China
Africa
Explore chapters and articles related to this topic
A Novel Formulation for Estimating Compressive Strength of High Performance Concrete Using Gene Expression Programming
Published in M.Z. Naser, Leveraging Artificial Intelligence in Engineering, Management, and Safety of Infrastructure, 2023
Iman Mansouri, Jale Tezcan, Paul O. Awoyera
A critical step in all conventional evolutionary algorithms is the evaluation of each candidate solution using a predefined fitness function to quantify the degree to which it meets the optimization goal. A fitness function can be defined as a mapping from a chromosome to a scalar value. It is closely related to the objective function used in mathematical optimization, and in the case of unconstrained optimization, the two terms can often be used interchangeably. As the optimization algorithm is guided by the fitness scores of the candidate solutions, proper selection of the fitness function is very important to the success of any genetic algorithm. A good fitness function should have a clear mathematical definition, and the values it generates should be easy to interpret in the context of the desired optimization goal. Traditionally, the fitness function is selected such that it produces a positive value, higher values corresponding to better solutions. Another important consideration in selecting a fitness function is its efficient implementation. As the fitness function is evaluated many times during the course of optimization, its computation should be sufficiently fast.
Introduction
Published in Laxmidhar Behera, Swagat Kumar, Prem Kumar Patchaikani, Ranjith Ravindranathan Nair, Samrat Dutta, Intelligent Control of Robotic Systems, 2020
Laxmidhar Behera, Swagat Kumar, Prem Kumar Patchaikani, Ranjith Ravindranathan Nair, Samrat Dutta
Crossover: Crossover is a genetic operator that amalgamates the genes of two chromosomes from the parents in order to create new chromosome. In GA, a chromosome is represented by a bit string. In n-point crossover, n is selected randomly. The strings are split up in two segments at the n position and the segments are exchanged to create two new strings as shown in the Fig. 1.11. Mutation: Mutation is another genetic operator. It brings random changes in a chromosome with a hope that the change will give good fitness to the new chromosome. Mutation is performed based on a probability factor. This sometimes saves the algorithm from being stuck in the local minima. Fitness: The fitness of a chromosome is evaluated based on a fitness function. The fitness function defines the optimization problem. After the crossover and mutation, the new chromosome goes through the fitness test. The fitness tells how close the chromosome is to the optimal solution. Selection: The elite offsprings are selected based on the fitness values of the chromosomes. These elites are considered the parental population for the new generation.
Evolutionary Optimization Techniques as Effective Tools for Process Modelling in Food Processing
Published in Surajbhan Sevda, Anoop Singh, Mathematical and Statistical Applications in Food Engineering, 2020
Lakshmishri Roy, Debabrata Bera, Vijay Kumar Garlapati
An evolutionary algorithm is a biologically inspired, generic, population-based optimization algorithm. Its mechanism includes: Reproduction/procreation: The process of producing new “offspring” from their “parents”.Mutation: Alteration in the order of the process being considered (e.g., organism, production or business process, code).Recombination: A process of exchange of information between two processes yielding a new combination of processes (e.g., operations in a workflow process).Selection: A method by which traits become either more or less common in a population as a function of the influence of traits concerning the intended goal (e.g., increased production efficiency in a production process). Selection is a key evolution mechanism. Probable solutions of the optimization problem for which an evolutionary algorithm is employed to arrive at, are viewed as entities in a population. A fitness function is used to assess its suitability as a solution. A fitness function is an objective function that is used to summarize how close a given solution is to fulfilling the optimization goals. All the stated operators are applied several times in the process and, hence, the term “evolutionary”.
Optimization of working position and posture of a 5-DOF hybrid automatic drilling system based on an improved GA-BP neural network
Published in International Journal of Computer Integrated Manufacturing, 2023
Zhihao Wang, Hongbin Li, Nina Sun
The classical genetic algorithm cannot guarantee that the population generated by each iteration is superior to the population of the original parent. To overcome this shortcoming, an improved genetic algorithm is proposed to improve the generation process of the new population. The improvement of improved generation algorithm is as follows. The initial population and fitness function are calculated, and the parent generation is generated.According to the prediction accuracy, the node number of the hidden layer is set to 7.Multiple iterations are performed to reach the optimal population.The initial population represents the initial pose generated on the basis of the algorithm output. The fitness function is used to define the optimization objective of the algorithm. Optimal population denotes PP with the smallest positioning error.
Particle Swarm Optimization Applied to the Nuclear Fuel Bundle Spacer Grid Spring Design
Published in Nuclear Technology, 2019
Victor C. Leite, Roberto Schirru, Miguel Mattar Neto
Note that a fitness function must represent in its terms all of the characteristics wanted to be optimized in a problem. In the present problem, the SI and the stiffness are the goals to be optimized, and as was explained, each one of these characteristics is represented by a term in the fitness function. The same thing could be done in order to take into account many other desirable characteristics to be optimized, for example, to consider thermal-hydraulic conditions. This characteristic could be represented in the fitness function by a third term, such as . In this term, is a constant that may fit the order of magnitude between the other terms of the equation, just like . The value of could represent a calculated heat transfer convective coefficient in a high-heat-flux region, considering a given geometry of SG, while the value of could represent a desired heat transfer convective coefficient for that region. The term characterized by the difference of and should develop the same functionality of the stiffness term explained before.
A framework for the analysis and synthesis of Swarm Intelligence algorithms
Published in Journal of Experimental & Theoretical Artificial Intelligence, 2021
Dávila Patrícia Ferreira Cruz, Renato Dourado Maia, Leandro Nunes de Castro
The exploration is performed by means of a random search for new solutions and random initialisation of the agents, while the exploitation is carried out by a guided search that leads the agents to good regions in the search space found by other agents. The quality of the solutions is evaluated by a fitness function related to the problem where the algorithm will be applied. The dynamics of the algorithm is directly linked to how solutions are represented computationally, algorithms with the same representation of solutions tend to present similar dynamics. For example, algorithms whose solutions are represented by a vector of features (vector space) manipulate solutions by means of operators that, in some cases, resemble evolutionary operators.