PSO GA Research Articles

Inverse Lithography Source Optimization via Particle Swarm Optimization and Genetic Combined Algorithm

Inverse lithography technologies (ILTs) are critical for improving the imaging performance of lithography in advanced technology nodes. Pixel-based source optimization (SO), as an efficient part of ILTs, can be implemented <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">via</i> heuristic approaches to achieve high-performance lithographic imaging. In this paper, a SO approach based on a combination of the particle-swarm optimization and genetic algorithms (PSO–GA) is proposed to determine the optimal intensity distribution of the source <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">via</i> iterations. The pixelated source can be decoded into the optimized variables of the merit functions in the SO model. The proposed PSO–GA algorithm, as a high-efficiency hybrid algorithm, can transform the discrete SO problem into the optimal search solution for the merit function, thereby inversely enhancing the lithographic-imaging performance. In the forward-imaging model in the lithography, the extraction of the mask's effective diffraction spectrum is implemented to calculate the layout of resist patterns. The simulation results highlight the superior performance of the proposed approach in achieving pixelated SO over the traditional GA and PSO algorithm in terms of convergence capacity.

An Improved Equilibrium Optimizer Algorithm and Its Application in LSTM Neural Network

An improved equilibrium optimizer (EO) algorithm is proposed in this paper to address premature and slow convergence. Firstly, a highly stochastic chaotic mechanism is adopted to initialize the population for range expansion. Secondly, the capability to conduct global search to jump out of local optima is enhanced by assigning adaptive weights and setting adaptive convergence factors. In addition 25 classical benchmark functions are used to validate the algorithm. As revealed by the analysis of the accuracy, speed, and stability of convergence, the IEO algorithm proposed in this paper significantly outperforms other meta-heuristic algorithms. In practice, the distribution is asymmetric because most logging data are unlabeled. Traditional classification models have difficulty in accurately predicting the location of oil layer. In this paper, the oil layers related to oil exploration are predicted using long short-term memory (LSTM) networks. Due to the large amount of data used, however, it is difficult to adjust the parameters. For this reason, an improved equilibrium optimizer algorithm (IEO) is applied to optimize the parameters of LSTM for improved performance, while the effective IEO-LSTM is applied for oil layer prediction. As indicated by the results, the proposed model outperforms the current popular optimization algorithms including particle swarm algorithm PSO and genetic algorithm GA in terms of accuracy, absolute error, root mean square error and mean absolute error.

Identification of significant bio-markers from high-dimensional cancerous data employing a modified multi-objective meta-heuristic algorithm

Identification of the most prominent bio-markers or genes with high classification accuracy in the high-dimensional cancerous data, is still remained as an emerging challenge for the machine learning researchers. As this challenge has two objectives, i.e. minimizing the number of genes (NoG) and maximizing the classification accuracy percentage (CAP), this problem can be modeled as binary multi-objective approach. In this work, a modified version of multi-objective Jaya algorithm, multi-objective chaotic Jaya (MOCJaya), is suggested to select the minimum NoG with high CAP. Initially, a filter approach, namely Fisher score is applied to pre-select the informative genes. Then, MOCJaya algorithm is employed for both selecting key genes and classifying the cancer data. To assess the efficacy of the designed algorithm, ten binary and multi-class cancerous datasets are considered. Here, the suggested algorithm has been compared with multi-objective chaotic Genetic Algorithm (MOCGA), multi-objective chaotic particle swarm optimization (MOCPSO), multi-objective Jaya (MOJaya), multi-objective PSO (MOPSO), and non-dominated sorting GA (NSGA-II) models. Moreover, a comparison of MOCJaya algorithm with other seventeen existing models, is also performed here. The experimental results and comparison analysis reveal that MOCJaya classifies both the positive and negative samples of the cancer datasets in high CAP with a smaller NoG.

Stability enhancement of a centrifugal compressor using inclined discrete cavities

Inclined discrete cavities are proposed to improve the operating stability of a centrifugal compressor while minimizing the flow loss, and optimized to simultaneously maximize the adiabatic efficiency and stall margin. Aerodynamic analysis was performed using three-dimensional Reynolds-averaged Navier–Stokes equations. The numerical results obtained for the total pressure ratio and adiabatic efficiency were validated with experimental data for the centrifugal compressor with a smooth casing. The angle of the cavity port, axial distance between cavities, and inclined angle of the cavity were selected as design variables based on sensitivity analysis. The stall margin and adiabatic efficiency at the design point were used as the objective functions for the design. Latin hypercube sampling was used to select 24 design points in the design space. Two neural network models, that is, the radial basis and deep neural networks, were tested with respect to surrogate models and their performances were compared using statistical analysis. A hybrid PSO–GA algorithm was used to identify the optimal solutions for the surrogate models. The deep neural network model has a better overall performance than the radial basis neural network model. The optimization results show that the stall margin increases, with increments of up to 6.41% and 11.32% compared with the reference design and compressor with a smooth casing, respectively.

