ENEmi Narita

Emi Narita

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Emi Narita has a total of 50 co-authors. Here are the top co-authors by number of shared publications:

  • Mitsuru Honda

    Graduate School of Engineering, Kyoto University, Nishikyo, Kyoto 615-8530, Japan

    8 shared publications
  • Maiko Yoshida

    Naka Fusion Institute, Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko, 801-1, Mukouyama, Naka, Ibaraki, 311-0193, JAPAN

    5 shared publications
  • Mitsuru Honda

    Fusion Energy Research and Development Directorate, Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko, 221 JT-60 Control Building, 801-1 Mukoyama, Naka, Ibaraki, 311-0193, JAPAN

    5 shared publications
  • Hidenobu Takenaga

    National Institutes for Quantum and Radiological Science and Technology, Naka, Ibaraki, Japan

    4 shared publications
  • Nobuhiko Hayashi

    Advanced Plasma Modeling Group, Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Naka Kaku Yugo Kenkyujo, Naka-gun, Ibaraki, JAPAN

    4 shared publications
  • Fumiyoshi Kin

    Kyoto University

    3 shared publications

Latest Publications

  • Open Access

    Transport model surrogates using stochastic variational Gaussian processes with deep kernels

    by M. Honda, S. Maeyama, E. Narita

    A surrogate model for turbulent transport models is proposed based on Gaussian process (GP) regression. To ensure applicability to medium- and large-scale datasets, the model incorporates the stochastic variational Gaussian process, a scalable GP framework. Expressive deep multi-layer kernels, constructed hierarchically in a layer-by-layer manner using an iterative formula, are employed to capture the complex input–output relationship characteristics of turbulent transport models. Integrated into the steady-state transport code GOTRESS, the GP-based surrogate models successfully reproduce the temperature profiles predicted by the original turbulent transport model TGLF. It is confirmed that the GP-based surrogate model remains effective even when trained on a very limited number of training data points.


  • Thermal transport and confinement improvements with a high power and high energy neutral beam in JT-60U ELMy H-mode plasmas

    by M Yoshida, E Narita, S Sumida

    Improvements in thermal transport and global energy confinement with a high power and high energy neutral beam (NB) have been discussed in JT-60U ELMy H-mode plasmas in terms of fast-ion effects on thermal transport. Fast-ion effects have been investigated in various tokamak devices and their mechanisms have been discussed with theoretical models. This paper presents the characteristics of thermal transport and global energy confinement with a power modulation of positive-ion-based NB (PNB) of 85 keV and negative-ion-based NB (NNB) of 0.4 MeV in JT-60U positive shear ELMy H-mode plasmas. NNB is one of the unique items in JT-60U but is hardly utilized for thermal transport studies. Both global thermal confinement and local thermal transport are improved by adding an NNB to the PNB injected H-mode plasmas. Power degradation of the global energy confinement was not observed with the NNB power scan in this series of experiments. The H-mode confinement enhancement factor increased with increasing power of the fast-ion beta. Less profile stiffness in the ion temperature, in other words an enhancement of internal transport barrier (ITB), has been identified with increasing heating power. The physical mechanisms of ITB formation are discussed based on the transport analyses; the ion thermal diffusivity ( χ i ) remained constant or decreased with increasing heating power, and tended to decrease at higher fast-ion pressures produced by NBs and the E×B shearing rate. Linear gyrokinetic simulations support the experimental observations, and the calculated linear growth rates of ion temperature gradient modes are reduced by adding fast-ion components.

  • Open Access

    Understanding of parameter dependence among the radiative collapse data in LHD plasmas with a causal discovery approach

    by A. Anzai, E. Narita, M. Honda, Y. Morishita, S. Murakami

    A causal discovery code, IEDS, has been developed and applied to data obtained from the Large Helical Device. IEDS can identify the dependence among variables quantitatively and construct a directed acyclic graph to represent their relations. The graph can be used to make graphical models, such as Bayesian networks, which can predict plasma behavior. The data used in this study include discharges with a radiative collapse and have been collected in a previous study to predict and control the radiative collapse. IEDS has demonstrated that the variables selected to predict the radiative collapse in the previous study are strongly connected to an indicator of the radiative collapse. The directed acyclic graph generated by IEDS also suggests that the relation between the line-averaged oxygen impurity emission intensity (OV) and the line-averaged carbon impurity emission intensity (CIV), which are included in the variables used to predict the radiative collapse, could be consistent with the experimental observation that shows OV increases before the increase in CIV.

