No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Ensemble-based extreme learning machine model for occupancy detection with ambient attributes
Abstract
Context-aware computing is a growing research domain in present circumstances due to technological advancements in the area of sensors technology, big data, artificial intelligence and robotics and automation. It has many applications for making the daily life of human beings sustainable, comfortable, and smooth. Context ware computing also includes ambient intelligence and applications such as occupancy detection, prediction, user recognition etc. Occupancy detection and recognition help in developing intelligent applications which help the energy management, intelligent decision making, that results in cost reduction and fault and failure prevention of services and products in advance. Several studies have been conducted to detect the occupancy with a different set of methodologies and approaches using varying types of data such as environmental parameters, image and video-based attributes, wireless or sensor based parameters, and noise-based parameters. This paper proposes a reliable, more accurate and efficient model based on the statistical analysis of the sensor based data for occupancy detection. Detailed quantification of the relationship of the ambient attributes is presented and the ensemble model is developed based on machine learning technique extreme learning machine to achieve the significant level of improvement in accuracy, efficiency, generalization and reliability. In addition to this, the paper also proposes one online and adaptive model-based online sequential extreme learning machine to perform occupancy detection on real-time data when complete data is not available and learning is done with recent data points coming in the form of streams. Results are compared with existing work in the domain and it is observed that proposed model perform better in terms of efficiency and accuracy over existing literature work.
... There are two known approaches to this topic: indirect (based on indoor environmental information) and direct (based on cameras). The main limitations of the direct approach are privacy concerns, hardware costs, intrusiveness, complex processing, and setup [5], [8]- [11]. Therefore, indirect approaches are preferred due to accessibility, lower hardware costs, and lower level of intrusiveness [8], [10], [12]. ...
... Many algorithms have been proposed for the detection of presence and occupancy with environmental data, such as feedforward neural networks [12], Multilayer Perceptron [16], Support Vector Machine (SVM) [8], [9], Gradient Boosting (GB) [9], Random Forests (RF) [4], [16], [17], or Extreme Learning Machine (ELM) [11], [17] with a mixed best result, between 0.6917 and 0.99 accuracy, 0.75 and 0.96 F1 score [14]. According to Chitu et al. [17], better results (accuracy of 0.91 and 0.69 for CO 2 ) are obtained for combinations of multiple environmental variables than considered separately, specifically for the combination of temperature, humidity, and CO 2 concentration. ...
... The accuracy of classifying the levels of occupancy (E, H, M, L) with Gradient Boosting is similar to those reported in [14] (accuracy of 0.9994 and 0.9834 for the Gym and Home data set, respectively) and in [13], [24] (0.9994 and 0.9540 for Gradient Boosting, in a gym and living room, respectively vs. 0.984 in and 0.95-0.99 for decision trees). However, we can observe mixed reports of the overall accuracy and F1 scores for different classifiers, similarly to those reported in [4], [8], [9], [11], [14], [16], [17]. The results indicate that Gradient Boosting is the most suitable method for indoor occupancy estimation based on raw ambient environment data, and SVM is the least suitable method, regardless of the applied kernel. ...
... Therefore, the International Energy Agency (IEA), in their Annex 66 [3] and Annex 79 [4], highlights the need for further explorations in this area. For example, the prediction of occupancy status in buildings and homes can lead to energy savings, between 30% and 50% of total consumption [2,5]. It is estimated that buildings represent from 30% to 40% of the total consumed energy and that such energy came mainly from fossil fuels. ...
... It is estimated that buildings represent from 30% to 40% of the total consumed energy and that such energy came mainly from fossil fuels. Thus, it is expected that in the next 20 years, the total energy consumption for buildings will increase by up to 40% [5][6][7]. ...
... Several systems are benefited by the occupancy information gathered in buildings, mainly the Heating, Ventilation, and Air Conditioners (HVAC) system. Similarly, real-time occupancy information can enable and benefit other building management services, such as security, emergency systems, fire systems, flow management, and automated energy management [5,[8][9][10]. However, occupancy detection systems face several challenges, such as reliability, adaptability, and accuracy. ...
