工业4.0时代下智能制造相关热点支撑技术趋势及进展（PPT课件）
工业4.0时代下，智能制造相关热点支撑技术趋势及进展,智能制造相关的几个基本概念 智能制造实施环境 智能制造相关热点支撑技术 ARC的建议,ARC咨询集团简介,成立于1986年，全球领先的研究及咨询机构，专注于制造业及基础设施领域。 主要研究领域： 传感设备，电机、驱动，自动化及控制，数字化油田，制造业信息化解决方案等。 智能制造及相关技术是ARC近年来关注的重点领域之一。 IIoT device management platforms IOT gateways IIoT devices total available market Industrial BI and analytics Cyber security Industrial IoT & Industrie 4.0 Viewpoints Newsletter,智能制造相关的几个概念,同为智能制造，Smart Manufacturing与Intelligent Manufacturing, Digital Manufacturing是一回事吗？参照下来自美国国家标准化委员会的定义。,智能制造相关的几个概念,智能制造 vs IIoT or Industrie 4.0 Smart manufacturing is more encompassing and includes all methodologies, processes and technologies that substantially improves the outcome of manufacturing, be it in the form of product value, quantity or quality, or in the form of productivity or reduced environmental footprint. 智能制造 vs 先进制造 先进制造 - This involves improvements in fundamental science or engineering, for example scientific advances such as photonics or chemical nanostructures or engineering improvements, such as modular production technology, additive manufacturing, or advanced forming. 智能制造 - This includes information, communication, or automation technologies applied to production processes and assembled to smart manufacturing systems.,智能制造相关的几个概念,下一步是什么？Industrie 4.1 or 5.0? 进化还是变革? 对于制造业用户而言，进化还是变革并不重要。稳步渐进的提升制造效率才是既可能也现实的生存之道。 并不仅仅是 IIoT or Industrie 4.0,智能制造相关的几个概念,智能制造战略及实施对于中国尤为重要 中国是制造业大国，制造业是立国之本 中国制造业面临日益加剧的竞争压力 还有一点，先进制造中短期内并非中国制造企业优势，即通过产品或者制造工艺的高技术附加值、复杂性提升来提高产品价值及竞争优势。短期内，先进制造领域发达国家仍保持很大优势。,智能制造实施环境,IIoT Smart Module,RemoteMonitoring & Services,Advanced Analytics,Smart Machines Connected Software-defined Agents, Apps Sensors e.g. vibration, ultrasound, infrared etc.,Connected Machine,智能制造是个生态系统,Production Flexible, responsive operations Maintenance optimization/Predictive techniques Connected workers, systems EH&S improvement Energy optimization,Customers Ongoing relationship vs. buy transaction OPEX vs. CAPEX Outsource non-core functions Service Level Agreements,Enterprise End-to-end, top-to-bottom, real-time performance New revenue opportunities Plan for disruptions,Services Remote monitoring, diagnostics, services CBM/predictive maintenance Field Service optimization New business models/services,Supply Chain Tracking, optimization Quality assessment/mgmt. Inventory tracking, optimization SC Control Tower,Design & Engineering Design for Connectivity In-service performance feedback improves design Software-based functionality Operating Ecosystem,工业云,Microsoft Azure, Amazon AWS, IBM BlueMix, 阿里云, 腾讯云等等 客户虽有担忧，但云服务供应商日子看起来过得挺滋润 工业云: GE Predix, Siemens MindSphere, Hitachi知造云等 工业云给制造企业带来客户价值 IT/OT融合的关键,GE Predix,2013年推出， 大致在今年2月对全球市场开放。 主要合作伙伴： 思科、微软、PTC等等 APM（资产性能管理）成为迁移至Predix云平台 的第一个应用。 创建了一个生态系统，合作伙伴可以基于平台开发 应用。超过2000名活跃的应用开放者（2016年2月）。 海尔在今年初成为Predix在中国第一个制造业客户。,Siemens MindSphere,2015年推出，服务于西门子数字化服务战略。 基于Sinalytics技术及SAP HANA平台 主要合作伙伴：SAP、Amazon、Atos、Accenture 目标客户是OEM及最终用户 服务类似：能源分析、机床分析、驱动设备分析及信息 安全分析,分析系统（Analytics Systems）框架,分析系统的关键功能,分析系统（Analytics Systems）框架,分析平台 vs. 分析应用,分析应用：基于多任务分析及ML平台,Predictive Maintenance Application,Inputs,Outputs,UI,Execution Code, APIs, Configuration Interface, etc,Analytics Engine (Platform),分析应用：基于云平台,Predictive Maintenance Application,Inputs,Outputs,UI,Execution Code, APIs, Configuration Interface, etc (as microservices),-GE Predix -SAP HANA Cloud Platform -IBM Bluemix -MS Azure -Etc.,基于云平台开发工业应用,Cloud Application Platform (PaaS),Launching an Application Application Code is pushed to version control system (Git remote) Then the platform: - builds the app artifact - constructs an application environment - deploys the application - provisions and links services - starts necessary processes,SAP HANA Cloud Platform,Create an App using platform or other tools,机器学习（Machine Learning）,Last Decade,Emerging and Future,Static Reports Static Dashboards Flowsheet Graphics Example: KPIs,Interactive Dashboards Trend Tools Excel X-Y Ploys Statistical Process (SPC),Machine Learning Algorithms Cross-Functional Context Applied to Process Data Supervised, Unsupervised,Big Data Aspects Asset Failure Codes SOPs Human Performance Data,什么是机器学习？,In Computer Science evolving form the study of pattern recognition and computational learning in Artificial Intelligence Goal is to have a machine mimic a human mind or behavior (abstract thinking, learning from past experience) For Problems that cannot be coded (Not Rule Based) ML Closely related to data mining and statistics Machine learning or AI (Artificial Intelligence) is a method of teaching computers to make predictions based on some data Leverages big data and predictive algorithms Determines a set of attributes to predict performance Supervised Regression- Predict a continuous value using labeled data Classification - Predict a class given a data set Un-Supervised Clustering Data (find groups or outliers) Recommender based on items can use all similar data sets,其他新兴热点技术,Augmented Reality,Cognitive Computing/AI,Predictive Analytics,Machine Learning,Str
