Data Quality for the LUX-ZEPLIN detector, and HV performance R&D

LUX-ZEPLIN 探测器的数据质量和 HV 性能 R

• 批准号: 2582902
• 金额: --
• 依托单位: University of Oxford
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2021
• 资助国家: 英国
• 起止时间: 2021 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2582902
• 关键词: Data; Quality; LUX; ZEPLIN; detector

Data quality studies in the LZ direct dark matter detector fall into two broad categories: issues that require eliminating certain time periods from analysis, and issues that require cutting particular events from analysis. This student will participate in studying data quality in the first LZ science run, and take a leading role in data quality assessment and cuts in subsequent science runs. The aim of this analysis task is to maximize LZ sensitivity to dark matter signals with the appropriate exclusion of non-dark matter backgrounds and inclusion of potential signal events in the dataset. Machine learning techniques looking for new background event populations that may leak into the signal region be evaluated to see if they save time and effort of human analysts, and are more effective than human analysts. Such cuts require studying event populations that generally fall far from our analysis region and are usually cut by energy or position reconstruction, but may contain many more examples of data quality problems than can be seen in the restricted search dataset. the LUX, ZEPLIN, and LZ analysis teams have not yet utilized machine learning in data quality studies, and other groups are primarily looking for waveform anomalies, or anomalies within the restricted dataset, not within the greater dataset. Data quality studies are required for all new detectors, and every new data taking period of the detector. Performing these analysis cuts is often the most resource intensive part of the science searches.The high voltage performance of the detector is a significant contributor to data quality issues with additional electron emission from HV components being the primary concern. The student will embark on a design, engineering, and testing project to prepare for high voltage electrode grids for future time projection chambers. A new experiment will require modular grids that can be assembled in situ, and engineering such electrodes with the maximization of light collection and electric field uniformity will be performed with light collection, engineering, and electric field simulations. Engineering prototypes of varies scales will be produced and tested. In addition, small scale test electrodes will be operated to build statistics on initial grid operation and burn in/conditioning to understand what the optimal commissioning plan is for a large scale electrode grid. multiple electrode technologies may be studied, from wires woven by experimentalists, commercial meshes, etched meshes, and thin film depositions.

LZ直接暗物质探测器的数据质量研究分为两大类：需要从分析中消除某些时间段的问题，以及需要从分析中删除特定事件的问题。这名学生将参与第一次LZ科学运行中的数据质量研究，并在后续科学运行中的数据质量评估和削减中发挥主导作用。该分析任务的目的是最大限度地提高LZ对暗物质信号的灵敏度，并适当排除非暗物质背景，并在数据集中包含潜在的信号事件。机器学习技术寻找可能泄漏到信号区域的新背景事件群体，以查看它们是否节省人类分析师的时间和精力，并且比人类分析师更有效。这种切割需要研究通常远离我们的分析区域的事件群体，并且通常通过能量或位置重建来切割，但可能包含比在受限搜索数据集中看到的更多的数据质量问题的示例。LUX、ZEPLIN和LZ分析团队还没有在数据质量研究中使用机器学习，其他团队主要在寻找波形异常，或者在受限数据集中而不是在更大数据集中寻找异常。需要对所有新探测器以及探测器的每个新数据采集周期进行数据质量研究。执行这些分析切割通常是科学研究中资源最密集的部分。探测器的高压性能是数据质量问题的重要因素，高压组件的额外电子发射是主要问题。学生将开始设计，工程和测试项目，为未来的时间投影室的高压电极网格做准备。一个新的实验将需要可以在原位组装的模块化网格，并且将通过光收集、工程和电场模拟来执行具有最大化光收集和电场均匀性的工程化这样的电极。不同规模的工程原型将被生产和测试。此外，将运行小规模测试电极，以建立初始电网运行和老化/调节的统计数据，以了解大规模电极电网的最佳调试计划。可以从实验者编织的线、商业网、蚀刻网和薄膜沉积来研究多电极技术。