The understanding of occupancy patterns has been identified as a key contributor to achieve improvements in energy efficiency in buildings since occupancy information can benefit different systems, such as HVAC (Heating, Ventilation, and Air Conditioners), lighting, security, and emergency. This has meant that in the past decade, researchers have focused on improving the precision of occupancy estimation in enclosed spaces. Although several works have been done, one of the less addressed issues, regarding occupancy research, has been the availability of data for contrasting experimental results. Therefore, the main contributions of this work are: (1) the generation of two robust datasets gathered in enclosed spaces (a fitness gym and a living room) labeled with occupancy levels, and (2) the evaluation of three Machine Learning algorithms using different temporal resolutions. The results show that the prediction of 3–4 occupancy levels using the temperature, humidity, and pressure values provides an accuracy of at least 97%.
... Many studies that have been published focused on the estimation of occupancy in enclosed buildings using machine learning algorithms. Amongst the recent research published in the past five years, environmental variables have been used extensively for the development of machine learning models for occupancy detection [4][5][6][7]. This is partly because these sensors used for the data collection are readily available in the market at economical prices. ...
... This is partly because these sensors used for the data collection are readily available in the market at economical prices. In a study conducted by Kumar et al., machine learning techniques were explored to develop an occupancy detection model based on environmental factors such as date and time, temperature, humidity, light, the level of carbon dioxide (CO 2 ) and humidity ratio [4]. Similarly, Vela et al. adjudged the KNN as the best performing machine learning classification model to estimate occupancy based on environmental variables (that is humidity, temperature, and pressure) [5]. ...
The control of thermostats of a heating, ventilation, and air-conditioning (HVAC) system installed in commercial and residential buildings remains a pertinent problem in building energy efficiency and thermal comfort research. The ability to determine the number of people at a particular time in an area is imperative for energy efficiency in order to condition only occupied regions and thermally deficient regions. In this study of the best features comparison for detecting the number of people in an area, feature extraction techniques including wavelet scattering, wavelet decomposition, grey-level co-occurrence matrix (GLCM) and feature maps convolution neural network (CNN) layers were explored using thermal camera imagery. Specifically, the pretrained CNN networks explored are the deep residual (Resnet-50) and visual geometry group (VGG-16) networks. The discriminating potential of Haar, Daubechies and Symlets wavelet statistics on different distributions of data were investigated. The performance of VGG-16 and ResNet-50 in an end-to-end manner utilizing transfer learning approach was investigated. Experimental results showed the classification and regression trees (CART) model trained on only GLCM and Haar wavelet statistic features, individually achieved accuracies of approximately 80% and 84%, respectively, in the detection problem. Moreover, k-nearest neighbors (KNN) trained on the combined features of GLCM and Haar wavelet statistics achieved an accuracy of approximately 86%. In addition, the performance accuracy of the multi classification support vector machine (SVM) trained on deep features obtained from layers of pretrained ResNet-50 and VGG-16 was between 96% and 97%. Furthermore, ResNet-50 transfer learning outperformed the VGG-16 transfer learning model for occupancy detection using thermal imagery. Overall, the SVM model trained on features extracted from wavelet scattering emerged as the best performing classifier with an accuracy of 100%. A principal component analysis (PCA) on the wavelet scattering features proved that the first twenty (20) principal components achieved a similar accuracy level instead of training on the whole feature set to reduce the execution time. The occupancy detection models can be integrated into HVAC control systems for energy efficiency and security systems, and aid in the distribution of resources to people in an area.
... The study also has methodological limitations and hence, other techniques such as Self Organising Map (SOM), Fuzzy C mean, Latent Class Analysis Model and Gaussian Mixture Model (GMM) techniques can be used alongside K-means that will help in the creation of more robust and distinct customer segments (Spoor, 2022). Future researchers can also explore that customer segmentation is done while taking cost as a constraint which is an essential attribute for meeting business goals more objectively (Kumar et al., 2020;Joun & Kim, 2023). ...
