This study focuses on the development of a robust Petri-fuzzy-neural-network (PFNN) control strategy applied to a linear induction motor (LIM) drive for periodic motion. Based on the concept of the nonlinear state feedback theory, a feedback linearization control (FLC) system is first adopted in order to decouple the thrust force and the flux amplitude of the LIM. However, particular system information is required in the FLC system so that the corresponding control performance is influenced seriously by system uncertainties. Hence, to increase the robustness of the LIM drive for high-performance applications, a robust PFNN control system is investigated based on the model-free control design to retain the decoupled control characteristic of the FLC system. The adaptive tuning algorithms for network parameters are derived in the sense of the Lyapunov stability theorem, such that the stability of the control system can be guaranteed under the occurrence of system uncertainties. The effectiveness of the proposed control scheme is verified by both numerical simulations and experimental results, and the salient merits are indicated in comparison with the FLC system

Robust Petri Fuzzy-Neural-Network Control for Linear Induction Motor Drive

Urban traffic systems that possess system states that are distributed, parallel, deterministic, stochastic, discrete, and continuous are well suited for a Petri net (PN) approach The literature survey conducted in this paper shows the vast applications of PNs in modeling and simulation, analyzing and evaluating performances, intelligent control and optimization, and congestion management in urban traffic systems This paper outlines the related works conducted using PNs and discusses its viability, such as its contributions and limitations Extendibility and future research potential to further the successful applications of PNs in traffic systems are discussed and proposed in this paper

A Review on the Applications of Petri Nets in Modeling, Analysis, and Control of Urban Traffic

This paper focuses on the development of a Petri-fuzzy-neural-network (PFNN) control for an indirect field-oriented linear-induction-motor (LIM) drive. First, an indirect field-oriented mechanism for a LIM drive is derived to preserve the decoupling control characteristic. Then, the concept of a Petri net (PN) is incorporated into a traditional FNN (TFNN) to form a new type of PFNN framework for alleviating the computation burden. Moreover, the supervised gradient descent method is used to develop the online training algorithm for the PFNN. In order to guarantee the convergence of tracking error, analytical methods based on a discrete-type Lyapunov function are proposed to determine the varied learning rates of the PFNN. With the proposed PFNN control system, the mover position of the controlled LIM drive possesses the advantages of good transient control performance and robustness to uncertainties for the tracking of periodic reference trajectories. In addition, the effectiveness of the proposed control scheme is verified by both numerical simulations and experimental results. Furthermore, the superiority of the proposed PFNN control system is indicated in comparison with the TFNN control system

Motion Control of Linear Induction Motor via Petri Fuzzy Neural Network

Traditional models unable to model adaptive software systems since they deal with fixed requirements only, but cannot handle the behaviors that change at runtime in response to environmental changes. In this paper, an adaptive Petri net (APN) is proposed to model a self-adaptive software system. It is an extension of hybrid Petri nets by embedding a neural network algorithm into them at some special transitions. The proposed net has the following advantages: 1) it can model a runtime environment; 2) the components in the model can collaborate to make adaption decisions while the system is running; and 3) the computation is done at the local component, while the adaption is for the whole system. We illustrate the proposed APN by modeling a manufacturing system.

Modeling Self-Adaptive Software Systems With Learning Petri Nets

Traditional models have limitation to model adaptive software systems since they build only for fixed requirements, and cannot model the behaviors that change at run-time in response to environmental changes. In this paper, an adaptive Petri net is proposed to model a self-adaptive software system. It is an extension of hybrid Petri nets by embedding a neural network algorithm into them at some special transitions. The proposed net has the following advantages: 1) It can model a runtime environment; 2) The components in the model can collaborate to make adaption decisions; and 3) The computing is done at the local, while the adaption is for the whole system. We illustrate the proposed adaptive Petri net by modeling a manufacturing system.

Modeling self-adaptive software systems with learning petri nets

According to recent knowledge of brain science it is suggested that there exists functions distribution, which means that specific parts exist in the brain for realizing specific functions. This paper introduces a new brain-like model called Learning Petri Network (LPN) that has the capability of functions distribution and learning. The idea is to use Petri net to realize the functions distribution and to incorporate the learning and representing ability of neural network into the Petri net. The obtained LPN can be used in the same way as a neural network to model and control dynamic systems, while it is distinctive to a neural network in that it has the capability of functions distribution. An application of the LPN to nonlinear crane control systems is discussed. It is shown via numerical simulations that the proposed LPN controller has superior performance to the commonly-used neural network one.

Learning Petri Network and Its Application to Non-linear System Control

Large-scale complicated systems are required to be controlled timely and appropriately. A human brain has similar functions to those of a controller of the large-scale complicated systems; it scans and recognizes sensory inputs and outputs responses to the environments. Why does a human brain work skillfully? The key is the capability of functions distribution and learning. Functions distribution means that a specific part exists in the brain, in order to realize a specific function. For example, a live neural network has different acting parts corresponding to different network inputs or stimuli. In this paper, we have proposed a new brain-like model that we call learning Petri network (LPN). The fundamental idea is to revise Petri net. Petri net is composed of state and transition and can control firing by tokens, so it is possible for this net to realize functions distribution. The revising point is to give Petri net the ability of learning as neural network (NN). And, it is the fundamental difference from NN, that learning of the proposed method is carried out on the only network pass of the token transfer.

Learning Petri network with route control

Learning Network Based on Petri Net with Functional Distributed Capability

A New Modelling Method of Functional Distribution of Brain by Modified Petri Net

軽油脱硫触媒の活性向上と，顧客ニーズへの迅速な対応を目的として，複数の軽油向け脱硫触媒を積層し，提案するシステムの検討を実施した。効果的な実験データの入手と効率的な設計を目的として，積層システムの検討に実験計画法と機械学習および多目的最適化手法を導入した。積層比率，実験条件を実験計画法により設定し，パイロットでの性能評価結果を教師データとして，応答局面法で機械学習を実施した。積層比率，性能評価条件，原料油性状を入力パラメーターとした生成油硫黄濃度を，RBFカーネルを用いて予測した。RBFカーネルは最も広く使用されるカーネルの一つで変数間の非線形性を考慮する関数である。機械学習により得られた最適な積層比率での性能の予測値は実験結果とよく一致しており，さらに使用した触媒単独および他積層比率よりも良好な結果を得ることができた。 

Shingo Sakai

Papers

Learning Petri Network and Its Application to Non-linear System Control

Learning Petri network with route control

Learning Network Based on Petri Net with Functional Distributed Capability

A New Modelling Method of Functional Distribution of Brain by Modified Petri Net

Development of Catalyst Grading System for Diesel Hydrodesulfurization Using Machine Learning Techniques