Effective parameter tuning for genetic algorithm to solve a real world transportation problem

Numerical modeling of experimental human fibrous cap delamination

Noise injection disturbs the scheduling of program threads in order to increase the probability that more of their different legal interleavings occur during the testing process. However, there exist many different types of noise heuristics with many different parameters that are not easy to set such that noise injection is really efficient. In this paper, we propose a new way of using genetic algorithms to search for suitable types of noise heuristics and their parameters. This task is formalized as the test and noise configuration search problem in the paper, followed by a discussion of how to represent instances of this problem for genetic algorithms, which objectives functions to use, as well as parameter tuning of genetic algorithms when solving the problem. The proposed approach is evaluated on a set of benchmarks, showing that it provides significantly better results than the so far preferred random noise injection.

To tackle several combinatorial optimization problems, genetic algorithms were frequently used as powerful evolutionary methods. However, performance of these algorithms is heavily dependent on finding appropriate parameters and operators. This phenomenon is known as the parameter tuning problem. The latter falls into the category of hard problems. Therefore, we develop a heuristic approach for effective parameter tuning of genetic algorithms. We also focus on determining the most appropriate genetic operators. Our approach detects the promising available set of parameters. For the practical demonstration, we tackle the sum coloring of graphs. Results show the practical performance of the proposed method to enhance the genetic algorithm.

Five-phase induction motors have the characteristics of high torque density, low torque ripple, and flexible control, making them suitable for medium- and low-voltage power supply situations. However, with the expansion of application scenarios, five-phase motors need to cope with increasingly complex operating conditions. Five-phase motors for propeller propulsion will face various complex sea conditions during actual use, and five-phase motors for electric vehicles will also face various complex road conditions and operating requirements during use. Therefore, as a propulsion motor, its speed control system must have strong robustness and anti-disturbance performance. The use of traditional PI algorithms has problems, such as poor adaptability and inability to adapt to various complex working conditions, but the use of an active disturbance rejection controller (ADRC) can effectively solve these problems. However, due to the significant coupling between the variables of induction motors and the large number of parameters in the ADRC, tuning the parameters of the ADRC is complex. Traditional empirical tuning methods can only obtain a rough range of parameter values and may have significant errors. Therefore, this paper uses ADRC based on genetic algorithm(GAADRC) to tune the parameters of the control and design an objective function based on multi-objective optimization. The parameters to be adjusted were obtained through multiple iterations. The simulation and experimental results indicate that GAADRC has lower startup overshoot, faster adjustment time, and lower load/unload speed changes compared to the empirically tuned PI controller and ADRC. Meanwhile, using a genetic algorithm for motor ADRC parameter tuning can obtain optimal control parameters while the control parameter range is completely uncertain; therefore, the method proposed in this paper has strong practical value.

A tuning algorithm for model predictive controllers based on genetic algorithms and fuzzy decision making

In Monte Carlo Markov Chain theory, a particle on a Markov chain is run for a ’long time’ until the particle is nearly in the stationary distribution. Coupling from the past is a methodology which allows samples to be taken exactly from the stationary distribution of the chain, but which requires that we have a completely coupling chain. One chain often used for the ferromagnetic Ising model is the Swendsen-Wang chain. We give the first method for running Swendsen-Wang as a completely coupling chain, thereby giving a procedure for taking exact samples from the ferromagnetic Ising model using Swendsen-Wang. Moreover, this chain completely couples in polynomial time when a parameter of the chain is inversely proportional to the maximum degree of the graph, or when this parameter is close to 1. This has two implications. First, the Swendsen-Wang chain is rapidly mixing over this set of parameter values, a fact recently (and independently) shown by Cooper and Frieze using path coupling. Second, over this set of parameter values the exact sampling procedure will have an expected running time which is polynomial. ∗This work was partially supported by an Office of Naval Research Fellowship and NSF grants CCR-9307391, CCR-9700029, and DMS-9505155, and ONR grant N00014-96-1-00500

Optimal tuning of power system stabilizer (PSS) parameters using genetic algorithm with single objective function is presented in this paper. A Single Machine Infinite Bus (SMIB) system is considered. The main objective of this research paper is to investigate the suitability of genetic algorithm for effective tuning of parameters of the power system stabilizer in a single machine infinite bus system. A conventional speed based lead‐lag PSS is used. A simple and effective method of tuning the parameters of PSS is proposed which is posed as an optimization formulation by maximizing the damping of modes of oscillations of the SMIB system over a wide range of loading conditions and different system configurations. It is found that GA based PSS with single objective design shows improved dynamic performance over Conventional PSS over a wide range of operating conditions and different system parameters.