This study aims to identify clients that share similar traits and develop a new micro-segmentation strategy. Drawn from two marketing theories i.e., customer relationship and personalisation and using the Recency Frequency Monetary (RFM) technique, clusters from the K-means technique are created to predict the behaviour of the best and least contributing retail customers. Transactional data was extracted from a Business to Customer (B2C) hyperretail store in India comprising 10, 20, 284 transactions done by 2140 regular customers taking into account their recency, frequency and total spending. Based on RFM metric values across three heterogeneous segments, customers were characterised as toppers, moderated and churners. Analysis reveals that the most valuable customers have RFM scores as HHH (high recency, high frequency and high monetary value). These are the most loyal customers and retailers cannot afford to lose them. In micro-segmentation, stores should also prioritise retaining customers who have a recent shopping experience (medium recency) but do so infrequently, while spending larger sums. This can be achieved through tailored marketing strategies. Implications stand for both offline and online retail businesses to understand customer behaviour and tailor-made marketing strategies.
... Experimental results show that using FFNN is effective for estimating indoor occupants' number with normalized mean square error (NMSE) of 0.23. Furthermore, in (Kumar, Singh, and Singh 2020), authors implemented an occupancy detection approach based on building environmental data by emerging Extreme Learning Machine (ELM) and Online sequential ELM (OSELM) techniques. In fact, reported results show the efficiency of the proposed method in detecting occupancy by achieving an accuracy of 99.2%. ...
Comprehensive occupancy information in smart buildings has become more imperative in order to develop new control strategies in energy management systems. Several techniques can be used to collect occupancy information considering accurate sensing techniques, such as passive infrared (PIR), carbon dioxide (CO2) and different types of cameras (i.e., thermal, or optical cameras). Recent studies show the usefulness of integrating occupancy information into energy management systems to reduce energy consumption while maintaining the occupants’ comfort. The purpose of this work is to elaborate a comprehensive review on occupancy detection systems in smart buildings. This study presents a set of comparison standards including methods, occupancy resolution, type of buildings and sensors. A classification of different approaches, which can be implemented and integrated into the building management system for detecting indoor occupancy, is introduced. Summary and discussions are given by highlighting the usefulness of machine learning for enabling predictive control of active systems in smart buildings.
... LITERATURE SURVEY AND ACCURACIES[18] ...
... The acceleration of the harmful effects of climate change and the potential impact of occupancy information in energy efficiency have caused the increasing interest of researchers in the field of occupancy detection and estimation in the last decade [1][2][3][4]. Within this field, indirect approaches (based on indoor environmental information) have prevailed over direct approaches (based mainly on cameras) as viable alternatives for occupancy detection since the latter present high implementation cost and intrusion problems of privacy [4][5][6][7][8]. Therefore, much of the specialized literature that can be found nowadays presents solutions based on indirect approaches. ...
The estimation of occupancy is a crucial contribution to achieve improvements in energy efficiency. The drawback of data or incomplete data related to occupancy in enclosed spaces makes it challenging to develop new models focused on estimating occupancy with high accuracy. Furthermore, considerable variation in the monitored spaces also makes it difficult to compare the results of different approaches. This dataset comprises the indoor environmental information (pressure, altitude, humidity, and temperature) and the corresponding occupancy level for two different rooms: (1) a fitness gym and (2) a living room. The fitness gym data were collected for six days between 18 September and 2 October 2019, obtaining 10,125 objects with a 1 s resolution according to the following occupancy levels: low (2442 objects), medium (5325 objects), and high (2358 objects). The living room data were collected for 11 days between 14 May and 4 June 2020, obtaining 295,823 objects with a 1 s resolution, according to the following occupancy levels: empty (50,978 objects), low (202,613 objects), medium (35,410 objects), and high (6822 objects). Additionally, the number of fans turned on is provided for the living room data. The data are publicly available in the Mendeley Data repository. This dataset can be used to train and compare different machine learning, deep learning, and physical models for estimating occupancy at enclosed spaces.
Data Set: 10.17632/kjgrct2yn3.3.
Data Set License: CC BY 4.0.
... Hence, In this work, Modern DNN technique called Convolutional Neural Network(CNN) are used to solve the problem of manual feature selection and extraction in conventional automatic systems for MI diagnosis. Recently, deep learning techniques have proved notably better classification performance compared to the traditional machine learning methods used for various tasks [32]- [40]. CNN are the modern deep learning techniques consists of multiple layers like an input, hidden, and output layers [41]. ...
