Hasan Erdal

Optimization of PID Controllers Using Ant Colony and Genetic Algorithms

Artificial neural networks, genetic algorithms and the ant colony optimization algorithm have become a highly effective tool for solving hard optimization problems. As their popularity has increased, applications of these algorithms have grown in more than equal measure. While many of the books available on these subjects only provide a cursory discussion of theory, the present book gives special emphasis to the theoretical background that is behind these algorithms and their applications. Moreover, this book introduces a novel real time control algorithm, that uses genetic algorithm and ant colony optimization algorithms for optimizing PID controller parameters. In general, the present book represents a solid survey on artificial neural networks, genetic algorithms and the ant colony optimization algorithm and introduces novel practical elements related to the application of these methods to process system control.

Trajectory tracking performance comparison between genetic algorithm and ant colony optimization for PID controller tuning on pressure process

The main goal of this study was to compare the performances of genetic algorithm (GA) and ant colony optimization (ACO) algorithm for PID controller tuning on a pressure control process. GA and ACO were used for tuning of the PID controller when predefined trajectory reference signal was applied. Offline learning approach was employed in both GA and ACO algorithms. Realized pressure process dynamic has nonlinear behavior, thus system was modeled by nonlinear auto regressive and exogenous input (NARX) type artificial neural network (ANN) approach. PID controller was also tuned by Ziegler–Nichols (Z–N) method to compare the results. A cost function was design to minimize the error along the defined cubic trajectory for the GA‐PID and ACO‐PID controller. Then PID controller parameters (Kp, Ki, Kd) were found by GA‐PID, ACO‐PID algorithms, which were adjusted with their optimal parameters. It was concluded that both ACO and GA algorithms could be used to tune the PID controllers in the pressure process with excellent performance. This material is suitable for an engineering course on neural networks, genetic algorithm, ant colony optimization and process control laboratory.

A study to classify Non-Dipper/Dipper blood pressure pattern of type 2 diabetes mellitus patients without Holter device

The aim of this study was to design an expert system to predict the Non-Dipping or Dipping pattern by using several basic clinical and laboratory data through an artificial intelligence algorithm. Data Mining is a technique which extracts information from data sets by using a combination of both statistical analysis methods and artificial intelligence algorithms. Also in this study, the decision tree and naivebayes classification algorithms of this technique were used. Firstly, sixty-five patients (mean age 51±7 years, 40 females,) were included in the study. Systolic and diastolic dipping were found in 13 and 15 % of the patients, respectively. In the advancing process of the experiment, the number of instances were reduced, because of some missing data of the patients. The data sets were tested using the J48 decision tree algorithm. This classification algorithm was implemented on 56 instances, and also the number of attributes was reduced from 35 to 23. 66 % of the instances (37) were reserved for training and 44 % of the instances (19) were reserved for testing. When the algorithm was run, the Non-Dipper/Dipper pattern of the instances were correctly predicted in a rate of 73.6842 %. Model was built in 0.02 seconds. This pilot study shows that a machine learning algorithm can help in the prediction of diurnal blood pressure pattern relying on some basic demographic, clinical and laboratory data, with a reasonable accuracy.

Ant Colony Optimization (ACO)

The ant colony optimization algorithm (ACO) is an evolutionary meta-heuristic algorithm based on a graph representation that has been applied successfully to solve various hard combinatorial optimization problems. Initially proposed by Marco Dorigo in 1992 in his PhD thesis [49], the main idea of ACO is to model the problem as the search for a minimum cost path in a graph. Artificial ants walk through this graph, looking for good paths. Each ant has a rather simple behavior so that it will typically only find rather poor-quality paths on its own. Better paths are found as the emergent result of the global cooperation among ants in the colony [13, 15, 50-52].

System control through the internet and a remote access laboratory implementation

Nowadays, remote access laboratories are getting popular in technical education. This paper illustrates a Remotely Access Control Laboratory (RACL) implementation in which students can make control experiments on a real plant through the Internet. Process Level Trainer is used as a plant, which is a practice environment for the fundamentals of the process control education. A Data Acquisition Card (DAQ Card) installed computer, as a control unit, is connected to the plant. A pan-tilt movable, CCD, IP Camera is preferred to transmit the look of the plant during the experiment. In order to control the plant through the Internet, special control software is developed using Delphi programming language and web pages including PHP scripts are designed. A Flash object is also developed for displaying the result graphics of experiments online. A user, after login from any computer that has an Internet connection, can easily study basic control algorithms, such as Open Loop, On-Off, PID with changing their parameters as well.

