Genetic-algorithm-based method to optimize spatial profile utilizing characteristics of electrostatic actuator deformable mirror

Natural populations are dynamic in both time and space. In biological populations such as insects, spatial distribution patterns are often studied as the first step to characterize population dynamics. In nature, the spatial distribution patterns of insect populations are considered as the emergent expression (property) of individual behaviors at population levels and are fine-tuned or optimized by natural selection. This inspiration prompts us to investigate the possibly similar mechanisms in Genetic Algorithms (GA) populations. In this study, we introduce the mathematical models for the spatial distribution patterns of insect populations to GA with the conjecture that the emulation of biological populations in nature may lead to computational improvement. In particular, we introduce three modeling approaches from the research of spatial distribution patterns of insect populations: (i) probability distribution modeling approach, (ii) aggregation index approach, and (iii) Taylor's (1961, 1977) Power Law, Iwao's (1968, 1976) Mean Crowding Model and Ma's (1991c) population aggregation critical density (PACD), to characterize populations in GA. With these three approaches, we investigate four mappings from the research field of insect spatial distribution patterns to GA populations in order to search for possible counterpart mechanisms or features in GA. They are: (i) mapping insect spatial distribution patterns to GA populations or allowing GA populations to be controlled by stochastic distribution models that describe insect spatial distributions; (ii) mapping insect population distribution to GA population fitness distribution via Power Law and PACD modeling; (iii) mapping population aggregation dynamics to GA fitness progression across generations (or fitness aggregation dynamics in GA) via insect population aggregation index; (iv) mapping insect population sampling model to optimal GA population sizing. With regard to the mapping (i), the experiment results show the significant improvements in GA computational efficiency in terms of the reduced fitness evaluations and associated costs. This prompts us to suggest using probability distribution models, or what we call stochastic GA populations, to replace the fixed-size population settings. We also found the counterpart for the second mapping, the wide applicability of Power Law and Mean Crowding model to the fitness distribution in GA populations. The testing of the third and fourth mappings is very preliminary; we use example cases to suggest two further research problems: the potential to use fitness aggregation dynamics for controlling the number of generations iterated in GA searches, and the possibility to use fitness aggregation distribution parameters [(obtained in mapping (ii)] in determining the optimum population size in GA. A third interesting research problem is to investigate the relationship between mapping (i) and (iii), i.e., the controlling of both population sizes and population generations.

VLSI floorplanning design automation based on soft computing is discussed. A novel approach based on the fusion of genetic algorithms and fuzzy inference for the automation of VLSI floorplanning design is proposed. The fuzzy rules are used to infer the initial position of the on-chip blocks based on analysis of the accumulated knowledge of the expert design engineer. Only the dominant combinations of place and block are inferred. Blocks deemed suitable candidates for placement at the center, relative to the four-corners of the chip, are inferred. These inferences are then reflected in the initial population of the genetic algorithms. The rest of the block placement phase is entrusted to the genetic algorithms. Experimental software to implement the proposed approach was developed. This was in turn used to perform computer-based experiments. The results of the experiments showed a level and quality of placement close to that of the expert design engineer.

We report on the development of a new class of electrostatic MEMS deformable mirror (DM) fabricated through a combination of bulk micromachining, wafer bonding, and surface micromachining. The combination of these fabrication technologies introduces four major improvements over previous MEMS DMs, which are fabricated using surface micromachining alone. First, the MEMS DM structural components (mirror surface and actuator array) are made entirely of single crystalline silicon by use of the device layer of a whole 4-inch silicon-on-insulator (SOI) wafer bonded together via anodic bonding. Unlike current MEMS DMs fabricated entirely using surface micromachining, bulk micromachining steps in this fabrication process require no etch access holes, print through is inexistent, and no polishing steps are required. This leads to reduced diffraction of light from the mirror surface, improved mirror surface optical quality, and elimination of manufacturing processing steps. Second, through-wafer interconnects are used to connect the densely-packed electrostatic actuator array to driver electronics. This eliminates the need for wirebonding at the periphery of the DM, increasing the surface area available for actuators and removes the need for bulky wire bundles to connect the device to its driver. Third, by using the full area of a silicon wafer for each mirror, these MEMS DMs offer a larger optical aperture than any previously-reported MEMS DM. The larger aperture will achieve higher angular resolution, providing larger wavefront correction. Finally, the mirror and actuator thicknesses are not limited to several micrometers, unlike in surface micromachining. The thickness limits using this fabrication process is prescribed by the device layer thickness in SOI wafers, which vary between several micrometers to several hundred of microns.