  • Open Access

    Simulation study of interaction between energetic particles and magnetohydrodynamic modes in the JT-60SA inductive scenario with a flat q≈1 profile

    by P. Adulsiriswad, Y. Todo, M. Sato, N. Aiba, E. Narita, H. Wang, M. Idouakass, J. Wang

    Interactions between energetic particles (EPs), an internal kink mode, and other magnetohydrodynamic (MHD) instabilities in the inductive scenario of JT-60SA (scenario # 2) are simulated with MEGA, a global EP–MHD hybrid code. For this scenario, it was predicted by TOPICS, an integrated transport code that the internal kink mode can be unstable and the sawtooth relaxation results in a flat safety factor (q) profile with q ≈ 1 for r / a ⩽ 0.6 . In this equilibrium, it is found in the simulation results that the stability of the internal kink mode depends strongly on the bulk plasma pressure gradient ( ∇ P b ). In the n = 1 simulations where the toroidal mode number is restricted to n = 0 and 1, the pressure-driven internal kink mode is dominant. In the presence of co-passing EPs generated by the negative-ion-based neutral beam (NB), these EPs transfer energy to the internal kink mode; however, the EP driving rate ( γ h ) is much lower than the driving rate from the bulk plasma pressure gradient ( γ P ). The mode’s frequency is less than 1 kHz because the toroidal orbit frequency (ω φ ) and poloidal orbit frequency (ω θ ) of the co-passing EPs are approximately equal within the q = 1 surfaces. This mode affects the EP and bulk plasma pressure redistributions. The feasibility of stabilizing the internal kink mode using trapped EPs is also investigated. It is found that the trapped EPs with energy 85 keV generated by the positive-ion-based NBs cannot stabilize the internal kink mode. Stabilization is observed when the injection energy is greater than 500 keV. In the multi-n simulations, where n ⩽ 8 modes are retained, the most unstable modes are high n interchange modes with poloidal number m = n whose linear growth rates exceed that of the pressure-driven internal kink mode observed in the n = 1 simulations. The overlapping of these modes creates a stochastic magnetic field, leading to stronger EP and bulk plasma pressure redistributions than those observed in the n = 1 simulations. During the nonlinear phase, the transition from the high m = n modes to the low m = n modes is observed where the dominant mode is the m / n = 1 / 1 mode with an internal kink-like structure. These low m = n modes are generated by the nonlinear coupling of the high m = n modes. The EP kinetic effect has a minor contribution to the dynamics of these nonlinearly generated m = n modes.


  • Open Access

    Simultaneous control of safety factor profile and normalized beta for JT-60SA using reinforcement learning

    by T. Wakatsuki, M. Yoshida, E. Narita, T. Suzuki, N. Hayashi

    Plasma with an internal transport barrier (ITB) will be developed in JT-60SA as an attractive operation appropriate for a steady-state fusion reactor. To achieve the ITB plasma while avoiding magnetohydrodynamic instabilities, it is advantageous to simultaneously control the safety factor (q) profile and the normalized beta ( β N ). In this study, a control system for simultaneous control of the q profile and β N is studied in simulations prior to the real experiment in JT-60SA. The bootstrap current dominates the total current in the ITB region, which results in a coupling between the pressure profile and the q profile. Thus, it is crucial to control the q profile and β N according to the strength of the ITB. A two-stage neural network (NN)-based control system was developed to address this problem. The first stage estimates the transport properties (i.e. ITB strength) of the plasma from measurements. The second stage consists of several NNs for control of the q profile and β N . According to the ITB strength estimated by the NN in the first stage, the appropriate NN for control is selected from those in the second stage. Each NN in the second stage is trained to control plasmas with different ITB strengths through reinforcement learning employing RAPTOR, an integrated transport code. To validate this system, it is tested in a simulation employing another integrated transport code, TOPICS, to mimic the plasma control in JT-60SA plasmas with various ITB strengths. Stable control of the q profile and β N is achieved in ITB plasmas simulated by both the RAPTOR and TOPICS codes.