Myocardial infarction also called heart attack, is the most dangerous Coronary heart disease for humans beings. Portable Electrocardiogram(ECG) device is useful for the identification and control of ECG signals for myocardial infarction. These ECG signals record heart electrical activity and reflect the unusual movement of the heart. Visually, it is difficult to identify a variation in ECG due to its small amplitude and period. Therefore in this paper, we implemented a convolutional neural network (CNN) made of two layers of convolution-pooling, two dense layers and one output layer for the diagnosis of myocardial infarction using ECG. For batter performance, this network uses Leaky ReLU neurons with categorical cross-entropy loss function and the ADAM optimizer algorithm. To avoid the problem of overfitting, we used L2 regularisation method for regularization of the dense layer of CNN. For experimentation, we use the Physikalisch-Technische Bundesanstalt (PTB) diagnostic database. In this database, we obtained results of sensitivity, specificity, and accuracy of 100 %, 99.65%, and 99.82%, respectively, for data taken from the training set. And sensitivity, specificity, and accuracy of 99.88 %, 99.65%, and 99.82%, respectively, on patients, it hasn’t seen before which indicating that the model can achieve excellent classification performance.
This study explores the dynamics of a (3+1)-dimensional extended Kairat-X equation, revealing its connections to differential geometry of curves and equivalence principles. Using the Hirota bilinear method and linear superposition principle (LSP), we derived lump-periodic solutions, higher-order solitons with bifurcations, resonant multi-soliton waves, and positive complexiton solutions. To highlight the significance of these results, we present 3D, density and contour plots for specific parameter values. Also, we analyzed the modulation instability (MI) by determining the stability region and gain spectrum. This novel work offers valuable insights into the integrable model introduced by Wazwaz, and provides algorithms for investigating newly developed systems related to plasma physics, optical communications, marine dynamics, and the geometry of curves.
Ensemble learning, which involves combining the opinions of multiple experts to arrive at a better result, has been used for centuries. In this work, a review of the major voting methods in ensemble learning is explored. This work will focus on a new method for combining the results of individual learners. The method depends on the relative accuracy and diversity of teams. Instead of trying to assign weight to each different trainer, the concept of diversity teams is presented. Each team will vote as one player; however, the individual accuracies of each learner still be implemented. The concept of relaxing parameters that deal with each team as one player is presented. Our experiments demonstrate that traditional ensemble voting methods outperform individual learners. There is a limit to the superiority of the ensemble learner that any ensemble learner cannot go beyond. The proposed voting method gives the same results as the traditional ensemble voting methods, however, a different diversity of the proposed method from the traditional voting method or for different values of the relaxing parameter can be achieved.
The emergence of digital and their products, such as Facebook, Twitter, Reddit, Instagram, and other social media entities, has become the dominant medium for exchanging information, connecting people, expressing emotion and feelings, and influencing audiences and communities. Sometimes, these platforms are used by some social media users to engage in inappropriate behavior and expression by using offensive, hateful, and harassing content to express their views and dissatisfaction. Negative, hateful, and abusive content can hurt or harm other social media users and communities and can lead to law and order problems in society. There is a need to stop and mitigate the effects of hate speech by developing intelligent systems and models to detect them. Artificial Intelligence/machine learning can help in the detection of social media posts, comments, and replies that are hateful, abusive, and offensive. Such content also correlated with race, gender, sexuality, religion, and age. The objective is to evaluate the performance of the proposed model in comparison to other machine learning and deep learning algorithms on the newly collected data. The experimental results demonstrate that the proposed model, CNN with word embeddings, exhibits outstanding performance in hate speech detection. It achieves a remarkable accuracy of 83%. CNN’s word embeddings’ contextual understanding of language and its ability to capture complex semantic relationships contribute significantly to its superior performance compared to other traditional machine learning and deep learning models.
Buildings account for a significant portion of the global energy consumption. Forecasting personnel occupancy is critical for reducing energy consumption in buildings. This study explored the general process of building occupancy prediction models, and specifically analyzed the evolution and application of various data collection methods and predictive algorithms. A comprehensive research framework is established. The main findings indicate that prediction accuracy can be substantially improved by leveraging the Internet of Things technology to enhance data collection and employing hybrid machine learning algorithms. These advancements are vital to optimize building operation strategies, reduce energy consumption, and minimize carbon dioxide emissions. Additionally, the assessment metrics for validating predictive models are discussed and a novel idea based on integrated selection methods is presented. Differing from existing research, this study explores data collection methods and predictive algorithms from a broader perspective, also examining their interplay. Finally, potential directions for further development and improvement in the field are identified. The findings emphasize the necessity to continually innovate in data collection and algorithm development to meet evolving environmental needs and sustainability goals. New insights for engineering design and energy system optimization are offered.
