Density limit disruption prediction using a long short-term memory network on EAST

Disruption prediction using a long short-term memory(LSTM)algorithm has been developed on EAST,due to its inherent advantages in time series data processing.In the present work,LSTM is used as the model and the AUC(area under receiver operation characteristic curve)is used as the evaluation index.When the model is trained on data from the plasma current flattop phase and tested on data from the same period multiple times,the highest AUC is 0.8646 and the training time is about 6900 s per epoch.For comparison,the last 1000 ms of the flattop phases are intercepted as short time sequences.When the model is trained on data from short time sequences and tested on data from the same period,the highest AUC is increased to 0.9379 and the training time is restricted to 36 s per epoch.When the best model trained on the short time sequences is applied to the flattop phase for testing,the AUC is up to 0.9189.The experiment results show that it is possible for LSTM to train the model on data from short time sequences and migrate the model to the entire flattop phase,with a shorter training time and higher AUC value.

Disruption prediction based on fusion feature extractor on J-TEXT

Predicting disruptions across different tokamaks is necessary for next generation device.Future large-scale tokamaks can hardly tolerate disruptions at high performance discharge,which makes it difficult for current data-driven methods to obtain an acceptable result.A machine learning method capable of transferring a disruption prediction model trained on one tokamak to another is required to solve the problem.The key is a feature extractor which is able to extract common disruption precursor traces in tokamak diagnostic data,and can be easily transferred to other tokamaks.Based on the concerns above,this paper presents a deep feature extractor,namely,the fusion feature extractor(FFE),which is designed specifically for extracting disruption precursor features from common diagnostics on tokamaks.Furthermore,an FFE-based disruption predictor on J-TEXT is demonstrated.The feature extractor is aimed to extracting disruption-related precursors and is designed according to the precursors of disruption and their representations in common tokamak diagnostics.Strong inductive bias on tokamak diagnostics data is introduced.The paper presents the evolution of the neural network feature extractor and its comparison against general deep neural networks,as well as a physics-based feature extraction with a traditional machine learning method.Results demonstrate that the FFE may reach a similar effect with physics-guided manual feature extraction,and obtain a better result compared with other deep learning methods.

Mitigation and prediction of disruption on the HL-2A Tokamak

Injection of high-Z impurities into plasma has been proved to be able to reduce the localized thermal load and mechanical forces on the in-vessel components and the vacuum vessel, caused by disruptions in Tokamaks. An advanced prediction and mitigation system of disruption is implemented in HL-2A to safely shut down plasmas by using the laser ablation of high-Z impurities with a perturbation real-time measuring and processing system. The injection is usually triggered by the amplitude and frequency of the MHD perturbation field which is detected with a Mirnov coil and leads to the onset of a mitigated disruption within a few milliseconds. It could be a simple and potential approach to significantly reducing the plasma thermal energy and magnetic energy before a disruption, thereby achieving safe plasma termination. The plasma response to impurity injection, a mechanism for improving plasma thermal and current quench in major disruptions, the design of the disruption prediction warner, and an evaluation of the mitigation success rate are discussed in the present paper.

Overview of machine learning applications in fusion plasma experiments on J-TEXT tokamak

Machine learning research and applications in fusion plasma experiments are one of the main subjects on J-TEXT.Since 2013,various kinds of traditional machine learning,as well as deep learning methods have been applied to fusion plasma experiments.Further applications in the real-time experimental environment have proved the feasibility and effectiveness of the methods.For disruption prediction,we started by predicting disruptions of limited classes with a short warning time that could not meet the requirements of the mitigation system.After years of study,nowadays disruption prediction methods on J-TEXT are able to predict all kinds of disruptions with a high success rate and long enough warning time.Furthermore,cross-device disruption prediction methods have obtained promising results.Interpretable analysis of the models are studied.For diagnostics data processing,efforts have been made to reduce manual work in processing and to increase the robustness of the diagnostic system.Models based on both traditional machine learning and deep learning have been applied to real-time experimental environments.The models have been cooperating with the plasma control system and other systems,to make joint decisions to further support the experiments.