Feature selection using multimodal optimization techniques

This paper investigates the effect of using Multimodal Optimization (MO) techniques on solving the Feature Selection (FSel) problem. The FSel problem is a high-dimensional optimization problem in the nature and thus needs a solver with high exploration power. On the other hand, if alternative optimal solutions could be provided for a problem, the implementation phase may become more selective depending on the cost and limitations of domain of the problem. The high exploration power and solution conservation capability of MO methods make them able to find multiple suitable solutions in a single run. Therefore, MO methods can be considered as a powerful tool of finding suitable feature subsets for FSel problem. In this paper, we made a special study on the use of MO methods in the feature selection problem. The binary versions of some existing Evolutionary Algorithm (EA) based MO methods like Dynamic Fitness Sharing (DFS), local Best PSO variants and GA_SN_CM, are proposed and used for selection of suitable features from several benchmark datasets. The results obtained by the MO methods are compared to some well-known heuristic approaches for FSel problem from the literature. The obtained results and their statistical analyses indicate the effectiveness of MO methods in finding multiple accurate feature subsets compared to existing powerful methods.

A novel hybrid PSO–GA meta-heuristic for scheduling of DAG with communication on multiprocessor systems

This work presents a novel hybrid meta-heuristic that combines particle swarm optimization and genetic algorithm (PSO---GA) for the job/tasks in the form of directed acyclic graph (DAG) exhibiting inter-task communication. The proposed meta-heuristic starts with PSO and enters into GA when local best result from PSO is obtained. Thus, the proposed PSO---GA meta-heuristic is different than other such hybrid meta-heuristics as it aims at improving the solution obtained by PSO using GA. In the proposed meta-heuristic, PSO is used to provide diversification while GA is used to provide intensification. The PSO---GA is tested for task scheduling on two standard well-known linear algebra problems: LU decomposition and Gauss---Jordan elimination. It is also compared with other states-of-the-art heuristics for known solutions. Furthermore, its effectiveness is evaluated on few large sizes of random task graphs. Comparative study of the proposed PSO-GA with other heuristics depicts that the PSO---GA performs quite effectively for multiprocessor DAG scheduling problem.

A hybrid self-adaptive Particle Swarm Optimization–Genetic Algorithm–Radial Basis Function model for annual electricity demand prediction

The present study proposes a hybrid Particle Swarm Optimization and Genetic Algorithm optimized Radial Basis Function (PSO–GA-RBF) neural network for prediction of annual electricity demand. In the model, each mixed-coding particle (or chromosome) is composed of two coding parts, binary and real, which optimizes the structure of the RBF by GA operation and the parameters of the basis and weights by a PSO–GA implementation. Five independent variables have been selected to predict future electricity consumption in Wuhan by using optimized networks. The results shows that (1) the proposed PSO–GA-RBF model has a simpler network structure (fewer hidden neurons) or higher estimation precision than other selected ANN models; and (2) no matter what the scenario, the electricity consumption of Wuhan will grow rapidly at average annual growth rates of about 9.7–11.5%. By 2020, the electricity demand in the planning scenario, the highest among the scenarios, will be 95.85billionkWh. The lowest demand is estimated for the business-as-usual scenario, and will be 88.45billionkWh.

Provincial carbon intensity abatement potential estimation in China: A PSO–GA-optimized multi-factor environmental learning curve method

This study aims to estimate carbon intensity abatement potential in China at the regional level by proposing a particle swarm optimization–genetic algorithm (PSO–GA) multivariate environmental learning curve estimation method. The model uses two independent variables, namely, per capita gross domestic product (GDP) and the proportion of the tertiary industry in GDP, to construct carbon intensity learning curves (CILCs), i.e., CO2 emissions per unit of GDP, of 30 provinces in China. Instead of the traditional ordinary least squares (OLS) method, a PSO–GA intelligent optimization algorithm is used to optimize the coefficients of a learning curve. The carbon intensity abatement potentials of the 30 Chinese provinces are estimated via PSO–GA under the business-as-usual scenario. The estimation reveals the following results. (1) For most provinces, the abatement potentials from improving a unit of the proportion of the tertiary industry in GDP are higher than the potentials from raising a unit of per capita GDP. (2) The average potential of the 30 provinces in 2020 will be 37.6% based on the emission's level of 2005. The potentials of Jiangsu, Tianjin, Shandong, Beijing, and Heilongjiang are over 60%. Ningxia is the only province without intensity abatement potential. (3) The total carbon intensity in China weighted by the GDP shares of the 30 provinces will decline by 39.4% in 2020 compared with that in 2005. This intensity cannot achieve the 40%–45% carbon intensity reduction target set by the Chinese government. Additional mitigation policies should be developed to uncover the potentials of Ningxia and Inner Mongolia. In addition, the simulation accuracy of the CILCs optimized by PSO–GA is higher than that of the CILCs optimized by the traditional OLS method.