Non-traditional algorithms are the realistic models that are used for solving many complex engineering optimization problems. Machining is one of the most important and widely used manufacturing processes which rely on optimization. Optimization problems may be either constrained or unconstrained in nature. The complication factor with constrained optimization is that there is a possible existence of one or more complex constraints. These constraints must be incorporated into the problem otherwise solution is unacceptable from a practical standpoint. Non-traditional algorithms are the best way for solving the constrained optimization problems. The advantages of non-traditional techniques are its tolerant of imprecision, uncertainty, and meta-heuristics. In this work, three different non-traditional algorithms such as genetic algorithm (GA), simulated annealing (SA), and particle swarm optimization (PSO) are used in multi-pass turning machining operation to identify the set of optimal parameter values for minimizing unit production cost. The mathematical model is taken from the literature (Chen and Tsai, Int J Prod Res 34:2803–2825, 1996 and it is used for the evaluation of algorithmic parameters. The main aim of this work is to analyze the effect of algorithmic parameters of non-traditional technique in turning operation. The analyses performed are (i) sensitivity analysis in GA, PSO, and SA, (ii) analysis to find the best reproduction method in GA (iii) analysis to analyze the performance of GA, PSO, and SA by varying algorithmic parameters, and (iv) analysis to compare the results of the algorithms with the literature, and the best one is proposed.

This paper proposes optimal tuning of Proportional Integral (PI) controller using Grey Wolf Optimizer (GWO) to overcome the slow convergence into local optimum associated with the Particle Swarm Optimization (PSO), and Genetic Algorithm (GA) tuning methods. The optimal gains of the PI controller obtained from tuning using GWO, PSO and GA were implemented in the pitch angle control of the fixed speed Wind Turbine integrated into the distribution system. The objectives are to enhance the efficiency and reduced the fluctuation in the Wind Turbine output. Similarly, the Ziegler Nichols (ZN) tuned PI controller, the PI-fuzzy logic and Fuzzy Logic controllers were implemented in the same distribution system. The implementation of the tuned PI controllers has enhanced the output power of the Wind Turbine compared to the value obtained using the Hybrid or Fuzzy controllers. Furthermore, the GWO tuned PI controller recorded the high average output power of the Wind Turbine with low mean power error compared to other controllers. Also, it recorded the second-position in smoothing the Wind Turbine output power. The significance of the research is it has enhanced the efficiency and smoothing the output power of the fixed speed Wind Turbine.

: A streaming statistical algorithm detects SSH client keystroke packets in a TCP connection on any port. Input data are time-stamps and TCP-IP header fields of packets in both directions, measured at a monitor on the path between the hosts. No packet content is included. The algorithm uses the packet dynamics just preceding and following a client packet with data to classify the packet as a keystroke or non-keystroke. The dynamics are described by classification variables derived from the arrival time-stamps and the packet data sizes, sequence numbers, acknowledgement numbers, and flags. The algorithm succeeds because a keystroke creates an identifiable dynamical pattern. Final testing of the algorithm is based on analysis of about 1 million connections covering all common network protocols. Data visualization and the statistical design of experiments play a critical role in the analysis. It is common to treat the choice of tuning parameters of a statistical or machine learning algorithm as an optimization that finds one set of parameter values. Instead, we run a designed experiment that treats the tuning parameters as statistical tuning factors, which yields valuable information about algorithm performance. One application of the algorithm is identification of any TCP connection as an SSH interactive session, allowing detection of backdoor SSH servers. More generally, the algorithm demonstrates the potential for the use of detailed packet dynamics to classify connections, important for network security. The algorithm is has been prototyped in the widely-used Argus traffic audit software system.

RESUMO O objetivo deste artigo é fazer uma avaliação crítica da hipótese de Arida-Bacha de que um prêmio de alto risco sobre a dívida externa e interna brasileira é consequência da conversibilidade limitada da conta de capital brasileira. Argumenta-se no presente trabalho que esta hipótese não é fundamentada nem em bases teóricas ou empíricas. A literatura teórica de conversibilidade da conta de capital não estabelece relação entre os controles de capital e o prêmio de risco sobre a dívida interna ou externa. A experiência recente da economia brasileira mostra que as taxas de juros reais ainda são muito altas, apesar da crescente liberalização da conta de capital. Testes econométricos envolvendo o prêmio de risco sobre a dívida externa brasileira e os índices de liberalização da conta de capital desenvolvidos por Cardoso e Goldfjan (1998) mostram que essas variáveis não cointegram. Assim, não é possível estabelecer a existência de uma relação de longo prazo entre a liberalização da conta de capital e o prêmio de risco, como seria de se esperar se a hipótese de Arida-Bacha estivesse correta.