Genetic algorithm in the wireless sensor networks localization has a problem that positioning errors of some nodes are larger, in this paper we propose an improved algorithm based on genetic algorithm with filter replenishment strategy(FRGA), we improve the regional constraint of the initial population of genetic algorithm, and introduce the filter and replenishment strategy, from the perspective of population differences in performance, we delete the poor individual to maintain population overall performance, and solve the problem that localization accuracy of some nodes is poor which caused by the premature convergence. Experiments show that the localization accuracy of the improved algorithm is better than the GA, and the improved algorithm has faster convergence speed, and suitable for large-scale wireless sensor networks.

A new method is developed for finite element (FE) domain decomposition. This method employs a hybrid graph-genetic algorithm for graph partitioning and correspondingly bisects finite element (FE) meshes.A weighted incidence graph is first constructed for the FE mesh, denoted by G0. A coarsening process is then performed using heavy-edge matching. A sequence of such operations is employed in n steps, which leads to the formation of Gn with a size suitable for genetic algorithm applications.Hereafter, Gn is bisected using conventional genetic algorithm. The shortest route tree algorithm is used for the formation of the initial population in genetic algorithm. Then an uncoarsening process is performed and the results are transferred to the graph Gn-1. The initial population for genetic algorithm on Gn-1 is constructed from the results of Gn. This process is repeated until G0 is obtained in the uncoarsening operation. Employing the properties of G1, the graph G0 is bisected by the genetic algorithm.

The efficiency improvement of topology optimization using genetic algorithms is necessary because of its high computation costs due to high dimension of topology optimization. In order to improve an efficiency of genetic algorithm for topology optimization, optimal solutions of Solid Isotropic Material with Penalization (SIMP) method was used for generation of initial population of genetic algorithms. For generation of initial population pool, we generated many similar individuals to optimal solutions of SIMP using another topology optimization using genetic algorithms. This study investigated the convergence and robustness of the proposed method.

Identification of planted (Z, d)-motifs is an important and hard challenging problem in computational biology. In this paper, we present an original algorithm that combines Genetic Algorithm (GA) and Random Projection Strategy (RPS) GARPS to identify (I, d)-motifs. We start with RPS to find good starting positions by introducing position-weight function and constructing a new hash function based on the function and return a set of candidate motifs. Then, we use the results(good candidate motifs) from RPS as the initial population of genetic algorithm to make series of iterations to refine motif candidates. We use the global search capability of GA and RPS are explored in GARPS. Experimental results on simulated data show that GARPS performs better than the Projection algorithm and solves the most of challenging planted motif finding problems and improves finding faint motifs.

Improved random selection strategy, self-adaptive crossover and mutation operators, population co-evolution algorithm are adopted in this paper to set up a new genetic algorithm(GA), guaranteeing the universality of population, the thoroughness of solution and the execution efficiency of algorithm in solving multi-process, process crossover, work-piece flow and other condition based scheduling problem of flexible manufacturing system(FMS). The problem of GA in solving FMS scheduling is summarized in this paper, reasons and countermeasures for GA's pre-maturing are studied, longitudinal and transverse comprehensive state-based correlative self-adaptive crossover and mutation operators are raised according to the analysis on GA to avoid local optimization and pre-maturity, the co-evolutionary GA is designed to solve the scheduling model for FMS, and the algorithm is validated for the scheduling of box-type parts for precise machine tools.

In this contribution we experimentally demonstrated femtosecond laser micro-processing of large arbitrary-shaped irradiance patterns in silicon without sample or beam scanning. This is mainly accomplished by carrying out spatial and temporal shaping of femtosecond pulses with the help of a phase-only spatial light modulator (SLM, PLUTO-NIR-II), and an acousto-optic programmable dispersive filter (Dazzler from FASTLITE). In particular, the spectral phase of the femtosecond pulse is changed by introducing different sets of dispersion parameters into the Dazzler, whereas the amplitude and spatial phase of the laser beam is controlled with a single phase mask encoded into the SLM. Each phase mask is designed following a specific method for encoding complex fields into a single phase element [1]. By using this diffraction-based method one can produce user-defined spatially uniform irradiance patterns with different spatial shapes at the surface of the silicon sample. Here, it should be mentioned that a suited setting of the position of the surface of the silicon sample with respect to the ablation plane is carried out by means of a multi-axis motorized stage (denoted by MS in Fig. 1). Before starting the micro-processing, the adequate laser fluence must be also determined. Then, in a single-shot regime controlled with a motorized shutter, the ablation process due to the interaction of above-mentioned irradiance patterns with the material surface is accomplished. Note that, under laser fluence range lower than 1 J/cm2 the dominate ablation mechanisms of spallation and fragmentation in metals [2] occur with minimal thermal or mechanical damages in the surrounding of the processed area. In addition, the lack of sequentially dot-by-dot scans over a sample considerably reduces the fabrication time and avoids boundary effects due to the ablation craters overlap.

