AUTONOM: Integrated through-life support for high-value systems

摘要：该提案扩展了新型传感、电子维护系统和决策策略的研究。它由EPSRC国家创新制造全寿命工程服务中心提供支持。维护广泛分散的资产是昂贵的，因为它涉及广泛的检查、检查和测量。在地理分散和结构较差的环境中集成基于传感器的信息在技术和成本方面提出了挑战。学术上的挑战包括嵌入式传感的改进，监测参数的可靠估计，数学和数据结构的统一方法，以及成本估计和效益分析的严格方法。产业驱动因素包括标准化、自动化、互联互通和单位成本降低。我们将在工业应用背景下建立自动化、智能维护的电子维护原则：从传感器级过渡到管理决策。在集成架构中，我们将深入研究一些研究挑战：-在固定和移动参与者之间交换维护数据的需求，例如铁路或航空航天应用中的轨道到列车数据交换。-在移动平台上的自动化监控，以及与其他参与者和固定系统的接口。-用于复杂分布式运行状况监控的新成本建模工具，特别是对业务的价值。其新颖之处在于实现了有效的信息流和分析技术，以支持优化和基于不确定性的资源调度，捕捉与数据滞后作斗争的业务“拉动”，以及将其转化为决策信息的工具不足。目的和目标：总体目标是在自动化、智能维护、响应业务拉动和展示价值的数据丰富的环境中，支持改进体系结构级别之间的集成。目标：-在技术丰富的数据源背景下，确定工厂和固定资产管理信息的工业拉动，以供维修决策；即从上到下设计基于状态的维护。-架构层之间的转换，从数据到信息再到决策，在适当的硬件（如无线平台和RFID）中嵌入自动功能。在固定演员和移动演员之间进行转换。-改进了大数据集状态监测问题的数据融合。-以业务流程的形式挖掘和优化重新配置维护数据的演示方法。-成本，价值和效益的评估工具，并经过验证。在一个由工业“拉动”指导的方案中，四个工作包将包括-整合，具有里程碑的“智库”事件报告；数据/信息和信息/决策转换的定义，用工业案例说明；设计策略文件；最终报告。-具有里程碑意义的数据融合和移动平台：融合多个非同步传感器数据的策略；融合多个非同步传感器数据的案例研究自顶向下监控系统设计策略报告/消息传递体系结构——概念验证。-基于具有里程碑的智能可重构业务流程的计划和调度：与合作伙伴合作，确定一组具有代表性的业务流程；多目标优化框架；以伴侣为榜样进行验证。-成本分析，具有里程碑：确定最佳实践成本建模方法，成本效益分析和价值分析；开发成本和价值模型；评估综合维修业务的成本和效益的框架；用于估计分布式异构高完整性资产的生命周期成本和收益的框架，以及工具集的验证。

Summary - This proposal extends research in novel sensing, e-maintenance systems, and decision-making strategies. It is supported by the EPSRC National Centre for Innovative Manufacturing in Through-life Engineering Services. Maintenance of widely-dispersed assets is expensive because it involves widespread inspection, checking and measurement. The integration of sensor-based information in geographically dispersed and less structured environments poses challenges in technology and cost justification. Academic challenges include improvement of embedded sensing, reliable estimation of monitoring parameters, a unified approach to the mathematics and data structures, and a rigorous approach to cost estimation and benefit analysis. The industrial drivers include standardisation, automation, connectivity, and reduction of unit cost. We will build on e-maintenance principles in automated, intelligent maintenance in the context of industrial application: transitions from sensor level to management decision-making. Within the integration architecture we will investigate some research challenges in depth:- Requirements for interchange of maintenance data between fixed and mobile actors, e.g. track-to-train data exchange in rail or aerospace applications.- Automation of monitoring on mobile platforms, and interface with other actors and fixed systems.- New cost modelling tools for complex distributed health monitoring, and especially the value to the business. The novelty lies in the implementation of effective information flows and analysis techniques to support optimisation, and uncertainty based resource scheduling, capturing business "pull" which struggles with lag in the data, and insufficient tools to turn it into information for decision making. Aims and Objectives: The overall aim is to enable improved integration between architectural levels in data-rich environments in automated, intelligent maintenance, responding to business pull and demonstrating value.Objectives:- Identification of the industrial pull for plant and fixed asset management information for maintenance decisions, in the context of technologically rich data sources; i.e. designing the condition based maintenance from the top down.- Transition between architectural levels, from data to information to decision, embedding automatic functionality in appropriate hardware such as wireless platforms and RFID. Transfer between fixed and mobile actors.- Improved data fusion in condition monitoring problems in large data sets.- A demonstrated methodology for mining and optimally reconfiguring maintenance data in the form of a business process.- An estimation tool for cost, value and benefits, with validated application.In a programme directed by industrial "pull", four work packages will cover - Integration, with milestones "Think tank" event report; Definition of data/information and information/decision transitions, illustrated by industrial cases; Strategy document for design; Final report.- Data fusion and mobile platforms, with milestones: Strategy for fusion of multiple unsynchronised sensor data; Case studies of fused multiple unsynchronised sensor data; Strategy for top-down monitoring system design; Reporting/message passing architecture - proof of concept.- Planning and Scheduling Based on Intelligent Reconfigurable Business Processes, with milestones: Identification of a representative set of business processes, in cooperation with the partners; The multi-objective optimisation framework; Validation on a partner's example.- Cost Analysis, with milestones: Identification of best practice cost modelling methods, and cost benefit analysis and value analysis; Development of cost and value models; Framework for estimating costs and benefits of integrated maintenance to the business; Framework for estimating through life costs and benefits of distributed heterogeneous high integrity assets, and Validation of tool set.