The Extended Kalman Filter (EKF) is currently a dominant method of sensor fusion used for navigation of mobile devices, robotics and autonomous vehicles. One navigation system involves the EKF fusion of an Inertial Navigation System (INS) with a Global Navigation Satellite System (GNSS) to perform 3D pose estimation, which is essential to practical applications like autonomous vehicles and UAVs. This system involves using the INS to predict pose changes and uses GNSS measurements when available to correct for estimation drifts and sensor errors. The performance of the state estimation is heavily dependent on the accurate selection of EKF parameters, leading to the optimal selection of parameters being a critical factor in the design and use of the EKF. In this paper, a methodical and efficient method of EKF parameter tuning is presented. The tuning framework uses nominal parameters generated by Gauss Markov (GM) and Allan Variance (AV) methods that are tuned by Genetic Algorithms (GA) accelerated by a Design of Experiment (DoE) technique to efficiently optimize EKF parameters. This framework has been implemented in MATLAB and tested using simulations and real data. The results demonstrate that GA tuned parameters increases accuracy substantially, and that the DoE technique consistently improves the convergence behavior of the GA.

Intelligent-PID controller design for speed track in automatic train operation system with heuristic algorithms

Study the optimization of aero-engine PID (Proportional-Integral-Derivative) controller parameters. In order to improve the tuning accuracy on aero-engine PID control parameters for the optimal solution, this paper presents a genetic algorithm-based PID parameter tuning method. Because the choice of crossover and mutation probabilities in genetic algorithm have a significant impact on the convergence speed and stability of the control system, genetic algorithm is adopted of which the crossover and mutation probability can automatically change with the fitness. Simulation results show that after variable crossover and mutation probability adaptive genetic algorithm to optimize PID control parameters, the average convergence algebra is significantly reduced and the overall control performance of the system is better. Aero engine work process is complex systems that big lag, nonlinear and multi-coupling throughout the flight envelope, Aero engine aerodynamic thermodynamic processes will occur a very big change, with the change of environmental conditions and flight status (airspeed, Mach number, etc.), so be quick, effective control is particularly important. PID control is one of the earliest developed control strategy, due to its simple structure, robustness, high reliability, easy to implement and other characteristics, widely used in process control and motion control, especially in the determination system that precise mathematical model can be established. Three parameters of proportional coefficient ( Kp ), integral time ( Ki ) and derivative time ( Kd ) tuning are direct influenced the control effect of the PID controller, the optimizing of PID controller parameters directly affect the control effect of control system. At present, there are many ways to optimize the PID parameters, such as Ziegler-Nichols method, indirect optimization method, the gradient method, climbing method, etc. The designer of these methods, mainly through the Mathematical model of controlled object, combined with practical experience, through repeated on-site equipment debugging obtained. Because these tuning method are relatively complicated, and affected by human factors, the final parameters of being got are not usually the optimal solution. The traditional tuning method based on a specific model is not applicable, for such aero engine as complex nonlinear systems. Therefore you can take advantage of intelligent control technology for optimizing PID parameters in order to achieve optimal control. Genetic Algorithm (GA) is an optimization method for parallel simulation of natural genetic mechanisms and biological evolution and the formation of a random search, survival of the fittest principle of biological evolution exist in nature, was introduced to optimize the parameters of a coded series groups.

Electronic nose (e-nose) is an artificial olfaction system that is being widely used in many industries. E-noses detect smells with the help of electronic signals produced by a number of sensors. The important part of an efficient e-nose system is to recognize these electronic signals accurately by some pattern classification algorithm. Recently developed nonparallel plane proximal classifier (NPPC) has shown its effectiveness in pattern classification task using kernel trick. In general the performance of such classifier depends on the values of optimal parameter set as well as the feature set. In this research work we have studied the effect of simultaneous parameter and feature weight selection on the accuracy of black tea quality estimation employing multiclass one vs. one NPPC. In order to choose the model parameters we have used genetic algorithm (GA). Experimental results show that GA-based tuning and feature weighting scheme increases the performance of NPPC by ~ 2% in the problem of black tea quality prediction.