As an active device, deformable mirror has been successfully used in adaptive optical systems to compensate phase aberrations and improve beam quality. However, it is often limited to correcting the phase aberrations for a flat reference wavefront. In this paper we present a new adaptive beam intensity profiles control technique based on the combination of a 19-element deformable mirror and a global genetic algorithm. This algorithm can adaptively adjust the voltages of the 19 actuators on the deformable mirror to reduce the difference between the target beam shape and the actual beam shape. The numerical simulations results show that within the mechanical constraints of the deformable mirror, this technique can create the given beam intensity profiles in a focal plane.

Ultrafast or ultrashort pulsed lasers have become integral in numerous industrial applications due to their high precision, non-thermal interaction with materials, and ability to induce nonlinear absorption. These characteristics have expanded their use in microfabrication, semiconductor processing, automotive engineering, and biomedical fields. Temporal pulse shaping reduces laser pulse durations, often to shorter timescales than many physical and chemical processes, enabling greater control. Meanwhile, spatial shaping improves efficiency and precision in micro- and nanofabrication and biomedical applications. Advances in optical parametric amplifiers (OPAs) and chirped-pulse amplifiers (CPAs) have allowed for more refined temporal and spatial shaping, ensuring the preservation of high peak power while achieving ultrashort pulse durations. Additionally, spatial light modulators (SLMs) have facilitated sophisticated beam shaping, which, when combined with ultrafast lasers, supports applications like computer-generated holography and nanoscale fabrication. These developments underscore the growing utility and versatility of ultrafast lasers in both research and industrial contexts.

Deformable mirror is a crucial component of adaptive optics. It can be used to vary the optical power of an optical image system, such as auto-focus and optical zoom function. An electrostatic type MEMS deformable mirror used in an optical image system has been promoted recently. However, high voltage in the range of hundreds volts is a serious concern for electrostatic type MEMS deformable mirror. Ionic-polymer metal composite (IPMC) is a polymer actuator with the advantage of large deformation under low actuation voltage. It is a sandwich structure composed of two metal electrodes and a layer of polymer film. The hydrated cation inside the polymer film moves toward the cathode. Because of the migration of ions and water inside the film, volume expansion and contraction induce the deformation of IPMC. In this paper, we design the IPMC type deformable mirror that is simulated by finite element method and then demonstrate its focus-varying function. It requires less than one volt to achieve over 100 diopters.

A new Heuristic Genetic Information based Ant Colony Genetic Algorithm (HGI-ACGA) to solve Traveling Salesman Problems(TSP) was proposed. HGI-ACGA's two sub-algorithms are Ant Colony Genetic Algorithm(ACGA) and Heuristic Genetic Information based Ant Colony Algorithm(HGI-ACA). ACGA enhances genetic algorithm's population diversity and reduces the search domain, while HGI-ACA eliminates the creation of invalid tours and also avoids depending on pheromone excessively. A combination of genetic information and pheromone leads to a significant improvement in performance. The strategy of HGI-ACGA algorithm can improve convergence rate and capacity of searching optimal solution. Experimental results show that the proposed algorithm generally exhibits a better solution and a higher rate of convergence for TSP than ACGA and ACA.

In typical applications, genetic algorithms (GAs) process populations of potential problem solutions to evolve a single population member that specifies an ‘optimized’ solution. The majority of GA analysis has focused on these optimization applications. In other applications (notably learning classifier systems and certain connectionist learning systems), a GA searches for a population of cooperative structures that jointly perform a computational task. This paper presents an analysis of this type of GA problem. The analysis considers a simplified genetics-based machine learning system: a model of an immune system. In this model, a GA must discover a set of pattern-matching antibodies that effectively match a set of antigen patterns. Analysis shows how a GA can automatically evolve and sustain a diverse, cooperative population. The cooperation emerges as a natural part of the antigen-antibody matching procedure. This emergent effect is shown to be similar to fitness sharing, an explicit technique for multimodal GA optimization. Further analysis shows how the GA population can adapt to express various degrees of generalization. The results show how GAs can automatically and simultaneously discover effective groups of cooperative computational structures.

Detecting quality community structures in complex networks is an important and highly active research area. Plenty of methods have been proposed for community detection in recent years. Among them, Genetic Algorithms (GAs) have been widely explored for community detection due to their strong competence at exploring the global discrete search space. However, existing GA algorithms for community detection still face major challenges when handling large and complex networks due to their use of random mutation operators. Whenever any candidate community structure in a GA population is mutated, a mutated node of the network under processing is often associated to a community with loose connections, seriously hurting GA’s effectiveness and scalability. To address this issue, a newly designed Leiden-based GA (LGA) with a novel mutation operator based on the Leiden algorithm is proposed in this paper to improve the effectiveness of the mutation operator and the performance of the GA approach. Experiment results clearly show that LGA can achieve highly competitive performance in comparison to several state-of-the-art GA and non-GA community detection algorithms on multiple synthetic and real-world networks.KeywordsCommunity detectionLeiden algorithmGenetic algorithm