Globally, cancer is the second largest cause of mortality. For the improvement of cancer diagnosis, gene expression data plays a significant role. Cancer detection using traditional approaches is too complex and time-consuming. As an alternative, machine learning techniques are the better option in terms of computational cost for this critical task. However, the analysis of these data is too complex as the raw data is huge, noisy, and contains redundant genes. Thus, effective preprocessing and a robust data classification strategy are required to be designed for cancer data analysis.
This research is based on designing a novel feature selection and data classification technique. The proposed technique begins by removing noisy genes from the data using gene selection techniques. For this, a combinational approach is followed, in which a pool of ordered four gene ranking approaches are adopted, ordered pipelines are built, and significant genes are extracted. Furthermore, not to bias with single classifier performance, a classifier ensemble is prepared considering five simple and improved extreme learning machine (ELM) models with the soft voting scheme for efficient data classification purposes. This experiment is conducted over seven microarray datasets.
After feature selection, ten frequently appearing features out of 16 pipelines are extracted for each dataset separately. Then, the proposed ensemble is compared with each individual classifier and the outcome is presented using four performance metrics. Overall, for all datasets, the performance of the ensemble is better over more than 74% of outcomes.
The concluding remark highlights the performance of the proposed design over a few considered state-of-the-art approaches.
Energy is a vital resource for smart cities and smart buildings. Maintaining Indoor temperature to a comfortable level requires energy consumption which comes from several resources renewable or nonrenewable. The indoor temperature depends on several parameters such as environmental attributes, building design and structural attributes, user activity etc. Predictions on energy consumption and indoor temperature have employed few data-driven model and machine learning approaches which require more accuracy, better generalisation and real-time estimation of energy and the indoor temperature. This paper proposed a hybrid model based on feature selection methods such as feature importance and support vector regression (SVR) to predict indoor temperature in a more accurate and efficient manner. The best model SVR gives MAE equals to 0.0174, MSE equals to 0.0006, RMSE equals to 0.0256 which are the least error in the prediction of indoor temperature. Random forest regressor (RFR) model gives MAE equals 0.0820, MSE equals 0.0101, RMSE equals 0.1006. RFR results are the worst in all prediction error matrices. Accuracy and efficiency have been compared with other machine learning models and the proposed model has outperformed them.
Digital technologies, their product and services have empowered the masses to generate information at a faster pace. Digital technologies based information sharing platforms such as news websites and social media platforms such as Facebook, Twitter, Instagram, What’s app etc have flooded the information space due to the easy generation of information and dissemination to the masses instantly. Information classification has been an important task, especially in newspapers and media organisations. In another area also, information or text classification has an important role to play so that important and vital information can be classified based on the already predefined categories. In journalism, editors and resources persons were allocated the task to recognise and classify the news stories so that they can be placed in the predefined categories of economy and business news, political news, social news, editorial section, education and career, and sports information etc. Nowadays the process of classification and segregation of textual information has become challenging due to the flow of diverse, vast information. Additionally, the pace of information and its updates, access and competition among the media House have made it more challenging. Hence automated and intelligent tools which can classify the information and text accurately and efficiently is needed to reduces human efforts, time and increase productivity. This paper presents an intelligent, efficient and robust intelligent machine learning model based on Multinomial Naive Bayes(MNB) to classify the current affairs news stories. The proposed Inverse Document Frequency(IDF) integrated MNB model achieves classification accuracy of 87.22 per cent. The experiment results are also compared with other machine learning models such as Logistics Regression(LR), Support Vector Machine(SVM), K-Nearest Neighbours(KNN) and Random forest(RF). The results demonstrate that the presented model is better in term of accuracy and may be deployed in real world information classification and media domain to improve the productivity, efficiency of the current affairs news classification process.
