Forecasting office indoor CO2 concentration using machine learning with a one-year dataset

Kallio, Johanna; Tervonen, Jaakko; Räsänen, Pauli; Mäkynen, Riku; Koivusaari, Jani; Peltola, Johannes

doi:10.1016/j.buildenv.2020.107409

Cited by 53 publications

(29 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A direct comparison of results obtained from the given experimental analysis with existing studies in the literature is limited due to several factors [ 76 ]. The first important aspect is the difference in datasets used for model training and the type of methods [ 76 , 77 ]. Several researchers in the past have used IAQ and thermal comfort datasets from meteorological websites and air pollution control boards [ 78 , 79 ].…”

Section: Resultsmentioning

confidence: 99%

ADFIST: Adaptive Dynamic Fuzzy Inference System Tree Driven by Optimized Knowledge Base for Indoor Air Quality Assessment

Saini

Dutta

Marques

2022

Sensors

View full text Add to dashboard Cite

Air quality levels do not just affect climate change; rather, it leaves a significant impact on public health and wellbeing. Indoor air pollution is the major contributor to increased mortality and morbidity rates. This paper is focused on the assessment of indoor air quality based on several important pollutants (PM10, PM2.5, CO2, CO, tVOC, and NO2). These pollutants are responsible for potential health issues, including respiratory disease, central nervous system dysfunction, cardiovascular disease, and cancer. The pollutant concentrations were measured from a rural site in India using an Internet of Things-based sensor system. An Adaptive Dynamic Fuzzy Inference System Tree was implemented to process the field variables. The knowledge base for the proposed model was designed using a global optimization algorithm. However, the model was tuned using a local search algorithm to achieve enhanced prediction performance. The proposed model gives normalized root mean square error of 0.6679, 0.6218, 0.1077, 0.2585, 0.0667 and 0.0635 for PM10, PM2.5, CO2, CO, tVOC, and NO2, respectively. This approach was compared with the existing studies in the literature, and the approach was also validated against the online benchmark dataset.

show abstract

Section: Resultsmentioning

confidence: 99%

ADFIST: Adaptive Dynamic Fuzzy Inference System Tree Driven by Optimized Knowledge Base for Indoor Air Quality Assessment

Saini

Dutta

Marques

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Despite the promising results, a pure performance-oriented approach (i.e., which mainly focuses on prediction accuracy) might limit the applicability of such solutions in real-world scenarios. Indeed, the maximization of the performance is typically achieved by using a significant amount of data collected over a long time period [7][8][9][10] or by leveraging on complex input variables from expensive cutting-edge sensors [9,11]. In this regard, such a scenario requires data collection for many days, even months, for training the artificial intelligence models.…”

Section: Cloudmentioning

confidence: 99%

“…In this work we mainly refer to [10]. Unlike all the above studies, it provides a detailed overview of the prediction of indoor CO 2 , starting from the data collection architecture to the forecast results of the models, taking into account the challenges of using edge devices in such a problem.…”

Section: Related Workmentioning

confidence: 99%

“…In this work we used the data provided and published by Kallio J. et al [10]. As reported by the authors, this dataset has been published for further analysis of CO 2 predictions; thus, it is suitable for our objectives.…”

Section: Datasetmentioning

confidence: 99%

“…This is a major limitation, as forecasting CO 2 is crucial for applying preventive actions, e.g., by means of intelligent systems capable of automatically regulating HVAC devices in advance. Thanks to the increasing interest for IAQ, over the last decade a number of studies regarding the forecast of indoor CO 2 have been published [7][8][9][10][11][12][13][14]. The proposed solutions mainly focus on maximizing the CO 2 prediction accuracy by means of different artificial intelligence (AI) techniques.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Practical and Adaptive Approach to Predicting Indoor CO2

Segala

Doriguzzi-Corin

Peroni³

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

COVID-19 has underlined the importance of monitoring indoor air quality (IAQ) to guarantee safe conditions in enclosed environments. Due to its strict correlation with human presence, carbon dioxide (CO2) represents one of the pollutants that most affects environmental health. Therefore, forecasting future indoor CO2 plays a central role in taking preventive measures to keep CO2 level as low as possible. Unlike other research that aims to maximize the prediction accuracy, typically using data collected over many days, in this work we propose a practical approach for predicting indoor CO2 using a limited window of recent environmental data (i.e., temperature; humidity; CO2 of, e.g., a room, office or shop) for training neural network models, without the need for any kind of model pre-training. After just a week of data collection, the error of predictions was around 15 parts per million (ppm), which should enable the system to regulate heating, ventilation and air conditioning (HVAC) systems accurately. After a month of data we reduced the error to about 10 ppm, thereby achieving a high prediction accuracy in a short time from the beginning of the data collection. Once the desired mobile window size is reached, the model can be continuously updated by sliding the window over time, in order to guarantee long-term performance.

show abstract

Utilization of Machine Learning Techniques in Hot‐Film Based Airflow Rate Sensors for Improving Flow Measurement

Shin,

Baek,

Choi

et al. 2024

Advanced Intelligent Systems

View full text Add to dashboard Cite

This article presents a novel airflow rate sensing method based on the principle of a hot‐film flow sensing device with data‐driven machine learning (ML) models. In addition, to combine the two signals of the sensor (the resistance changes of the heaters) and predict the output (airflow rate), three different ML multivariate regression models, i.e., multiple linear regression (MLR), k‐nearest neighbor (KNN), and deep neural network (DNN) models, are trained and compared using 8400 experimentally obtained data. Using sensor fusion techniques, the average mean absolute error (MAE) and mean squared error (MSE) of the KNN model are determined to be 0.01522 and 0.00132, respectively, in the range of 0–5.07 standard liters per minute. Compared with the average results obtained using only a single input, those obtained using a dual input indicate a significant decrease in the MAE and MSE by 85.69% and 96.68%, respectively. Furthermore, a transient analysis of the ML‐based flow sensor is conducted to investigate the response time and transient characteristics of the MLR, KNN, and DNN models. The results of this study contribute to the advancement of airflow management systems for various industrial applications, such as building ventilation, gas leakage detection, and energy systems.

show abstract

Forecasting office indoor CO2 concentration using machine learning with a one-year dataset

Cited by 53 publications

References 23 publications

ADFIST: Adaptive Dynamic Fuzzy Inference System Tree Driven by Optimized Knowledge Base for Indoor Air Quality Assessment

ADFIST: Adaptive Dynamic Fuzzy Inference System Tree Driven by Optimized Knowledge Base for Indoor Air Quality Assessment

A Practical and Adaptive Approach to Predicting Indoor CO2

Utilization of Machine Learning Techniques in Hot‐Film Based Airflow Rate Sensors for Improving Flow Measurement

Contact Info

Product

Resources

About