static and dynamic functions have been used for the simulations. Recently wavenet has been used as an alternative of the neuralnet because interpretation of the model with neuralnet is so hard. For wavenet learning approach, training algorithms require smaller number of iterations when compared with neuralnet. Consequently, according to the number of training iteration and TMSE, dynamic and static system modeling with wavenet is better as shown in results.

Reactor Controller Design Using Genetic Algorithms with Simulated Annealing

This chapter presents a digital control system for ITU TRIGA Mark-II reactor using genetic algorithms with simulated annealing. The basic principles of genetic algorithms for problem solving are inspired by the mechanism of natural selection. Natural selection is a biological process in which stronger individuals are likely be winners in a competing environment. Genetic algorithms use a direct analogy of natural evolution. Genetic algorithms are global search techniques for optimisation but they are poor at hill-climbing. Simulated annealing has the ability of probabilistic hill-climbing. Thus, the two techniques are combined here to get a fine-tuned algorithm that yields a faster convergence and a more accurate search by introducing a new mutation operator like simulated annealing or an adaptive cooling schedule. In control system design, there are currently no systematic approaches to choose the controller parameters to obtain the desired performance. The controller parameters are usually determined by test and error with simulation and experimental analysis. Genetic algorithm is used automatically and efficiently searching for a set of controller parameters for better performance.

A Hybrid Implementation of Genetic Algorithm for Path Planning of Mobile Robots on FPGA

This paper proposes a hybrid design and implementation of Genetic Algorithm (GA) for the path planning of mobile robots on a Field Programmable Gate Array (FPGA). GAs have been widely used to generate an optimal path by taking the advantage of its strong optimization ability; however, GA’s computation time may be longer for complex problems. Especially, calculation of the fitness function takes a long time. A solution to accelerate it is to implement the GA in hardware. Intellectual Property (IP) hard core provides faster computation. In this study, fitness function of the GA is implemented on IP hard core while the other operators of GA run on a Microblaze soft processor. The experimental results showed that the fitness module by IP hard core can run 98.95 times faster than the fitness module by the Microblaze soft processor. The overall performance of the GA is accelerated 37.5 % by hybrid implementation with both hard and soft cores. We used the Pioneer P3-DX Mobile Robot and Xilinx XUPV5-LX110T FPGA device.