Given a coarse satellite image and a fine satellite image of a particular location taken at the same time, the high-resolution spatiotemporal image fusion technique involves understanding the spatial correlation between the pixels of both images and using it to generate a finer image for a given coarse (or test) image taken at a later time. This technique is extensively used for monitoring agricultural land cover, forest cover, etc. The two key issues in this technique are: (i) handling missing pixel data and (ii) improving the prediction accuracy of the fine image generated from the given test coarse image. This paper tackles these two issues by proposing an efficient method consisting of the following three basic steps: (i) imputation of missing pixels using neighborhood information, (ii) cross-scale matching to adjust both the Point Spread Functions Effect (PSF) and geo-registration errors between the course and high-resolution images, and (iii) error-based modulation, which uses pixel-based multiplicative factors and residuals to fix the error caused due to modulation of temporal changes. The experimental results on the real-world satellite imagery datasets demonstrate that the proposed model outperforms the state-of-art by accurately producing the high-resolution satellite images closer to the ground truth.
The control of Heating, Ventilation, and Air Conditioning (HVAC) system automatically is one of the progressive areas of research. The collective importance of the HVAC system is to maintain indoor thermal comfort while ensuring energy efficiency. This study explores the thermal comfort, acceptability, preference, and sensation of fifteen subjects from February to September 2021. Multiclass-multioutput Decision Tree, Extra Trees, K-Nearest Neighbors and Random Forest classification models were developed to predict the thermal comfort metrics, of subjects in a room based on gender, age, indoor temperature, humidity, carbon dioxide concentration, activity level and time series features. It is important to understand occupants’ thermal comfort in real time to automatically control the environment. The best mean accuracy and mean squared error of 68% and 2.15 respectively was achieved by multiclass-multioutput Extra Tree classification model, when all the features were used in training and testing. Through this study, the feasibility of using machine learning techniques to predict thermal comfort, preference, acceptability, and sensation at the same time for HVAC control was established.
In the present-day technology-driven world, information reaching at the individual’s doorstep sometimes becomes complex, haphazard and difficult to classify to get the insights. The endpoint consumer of the information requires processed information which is contextually suited to their needs, interests and is properly formatted and categorised. Interests and need-based categorization of news and stories would enable the user beforehand to further evaluate information deeply. For instances, the type current affairs related issues and news to read or not to read. This research work proposes an advanced current affairs classification model based on deep learning approaches called Intelligent Current Affairs Classification Using Deep Learning (iCACD). The proposed model is better than already proposed models based on machine learning approached which have been compared on accuracy and performance criteria. The proposed model is better in the following ways. Firstly, It is based on advanced deep neural network architecture. Secondly, the model advances the work to include both headline and body of the information/news articles rather than only processing headlines. Thirdly, A detailed comparative analysis and discussion on accuracy and performance with other machine leaning models have also been presented.
The field of service operations management has a plethora of research opportunities to capitalise on, which are nowadays heightened by the presence of big data. In this research, we review and analyse the current state-of-the-art of the literature on big data for service operations management. To this aim, we use the Scopus database and the VOSviewer visualisation software for bibliometric analysis to highlight developments in research and application. Our analysis reveals patterns in scientific outputs and serves as a guide for global research trends in big data for service operations management. Some exciting directions for the future include research on building big data-driven analytical models which are deployable in the Cloud, as well as more interdisciplinary research that integrates traditional modes of enquiry with for example, behavioural approaches, with a blend of analytical and empirical methods.
Numerous research studies have claimed that search-based algorithms have the potential to be effectively used in various software engineering domains. An important task in software organizations is to efficiently recognize change prone classes of a software, as it is crucial to plan efficient resource utilization and to take precautionary design measures as early as possible in the software product lifecycle. This assures development of good quality software products at lower costs. The current study attempts to evaluate the capability of search-based algorithms while developing prediction models for identification of the change prone classes in a software. Though previous literature has evaluated the use of statistical category and machine learning category of algorithms in this domain, the suitability of search-based algorithms needs extensive investigation in this area. Furthermore, the study compares the performance of search-based classifiers with statistical and machine learning classifiers, by empirically validating the results on fourteen open source data sets. The results indicate comparable and in some cases even better performance of search based algorithms in comparison to other evaluated categories of algorithms.
