• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.5
• /
• pp.474-480
• /
• 2007

This paper presents an application of intelligent modeling method to manual control system with human operator. Human operator as a part of controller is difficult to be modeled because of changes in individual characteristics and operation environment. So in these situation, a fuzzy model developed relying on the expert's experiences or trial and error may not be acceptable. To supplement the fuzzy model block, a neural network based modeling error compensator is incorporated. The feasibility of the present fuzzy-neural modeling scheme has been investigated for the real human based target tracking system.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.28 no.3
• /
• pp.108-119
• /
• 2024

The Reynolds-averaged Navier-Stokes (RANS) simulations are commonly used in industrial applications due to their computational efficiency. However, the linear eddy viscosity model (LEVM) used in RANS often fails to accurately capture the anisotropy of Reynolds stress in complex flow conditions. To enhance RANS predictive accuracy, data-driven closure models, such as Tensor Basis Neural Network (TBNN) and Tensor Basis Random Forest (TBRF), have been proposed. However existing models, including TBNN and TBRF, have limitations in capturing the nonlocal patterns of turbulence models, resulting in irregular and unsmooth predictions. Convolutional neural networks (CNNs) are considered as an alternative approach, but their reliance on discretization poses challenges when dealing with arbitrarily designed meshes in RANS simulations. In this study, we propose a nonlinear convolutional neural operator as the RANS closure model. Our model satisfies Galilean invariance, can learn nonlocal physics, and recovers high-resolution physics even when trained on undersampled grids. The model outperforms existing TBNN and TBRF models, successfully predicting smooth fields of Reynolds stress in flows with adverse pressure gradients, separations, and streamline curvature, where existing models struggle or fail to provide accurate predictions.

• ETRI Journal
• /
• v.45 no.2
• /
• pp.318-328
• /
• 2023

Recently, embedded systems, such as mobile platforms, have multiple processing units that can operate in parallel, such as centralized processing units (CPUs) and neural processing units (NPUs). We can use deep-learning compilers to generate machine code optimized for these embedded systems from a deep neural network (DNN). However, the deep-learning compilers proposed so far generate codes that sequentially execute DNN operators on a single processing unit or parallel codes for graphic processing units (GPUs). In this study, we propose PartitionTuner, an operator scheduler for deep-learning compilers that supports multiple heterogeneous PUs including CPUs and NPUs. PartitionTuner can generate an operator-scheduling plan that uses all available PUs simultaneously to minimize overall DNN inference time. Operator scheduling is based on the analysis of DNN architecture and the performance profiles of individual and group operators measured on heterogeneous processing units. By the experiments for seven DNNs, PartitionTuner generates scheduling plans that perform 5.03% better than a static type-based operator-scheduling technique for SqueezeNet. In addition, PartitionTuner outperforms recent profiling-based operator-scheduling techniques for ResNet50, ResNet18, and SqueezeNet by 7.18%, 5.36%, and 2.73%, respectively.

• Journal of Applied and Pure Mathematics
• /
• v.4 no.3_4
• /
• pp.185-209
• /
• 2022

Here we exhibit multivariate quantitative approximations of Banach space valued continuous multivariate functions on a box or ℝ^N, N ∈ ℕ, by the multivariate normalized, quasi-interpolation, Kantorovich type and quadrature type neural network operators. We treat also the case of approximation by iterated operators of the last four types. These approximations are achieved by establishing multidimensional Jackson type inequalities involving the multivariate modulus of continuity of the engaged function or its high order Fréchet derivatives. Our multivariate operators are defined by using a multidimensional density function induced by the generalized symmetrical sigmoid function. The approximations are point-wise and uniform. The related feed-forward neural network is with one hidden layer.

