Neural network based prediction of heat flux profiles on STRIKE

Abstract

The instrumented calorimeter STRIKE (Short-Time Retractable Instrumented Kalorimeter Experiment) has been designed with the main purpose of characterizing the SPIDER (Source for Production of Ion of Deuterium Extracted from Radio Frequency plasma) negative ion beam in terms of beam uniformity and divergence during short pulse operations. STRIKE is made of 16 1D Carbon Fiber Composite (CFC) tiles, intercepting the whole beam and observed on the rear side by infrared (IR) cameras. The front observation presents some drawbacks due to optically emitting layer caused by the excited gas between the beam source and the calorimeter, and the material sublimated from the calorimeter surfaces due to the heating itself. This paper proposes a Neural Network-based approach to solve the inverse non-linear problem of determining the energy flux profile impinging on the calorimeter, considering the 2D temperature pattern measured on the rear side of the tiles. Most of the conventional methods used to evaluate the inverse heat flux are unbearably time consuming; since the objective is having a tool for heat flux evaluation for STRIKE real time operation, the need to have a ready-to-go instrument to understand the beam condition becomes stringent. For this reason, in this paper, a Multi-Layer Perceptron has been used to solve the problem. Once properly trained, the neural networks provide a fast evaluation of the impinging flux. Furthermore, there is no need to optimize any parameter since this operation is already included in the self-adjustment of the network weights during the training. The achieved results show the reliability of the proposed method both with stationary and non-stationary heat fluxes.

Introduction

A negative ion source prototype, SPIDER (Source for Production of Ion of Deuterium Extracted from Radio Frequency plasma), has been operating in Padova since May 2018, with the aim of testing and developing the ITER-scale radio-frequency negative ion source, of studying the beam characteristics and to verify the source proper operation [1,2]. The SPIDER beam (40 A, 100 kV, formed by 1280 beamlets) is characterized by a full set of diagnostics [3], in particular by the instrumented calorimeter STRIKE [4] (Short Time Retractable Instrumented Kalorimeter Experiment) whose main components are anisotropic carbon fiber composite tiles. STRIKE will be used to characterize the SPIDER negative ion beam during short pulse operation (<10 s) to verify ITER requirements about the maximum allowed beam non-uniformity. A representation of a reduced set of the beam pattern is shown in Fig. 1(b) which represents 1/16 of the whole set, while the complete set can be identified in Fig. 2, where the entire Beam Source is shown before the assembly. The calorimeter is designed for measuring the uniformity of the ion current and to investigate the beamlet divergence and halo. The beam impinges directly onto the front side of the calorimeter, but the measurements, carried out by a set of two infrared cameras (IR), see Fig. 1(c), will be performed on the back side since the observation of the front would be disturbed by the optically emitting layer created by the beam interaction with the background gas, located between the beam source and STRIKE, and by the debris coming off the tile surface.

To have a better evaluation of the impinging energy flux, the tile material must then transfer the thermal pattern from the front to the back side with a distortion as small as possible; this requires the thermal conductivity along the beam direction to be much higher than the conductivity in the plane perpendicular to the beam; the tiles tested so far have a ratio of 20 [5].

Several methods are indicated in literature (a short list is reported [[6], [7], [8], [9]]) as able to determine the impinging flux with the lowest possible error, but they all exhibit a major drawback: significant computational time to get the convergence for every single thermal pattern considered. The need to have a ready-to-go instrument to understand the beam condition while operating STRIKE is a stringent requirement, therefore any used method must comply with the requirement of carrying out the analysis within few minutes for the whole set of IR frames.