Type-1 and Type-2 fuzzy logic controller design using a Hybrid PSO–GA optimization method

In this paper we propose a Hybrid PSO–GA optimization method for automatic design of fuzzy logic controllers (FLC) to minimize the steady state error of a plant’s response. We test the optimal FLC obtained by the Hybrid PSO–GA method using benchmark control plants and an autonomous mobile robot for trajectory tracking control. The bio-inspired method is used to find the parameters of the membership functions of the FLC to obtain the optimal controller for the respective plants. Simulation results show the feasibility of the proposed approach for these control applications. A comparison is also made among the proposed Hybrid PSO–GA, with GA and PSO to determine if there is a significant difference in the results.

Performance Comparison of Hybrid GA-PSO Based Tuned IMMs for Maneuver Target Tracking

Target tracking is very important field of research as it has wider applications in defense as well as civilian applications. Kalman filter is generally used for such applications. When the process and measurements are non linear extensions of Kalman filters like Extended Kalman Filter, Unscented Kalman Filters are widely used. UKF can give estimations up to second order characteristics of random process. The target is maneuvering and switching among different models like constant velocity (CV), constant acceleration (CA) or constant turn (CT), Interactive Multiple Models (IMM) are employed. Implementation of IMM filters for any application is difficult because of initialization of Kalman filter i,e, tuning of filter has to be performed before applying to real time situations. It demands prior estimations of Noise covariance matrices which are left for engineering intuitions. This paper presents the nonlinear state estimation using IMM and tuning of the filter is done using bio-inspired algorithms like PSO GA and Hybrid GA-PSO.

IHPG Algorithm for Efficient Information Fusion in Multi-Sensor Network via Smoothing Parameter Optimization

This investigate proposed a innovative Improved Hybrid PSO-GA (IHPG) algorithm which it combined the advantages of the PSO algorithm and GA algorithm. The IHPG algorithm uses the velocity and posit...

IHPG Algorithm for Efficient Information Fusion in Multi-Sensor Network via Smoothing Parameter Optimization

This investigate proposed a innovative Improved Hybrid PSO-GA (IHPG) algorithm which it combined the advantages of the PSO algorithm and GA algorithm. The IHPG algorithm uses the velocity and position update rules of the PSO algorithm and the GA algorithm in selection, crossover and mutation thought. This study explores the quality monitoring experiment by three existing neural network approaches to data fusion in wireless sensor module measurements. There are ten sensors deployed in a sensing area, the digital conversion and weight adjustment of the collected data need to be done. This experiment result can improve the accuracy of the estimated data and reduce the randomness of computing by adjustment optimization of smoothing parameter. According to the experimental analysis, the IHPG is better than the single PSO and GA in comparison the various neural network learning model.

A PSO–GA optimal model to estimate primary energy demand of China

To improve estimation efficiency for future projections, the present study has proposed a hybrid algorithm, Particle Swarm Optimization and Genetic Algorithm optimal Energy Demand Estimating (PSO–GA EDE) model, for China. The coefficients of the three forms of the model (linear, exponential, and quadratic) are optimized by PSO–GA using factors, such as GDP, population, economic structure, urbanization rate, and energy consumption structure, that affect demand. Based on 20-year historical data between 1990 and 2009, the simulation results of the proposed model have greater accuracy and reliability than other single optimization methods. Moreover, it can be used with optimal coefficients for the energy demand projections of China. The departure coefficient method is applied to get the weights of the three forms of the model to obtain a combinational prediction. The energy demand of China is going to be 4.79, 4.04, and 4.48 billion tce in 2015, and 6.91, 5.03, and 6.11 billion tce (“standard” tons coal equivalent) in 2020 under three different scenarios. Further, the projection results are compared with other estimating methods.

Energy demand projection of China using a path-coefficient analysis and PSO–GA approach

Energy demand projection is fundamental to rational energy planning formulation. The present study investigates the direct and indirect effects of five factors, namely GDP, population, proportion of industrial, proportion of urban population and coal percentage of total energy consumption on China’s energy demand, implementing a path-coefficient analysis. On this basis, a hybrid algorithm, Particle Swarm Optimization and Genetic Algorithm optimal Energy Demand Estimating (PSO–GA EDE) model, is proposed for China. The coefficients of the three forms of the model (linear, exponential and quadratic model) are optimized by proposed PSO–GA. To obtain a combinational prediction of three forms, a departure coefficient method is applied to get the combinational weights. The results show that the China’s energy demand will be 4.48 billion tce in 2015. Furthermore; the proposed method forecast shows its superiority compared with other single optimization method such as GA, PSO or ACO and multiple linear regressions.