• Journal of Applied and Pure Mathematics
• /
• v.5 no.1_2
• /
• pp.69-93
• /
• 2023

Here we give multivariate quantitative approximations of Banach space valued continuous multivariate functions on a box or ℝ^N, N ∈ ℕ, by the multivariate normalized, quasi-interpolation, Kantorovich type and quadrature type neural network operators. We treat also the case of approximation by iterated operators of the last four types. These approximations are derived by establishing multidimensional Jackson type inequalities involving the multivariate modulus of continuity of the engaged function or its high order Fréchet derivatives. Our multivariate operators are defined by using a multidimensional density function induced by a parametrized Gudermannian sigmoid function. The approximations are pointwise and uniform. The related feed-forward neural network is with one hidden layer.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.2925-2948
• /
• 2019

In order to improve the utilization of irrigation water resources of greenhouse tomatoes, a water-saving irrigation decision-making model based on genetic optimization T-S fuzzy neural network is proposed in this paper. The main work are as follows: Firstly, the traditional genetic algorithm is optimized by introducing the constraint operator and update operator of the Krill herd (KH) algorithm. Secondly, the weights and thresholds of T-S fuzzy neural network are optimized by using the improved genetic algorithm. Finally, on the basis of the real data set, the genetic optimization T-S fuzzy neural network is used to simulate and predict the irrigation volume for greenhouse tomatoes. The performance of the genetic algorithm improved T-S fuzzy neural network (GA-TSFNN), the traditional T-S fuzzy neural network algorithm (TSFNN), BP neural network algorithm(BPNN) and the genetic algorithm improved BP neural network algorithm (GA-BPNN) is compared by simulation. The simulation experiment results show that compared with the TSFNN, BPNN and the GA-BPNN, the error of the GA-TSFNN between the predicted value and the actual value of the irrigation volume is smaller, and the proposed method has a better prediction effect. This paper provides new ideas for the water-saving irrigation decision in greenhouse tomatoes.

• Journal of Ocean Engineering and Technology
• /
• v.34 no.3
• /
• pp.167-179
• /
• 2020

Recently, the increasing importance of artificial intelligence (AI) technology has led to its increased use in various fields in the shipbuilding and marine industries. For example, typical scenarios for AI include production management, analyses of ships on a voyage, and motion prediction. Therefore, this study was conducted to predict a response amplitude operator (RAO) through AI technology. It used a neural network based on one of the types of AI methods. The data used in the neural network consisted of the properties of the vessel and RAO values, based on simulating the in-house code. The learning model consisted of an input layer, hidden layer, and output layer. The input layer comprised eight neurons, the hidden layer comprised the variables, and the output layer comprised 20 neurons. The RAO predicted with the neural network and an RAO created with the in-house code were compared. The accuracy was assessed and reviewed based on the root mean square error (RMSE), standard deviation (SD), random number change, correlation coefficient, and scatter plot. Finally, the optimal model was selected, and the conclusion was drawn. The ultimate goals of this study were to reduce the difficulty in the modeling work required to obtain the RAO, to reduce the difficulty in using commercial tools, and to enable an assessment of the stability of medium/small vessels in waves.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.861-865
• /
• 1991

In general, the fusion of neural network and fuzzy logic theory is based on the fact that neural network and fuzzy logic theory have the common properties that 1) the activation function of a neuron is similar to the membership function of fuzzy variable, and 2) the functions of summation and products of neural network are similar to the Max-Min operator of fuzzy logics. In this paper, a fuzzy-neural network will be proposed and a force controller of the robot gripper, utilizing the fuzzy-neural network, will be presented. The effectiveness of the proposed strategy will be demonstrated by computer simulation.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.4
• /
• pp.1066-1073
• /
• 1998

In this paper, we propose a new efficient operator named DBAH (difference between arithmetic mean and harmonic mean) and a technique for extracting sketch features through learning fuzzy inference rules with a neural network. The DBAH operator provide some advantages; sensitivity dependence on local intensities and insensitivity on small rates of intensity change in very dark regions. Also, the proposed fuzzy reasoning technique by a neural network has a good performance in extracting sketch features without human intervention.

• Journal of Intelligence and Information Systems
• /
• v.2 no.2
• /
• pp.69-83
• /
• 1996

The objective of this research is to develop a knowledge acquisition and refinement method for a multi-objective and multi-decision FMS scheduling problem. A competitive neural network and an inductive learning algorithm are integrated to extract and refine necessary scheduling knowledge from simulation outputs. The obtained scheduling knowledge can assist the FMS operator in real-time to decide multiple decisions simultaneously, while maximally meeting multiple objective desired by the FMS operator. The acquired scheduling knowledge for an FMS scheduling problem is tested by comparing the desired and the simulated values of the multiple objectives. The result show that the knowledge acquisition and refinement method is effective for the multi-objective and multi-decision FMS scheduling problems.