Modeling of hybrid wind‐gas power generation system and adaptive neuro‐fuzzy controller to improve the system performance

In this study, dynamic modeling of the hybrid wind‐gas power generation system (HWGPGS) is realized by using the MATLAB/Simulink program for purpose of meeting the electric energy needs of small settlement units far from city centers or energy distribution networks. Besides, Adaptive Neuro‐Fuzzy Inference System (ANFIS) is used to ensure electrical output magnitudes of the hybrid power generation system at a desired operating performance. The components that build up the hybrid power generation system and its functions in the system are explained. In case the HWGPGS is loaded with different consumer loads, analysis of electrical magnitudes of the system is made through simulation results. As it can be seen from the results of simulation study realized for the hybrid power generation system, output electrical magnitudes of the ANFIS controlled system reach to desired operating values in a short time.

Interface design for automation of the scanning EMG method

Scanning EMG is a method developed for examining the electro-physiological cross section and the size of the motor unit of a human muscle. Electrical specifications of the motor unit can be obtained as well as anatomical distribution of muscle fibers and pathological changes between different muscles can be examined by the help of this method. In this paper; an automation system which is designed for the execution of scanning EMG method, whether manually or automatically, and a user-interface are described. Parameters like step count and step size which are about the movement of an electrode, moved by a linear actuator, through muscle fibers can be defined as reference by user via designed interface. As a result, acquired signals are digitalized by data-acquisition card (DAQ) and saved as text file for the future signal process tasks.

Liquid level process control with fuzzy logic based embedded system

In this study, the design of embedded Sugeno fuzzy logic based controller for controlling non-linear liquid level process is realized. First, the system was activated on Matlab-Simulink platform and uploaded to Arduino Mega. With this study, activation of modern control methods on embedded systems is simplistically proven. When the control system operation was checked out, it has been observed that it was pretty successful.

Performance evaluation of classification algorithms by excluding the most relevant attributes for dipper/non-dipper pattern estimation in Type-2 DM patients

Diabetes Mellitus (DM) is a high prevalence disease that causes cardiovascular morbidity and mortality. On the other hand, the absence of physiologic night-time blood pressure decrease can further lead to morbidity problems such as target organ damage both in diabetics and non-diabetics patients. However, the Non-dipping pattern can only be measured by the 24-hour ambulatory blood pressure monitoring (ABPM) device. ABPM has certain challenges such as insufficient devices to distribute to patients, lack of trained staff or high costs. Therefore, in this study, it is aimed to develop a classifier model that can achieve a sufficiently high accuracy percentage for Dipper/non-Dipper blood pressure pattern in patients by excluding ABPM data. The study was conducted with 56 Turkish patients in Marmara University Hypertension and Atherosclerosis Center and School of Medicine Department of Internal Medicine, Division of Endocrinology between the years 2010 and 2012. Our purpose was to find out if the proposed method would be able to detect non-dipping/dipping pattern through various data mining algorithms in WEKA platform such as J48, NaiveBayes, MLP, RBF. All algorithms were run to get accurate Dipper/non-Dipper pattern estimation excluding the attributes of ABPM data. The results show that Neural Network (MLP and RBF) algorithms mostly produced reasonably high classification accuracy, sensitivity and specificity percentages reaching up to 90.63% when the attributes were reduced. However in medical sciences, sensitivity is taken as a valid and reliable indication for diagnosis. Therefore, MLP had a highersensitivity percentage (83.3%) than others. Also, ROC values, which had the closest values to 1, were achieved by RBF for each selection mode. ROC was 0.872 for 10 fold CV mode and 0.856 for percentage split mode. Finally, ANN MLP and RBF algorithms were used, and was observed that RBF algorithm had the highest success rate in terms of sensitivity that was 83.3%. In medical diagnosis, a higher sensitivity performance is regarded as more valid indication of metric than a higher specificity. The proposed model could represent an innovative approach that might simplify and fasten the diagnosis process by skipping some steps in Dipper/non-Dipper diagnosis/prognosis.

An Application for Process System Control

The purpose method is implemented to stabilize the pressure of the tank at the desired pressure level adjusting the input air flow despite the continuous exhaust output flowing as a disturbance. Because of compressibility of air and nonlinear characteristic of valves, realized system has nonlinear dynamics. Cubic trajectory function was used as an input reference signal, to prevent the pressure fluctuations and large overshoot in tank which could be harmful in some process [56].