  • Toward efficient runs of nonlinear gyrokinetic simulations assisted by a convolutional neural network model recognizing wavenumber-space images

    by E. Narita, M. Honda, S. Maeyama, T.-H. Watanabe

    A neural-network based innovative model recognizing the wavenumber space images has been developed to accurately forecast when the saturation of turbulent heat fluxes commences, i.e., the saturation time, in nonlinear gyrokinetic simulations. The wavenumber space images of the perturbed distribution function are focused on, which better represent the characteristics of turbulence. The model exploiting the state-of-the-art convolutional neural network model is capable of detecting minuscule differences between the images. Once the wavenumber space image is fed into the developed model, it can quickly and almost perfectly classify which phase of the turbulence evolution in nonlinear gyrokinetic simulations the image is in: the linearly and nonlinearly growing phases and the saturation phase. It can also predict the simulation time at which the image was processed with significantly high accuracy. The model enables us to forecast the saturation time of the gyrokinetic simulation in question by feeding an image at an early stage of the simulation and receiving the degree of progress toward the saturation. The ability of the model makes it possible to easily search out a desirable initial condition that rapidly conducts the simulation to a saturation phase. Such a pre-prediction model is important for running long time simulations on a large scale supercomputer like Fugaku in view of the efficient use of computational resources. In order to improve the predictive capability for the simulation that is going to be performed, several prediction models are trained by data with different major instabilities. The best predictor is selected to be in use based on the result of the pre-performed linear stability calculation with low computational cost.


  • Quasilinear turbulent particle and heat transport modelling with a neural-network-based approach founded on gyrokinetic calculations and experimental data

    by E. Narita, M. Honda, M. Nakata, M. Yoshida, N. Hayashi

    A novel quasilinear turbulent transport model DeKANIS has been constructed founded on the gyrokinetic analysis of JT-60U plasmas. DeKANIS predicts particle and heat fluxes fast with a neural network (NN) based approach and distinguishes diffusive and non-diffusive transport processes. The original model only considered particle transport, but its capability has been extended to cover multi-channel turbulent transport. To solve a set of particle and heat transport equations stably in integrated codes with DeKANIS, the NN model embedded in DeKANIS has been modified. DeKANIS originally determined turbulent saturation levels semi-empirically based on JT-60U experimental data, but now it can also estimate them using a theory-based saturation rule. The new saturation model is still partly connected to experimental data, but it offers the potential for applying DeKANIS independently of the device.


  • Open Access

    Application of transient transport analysis method for modulation experiment

    by Y. Ohtani, M. Yoshida, M. Honda, E. Narita

    A transient TRansport Analysis method for Modulation (t-TRAM) has been developed. This method consists of two methods. One is an instantaneous modulation analysis method using the Hilbert transform (HT) with the analytical mode decomposition (AMD). In the modulation experiments, plasma transport can be changed periodically by modulated external input. Profiles of the modulation amplitude and phase change in time due to the change in the plasma transport. The HT with AMD method can evaluate an instantaneous analytic signal in the presence of changes in the modulation amplitude and phase in time. The other is a transport analysis method to evaluate the diffusion coefficient and the convection velocity from the analytic signal. The t-TRAM has enabled us to evaluate a momentum diffusion coefficient and momentum convection velocity in the presence of a momentum source. The t-TRAM is applied to momentum modulation experiments in JT-60U. The evaluation reveals that a normalized momentum flux has a linear dependence on a normalized momentum gradient during a half cycle of the modulation.