The performance and learning speed of the Cascade Correlation neural network (CasCor) may not be optimal because of redundant hidden units’ in the cascade architecture and the tuning of connection weights. This study explores the limitations of CasCor and its variants and proposes a novel constructive neural network (CNN). The basic idea is to compute the input connection weights by generating linearly independent hidden units from the orthogonal linear transformation, and the output connection weights by connecting hidden units in a linear relationship to the output units. The work is unique in that few attempts have been made to analytically determine the connection weights on both sides of the network. Experimental work on real energy application problems such as predicting powerplant electrical energy, predicting seismic hazards to prevent fatal accidents and reducing energy consumption by predicting building occupancy detection shows that analytically calculating the connection weights and generating non-redundant hidden units improves the convergence of the network. The proposed CNN is compared with that of the state-of-the-art machine learning algorithms. The work demonstrates that proposed CNN predicts a wide range of applications better than other methods.
The operating efficiency of heating, ventilation and air conditioning (HVAC) system is critical for building energy performance. Demand-based control is an efficient HVAC operating strategy, which can provide an appropriate level of HVAC services based on the recognition of actual cooling “demand.” The cooling demand primarily relies on the accurate detection of occupancy. The current researches of demand-based HVAC control tend to detect the occupant count using cameras or other sensors, which often impose high computation and costs with limited real-life applications. Instead of detecting the occupant count, this paper proposes to detect the occupancy density. The occupancy density (estimated by image foreground moving pixels) together with the indoor and outdoor information (acquired from existing sensors) are used as inputs to an artificial neural network model for cooling demand estimation. Experiments have been implemented in a university design studio. Results show that, by adding the occupancy density, the cooling demand estimation error is greatly reduced by 67.4% and the R value is improved from 0.75 to 0.96. The proposed approach also features low-cost, computationally efficient, privacy-friendly and easily implementable. It shows good application potentials and can be readily incorporated into existing building management systems for improving energy efficiency.
The accuracy of the prediction of occupancy in an office room using data from light, temperature, humidity and CO2 sensors has been evaluated with different statistical classification models using the open source program R. Three data sets were used in this work, one for training, and two for testing the models considering the office door opened and closed during occupancy. Typically the best accuracies (ranging from 95% to 99%) are obtained from training Linear Discriminant Analysis (LDA), Classification and Regression Trees (CART) and Random Forest (RF) models. The results show that a proper selection of features together with an appropriate classification model can have an important impact on the accuracy prediction. Information from the time stamp has been included in the models, and usually it increases the accuracy of the detection. Interestingly, using only one predictor (temperature) the LDA model was able to estimate the occupancy with accuracies of 85% and 83% in the two testing sets.
Commercial office buildings represent the largest in floor area in most developed countries and utilize substantial amount of energy in the provision of building services to satisfy occupants’ comfort needs. This makes office buildings a target for occupant-driven demand control measures, which have been demonstrated as having huge potential to improve energy efficiency. The application of occupant-driven demand control measures in buildings, most especially in the control of thermal, visual and indoor air quality providing systems, which account for over 30% of the energy consumed in a typical office building is however hampered due to the lack of comprehensive fine-grained occupancy information. Given that comprehensive fine-grained occupancy information improves the performance of demand-driven measures, this paper presents a review of common existing systems utilized in buildings for occupancy detection. Furthermore, experimental results from the performance evaluation of chair sensors in an office building for providing fine-grained occupancy information for demand-driven control applications are presented.
Extreme learning machine (ELM) has gained increasing interest from various research fields recently. In this review, we aim to report the current state of the theoretical research and practical advances on this subject. We first give an overview of ELM from the theoretical perspective, including the interpolation theory, universal approximation capability, and generalization ability. Then we focus on the various improvements made to ELM which further improve its stability, sparsity and accuracy under general or specific conditions. Apart from classification and regression, ELM has recently been extended for clustering, feature selection, representational learning and many other learning tasks. These newly emerging algorithms greatly expand the applications of ELM. From implementation aspect, hardware implementation and parallel computation techniques have substantially sped up the training of ELM, making it feasible for big data processing and real-time reasoning. Due to its remarkable efficiency, simplicity, and impressive generalization performance, ELM have been applied in a variety of domains, such as biomedical engineering, computer vision, system identification, control and robotics, etc. In this review, we try to provide a comprehensive view of these advances in ELM together with its future perspectives.
