The impact of disruptions in JET became even more important with the replacement of the previous Carbon Fiber Composite (CFC) wall with a more fragile full metal ITER-like wall (ILW). The development of robust disruption mitigation systems is crucial for JET (and also for ITER). Moreover, a reliable real-time (RT) disruption predictor is a pre-requisite to any mitigation method. The Advance Predictor Of DISruptions (APODIS) has been installed in the JET Real-Time Data Network (RTDN) for the RT recognition of disruptions. The predictor operates with the new ILW but it has been trained only with discharges belonging to campaigns with the CFC wall. 7 real-time signals are used to characterize the plasma status (disruptive or non-disruptive) at regular intervals of 1ms. After the first 3 JET ILW campaigns (991 discharges), the success rate of the predictor is 98.36% (alarms are triggered in average 426ms before the disruptions). The false alarm and missed alarm rates are 0.92% and 1.64%. IntroductIonDue to the complex and highly non-linear coupling of events that lead to a disruption, predictive physics-driven models of these phenomena have not been established from theoretical considerations so far. As an alternative, data-driven models allow the estimation of useful relationships among several quantities to recognize the signature of an incoming disruption.As mentioned in the abstract, the impact of disruptions in JET is a more serious issue with the ILW [1]. This article shows the results of a disruption predictor at JET, APODIS, that has been in operation in the JET RTDN during the three initial ILW campaigns (C28-C30, between August 2011 and July 2012). The objective has been to assess its prediction capabilities (success rate, missed alarms, false alarms and prediction times) for later use in next campaigns as trigger for mitigation actions. In the above ILW campaigns, the alarm generated by APODIS has been distributed through the RTDN and recorded for off-line analysis, but it has not been used to close any feedback loop.

A power-balance model, with radiation losses from impurities and neutrals, gives a unified description of the density limit (DL) of the stellarator, the L-mode tokamak, and the reversed field pinch (RFP). The model predicts a Sudo-like scaling for the stellarator, a Greenwald-like scaling, , for the RFP and the ohmic tokamak, a mixed scaling, , for the additionally heated L-mode tokamak. In a previous paper (Zanca et al 2017 Nucl. Fusion 57 056010) the model was compared with ohmic tokamak, RFP and stellarator experiments. Here, we address the issue of the DL dependence on heating power in the L-mode tokamak. Experimental data from high-density disrupted L-mode discharges performed at JET, as well as in other machines, are taken as a term of comparison. The model fits the observed maximum densities better than the pure Greenwald limit.

The 2014–2016 JET results are reviewed in the light of their significance for optimising the ITER research plan for the active and non-active operation. More than 60 h of plasma operation with ITER first wall materials successfully took place since its installation in 2011. New multi-machine scaling of the type I-ELM divertor energy flux density to ITER is supported by first principle modelling. ITER relevant disruption experiments and first principle modelling are reported with a set of three disruption mitigation valves mimicking the ITER setup. Insights of the L–H power threshold in Deuterium and Hydrogen are given, stressing the importance of the magnetic configurations and the recent measurements of fine-scale structures in the edge radial electric. Dimensionless scans of the core and pedestal confinement provide new information to elucidate the importance of the first wall material on the fusion performance. H-mode plasmas at ITER triangularity (H  =  1 at βN ~ 1.8 and n/nGW ~ 0.6) have been sustained at 2 MA during 5 s. The ITER neutronics codes have been validated on high performance experiments. Prospects for the coming D–T campaign and 14 MeV neutron calibration strategy are reviewed.

Prediction of disruptions from scratch is an ITER-relevant topic. The first operations with the new ITER-like wall constitute a good opportunity to test the development of new predictors from scratch and the related methodologies. These methodologies have been based on the Advanced Predictor Of DISruptions (APODIS) architecture. APODIS is a real-time disruption predictor that is in operation in the JET real-time network. Balanced and unbalanced datasets are used to develop real-time predictors from scratch. The discharges are used in chronological order. Also, different criteria to decide when to retrain a predictor are discussed. The best results are obtained by applying a hybrid method (balanced/unbalanced datasets) for training and with the criterion of retraining after every missed alarm. The predictors are tested off-line with all the discharges (disruptive/non-disruptive) corresponding to the first three JET ITER-like wall campaigns. The results give a success rate of 93.8% and a false alarm rate of 2.8%. It should be considered that these results are obtained from models trained with no more than 42 disruptive discharges.

The development of accurate real-time disruption predictors is a pre-requisite to any mitigation action. Present theoretical models of disruptions do not reliably cope with the disruption issues. This article deals with data-driven predictors and a review of existing machine learning techniques, from both physics and engineering points of view, is provided. All these methods need large training datasets to develop successful predictors. However, ITER or DEMO cannot wait for hundreds of disruptions to have a reliable predictor. So far, the attempts to extrapolate predictors between different tokamaks have not shown satisfactory results. In addition, it is not clear how valid this approach can be between present devices and ITER/DEMO, due to the differences in their respective scales and possibly underlying physics. Therefore, this article analyses the requirements to create adaptive predictors from scratch to learn from the data of an individual machine from the beginning of operation. A particular algorithm based on probabilistic classifiers has been developed and it has been applied to the database of the three first ITER-like wall campaigns of JET (1036 non-disruptive and 201 disruptive discharges). The predictions start from the first disruption and only 12 re-trainings have been necessary as a consequence of missing 12 disruptions only. Almost 10 000 different predictors have been developed (they differ in their features) and after the chronological analysis of the 1237 discharges, the predictors recognize 94% of all disruptions with an average warning time (AWT) of 654 ms. This percentage corresponds to the sum of tardy detections (11%), valid alarms (76%) and premature alarms (7%). The false alarm rate is 4%. If only valid alarms are considered, the AWT is 244 ms and the standard deviation is 205 ms. The average probability interval about the reliability and accuracy of all the individual predictions is 0.811 ± 0.189.