Feasibility of random neural networks as an intelligent self-learning platform for cost effective deployment of energy harvesting compatible wireless sensors applied to building management systems

随机神经网络作为智能自学习平台的可行性，用于经济有效地部署应用于建筑管理系统的能量收集兼容无线传感器

• 批准号: 131608
• 金额: $ 14.45万
• 依托单位: GAS SENSING SOLUTIONS LTD.
• 依托单位国家: 英国
• 项目类别: Feasibility Studies
• 财政年份: 2014
• 资助国家: 英国
• 起止时间: 2014 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=131608
• 关键词: Feasibility; random; neural; networks; intelligent

This project addresses the need for energy harvesting compatible combined carbon dioxide / temperature /humidity wireless smart sensors, enabling people occupancy & indoor air quality (IAQ) monitoring in buildings and deployed within an intelligent self learning (ISL) network. Project output provides automated adaptive management of air conditioning systems to achieve required IAQ and minimise energy usage. The project focusses on establishing feasibility of using intelligent self learning networks based on a random neural network (RNN) approach, providing a control platform for simultaneous measurement of multiple wireless CO2/temperature/humidity sensor inputs The project combines unique patented energy harvesting compatible CO2/temperature/humidity wireless sensors, developed by project lead Gas Sensing Solutions, combined with innovative patented ISL network capability from Glasgow Caledonian University. The RNN wireless network methodology provides potential for low cost “fit and forget” deployment of CO2/humidity/ temperature wireless sensor ISL networks into existing and new buildings. The consortium includes non-funded end user assessment.

该项目解决了能量收集兼容的二氧化碳/温度/湿度无线智能传感器的需求，使建筑物中的人员占用和室内空气质量（IAQ）监测成为可能，并部署在智能自学习（ISL）网络中。项目输出提供空调系统的自动化自适应管理，以实现所需的IAQ并最大限度地减少能源使用。该项目的重点是建立基于随机神经网络（RNN）方法的智能自学习网络的可行性，为同时测量多个无线CO2/温度/湿度传感器输入提供控制平台该项目结合了独特的专利能量收集兼容CO2/温度/湿度无线传感器，由项目负责人开发气体传感解决方案，结合格拉斯哥喀里多尼亚大学的创新专利ISL网络功能。RNN无线网络方法提供了将CO2/湿度/温度无线传感器ISL网络低成本“适应和忘记”部署到现有和新建筑物中的潜力。该联合体包括非供资的最终用户评估。