Market Assessment for Deep Reinforcement Learning (DRL) based Shape Optimization for Heat Exchangers

基于深度强化学习 (DRL) 的热交换器形状优化的市场评估

• 批准号: 576544-2022
• 负责人: Hamdullahpur, FeridunX
• 金额: $ 1.07万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Idea to Innovation
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=760348
• 关键词: Market; Assessment; Deep; Reinforcement; Learning

University of Waterloo (UW) researchers Hamdullahpur (Professor, Department of Mechanical and Mechatronics Engineering) Kerameti (PhD candidate, Department of Mechanical and Mechatronics Engineering) have developed a new software-based methodology for the design of complex heat exchangers used in various industrial applications. The invention enables designers to design and build more effective heat exchanger (e.g., higher heat transfer rate, lower pressure drop) without using expensive and time consuming computational fluid dynamics or other numerical method software. Among several, the main advantage of this novel system include changing the geometry by reconfiguring the CAD file automatically toward a desired shape where other methods require conversion to CAD file for downstream tasks such as manufacturing. The combination of AI agents optimizes shapes that construct the geometry of the heat exchanger. This method is developed for the first time by the creators of the current invention. As part of this Market Assessment grant, the selected consultant will perform a thorough market analysis to identify the most suitable market opportunities along with technical challenges associated with the application of the technology in a particular market segment. A favorable market assessment result will further guide the development of a commercialization strategy and in attracting investments.

滑铁卢大学 (UW) 研究人员 Hamdullahpur（机械和机电工程系教授）Kerameti（机械和机电工程系博士研究生）开发了一种新的基于软件的方法，用于设计各种工业应用中使用的复杂热交换器。 本发明使设计者能够设计和构建更有效的热交换器（例如，更高的传热速率、更低的压降），而无需使用昂贵且耗时的计算流体动力学或其他数值方法软件。其中，该新颖系统的主要优点包括通过将 CAD 文件自动重新配置为所需的形状来更改几何形状，而其他方法需要转换为 CAD 文件以执行制造等下游任务。 人工智能代理的组合优化了构建热交换器几何形状的形状。该方法是由本发明的创造者首次开发的。作为市场评估拨款的一部分，选定的顾问将进行全面的市场分析，以确定最合适的市场机会以及与特定细分市场中的技术应用相关的技术挑战。良好的市场评估结果将进一步指导商业化战略的制定和吸引投资。