The quality of the environment inside in buildings matters as much as the quality of the environment outdoors. Actually even more, as the health andcomfort of the building occupants depend on it. Indoor air quality (IAQ) and thermal comfort conditions (ITQ) as well as hygrometric aspect areconcerned to define indoor climate quality (ICQ) which is just part of Indoor Environmental Quality (IEQ). An enlarged concept of IEQ comprises fourmore quality aspects; sound (ISQ), lighting (ILQ), odor (IOQ), and vibration (IVQ). A high indoor environmental quality can increase health,wellbeing and productivity of building occupants, but also decrease costs for energy and building maintenance. The main aim of this paper is to presentmaterials and methods used on indoor environmental research, particularly focused on indoor air quality applications in data analysis. In the studies ofour research group measured parameters incorporate outdoor climate conditions, indoor thermal, hygrometric, and air quality aspects as well as energyconsumption readings including electricity, district heating, and water. Data analysis were performed using computational, visualization and groupingartificial neural network methods. Furthermore, additional external study cases utilizing artificial neural networks (ANN), model-based control, andbig data analysis are presented and compared. In this paper, on the whole 17 examples of intelligent data analysis and knowledge deployment arereviewed in the sense of the benefits of methods used. The results demonstrate that several intelligent computational methods can be applied to study andsolve indoor environmental problems. Artificial neural networks can be categorized by the way they process data through the network. Feed forward, ornonrecurrent, methods like Multilayer perceptron (MLP) were applied in the cases of electricity or cooling load prediction. Support vector machine (SVM)was used as activity classified for controlling an intelligent HVAC system. Several recurrent ANN-methods were applied for modelling andvisualization in the cases of studying indoor air quality (self-organizing maps, SOM), building electricity consumption (Principal component analysis,PCA compared to Case-base reasoning, CBR), and multi-variable data visualization using Sammon’s mapping as well as enhanced data visualizationusing Weighted Voting Superpositioning (WeVoS) methods called WeVOS-SOM and WeVos-ViSOM. All methods reviewed possess their ownbenefits in multiple IEQ application fields. The measurements and technology presented in this paper relies heavily on building automation systems(BAS) based big data, and the concept termed the “Internet of Things” (IoT). Summarized, this paper considers how to interpret and exploit the resultsof continuous measurements and knowledge discovery for researching and developing new information services.
Citation: ASHRAE and AIVC IAQ 2016 Conf
Product Details
- Published:
- 2016
- Number of Pages:
- 8
- Units of Measure:
- Dual
- File Size:
- 1 file , 470 KB
- Product Code(s):
- D-2016IAQ-24