We address the problem of estimating the occupancy levels in rooms using the information available in standard HVAC systems. Instead of employing dedicated devices, we exploit the significant statistical correlations between the occupancy levels and the CO2 concentration, room temperature, and ventilation actuation signals in order to identify a dynamic model. The building occupancy estimation problem is formulated as a regularized deconvolution problem, where the estimated occupancy is the input that, when injected into the identified model, best explains the currently measured CO2 levels. Since occupancy levels are piecewise constant, the zero norm of occupancy is plugged into the cost function to penalize non-piecewise constant inputs. The problem then is seen as a particular case of fused-lasso estimator by relaxing the zero norm into the ℓ1 norm. We propose both online and offline estimators; the latter is shown to perform favorably compared to other data-based building occupancy estimators. Results on a real testbed show that the MSE of the proposed scheme, trained on a one-week-long dataset, is half the MSE of equivalent Neural Network (NN) or Support Vector Machine (SVM) estimation strategies.
Heating, ventilation, and air conditioning (HVAC) is a major energy consumer in buildings, and implementing demand driven HVAC operations is a way to reduce HVAC related energy consumption. This relies on the availability of occupancy information, which determines peak/off-hour modes that impact cooling/heating loads of HVAC systems. This research proposes an occupancy estimation model that is built on a combination of non-intrusive sensors that can detect indoor temperature, humidity, CO2 concentration, light, sound and motion. Sensor data is processed in real time using a radial basis function (RBF) neural network to estimate the number of occupants. Field tests carried out in two shared lab spaces for 20 consecutive days report an overall detection rate of 87.62% for self-estimation and 64.83% for cross-estimation. The results indicate the ability of the proposed system to monitor the occupancy information of multi-occupancy spaces in real time, supporting demand driven HVAC operations.
Extreme learning machine (ELM) is a new learning algorithm for the single hidden layer feedforward neural networks. Compared with the conventional neural network learning algorithm it overcomes the slow training speed and over-fitting problems. ELM is based on empirical risk minimization theory and its learning process needs only a single iteration. The algorithm avoids multiple iterations and local minimization. It has been used in various fields and applications because of better generalization ability, robustness, and controllability and fast learning rate. In this paper, we make a review of ELM latest research progress about the algorithms, theory and applications. It first analyzes the theory and the algorithm ideas of ELM, then tracking describes the latest progress of ELM in recent years, including the model and specific applications of ELM, finally points out the research and development prospects of ELM in the future.
There has been extensive research focusing on developing smart environments by integrating data mining techniques into environments that are equipped with sensors and actuators. The ultimate goal is to reduce the energy consumption in buildings while maintaining a maximum comfort level for occupants. However, there are few studies successfully demonstrating energy savings from occupancy behavioural patterns that have been learned in a smart environment because of a lack of a formal connection to building energy management systems. In this study, the objective is to develop and implement algorithms for sensor-based modelling and prediction of user behaviour in intelligent buildings and connect the behavioural patterns to building energy and comfort management systems through simulation tools. The results are tested on data from a room equipped with a distributed set of sensors, and building simulations through EnergyPlus suggest potential energy savings of 30% while maintaining an indoor comfort level when compared with other basic energy savings HVAC control strategies.
Buildings are among the largest consumers of electricity in the US. A significant portion of this energy use in buildings can be attributed to HVAC systems used to maintain comfort for occupants. In most cases these building HVAC systems run on fixed schedules and do not employ any fine grained control based on detailed occupancy information. In this paper we present the design and implementation of a presence sensor platform that can be used for accurate occupancy detection at the level of individual offices. Our presence sensor is low-cost, wireless, and incrementally deployable within existing buildings. Using a pilot deployment of our system across ten offices over a two week period we identify significant opportunities for energy savings due to periods of vacancy. Our energy measurements show that our presence node has an estimated battery lifetime of over five years, while detecting occupancy accurately. Furthermore, using a building simulation framework and the occupancy information from our testbed, we show potential energy savings from 10% to 15% using our system.
Current climate control systems often rely on building regulation maximum occupancy numbers for maintaining proper temperatures. However, in many situations, there are rooms that are used infrequently, and may be heated or cooled needlessly. Having knowledge regarding occupancy and being able to accurately predict usage patterns may al-low significant energy-savings by intelligent control of the L-HVAC systems. In this paper, we report on the deploy-ment of a wireless camera sensor network for collecting data regarding occupancy in a large multi-function building. The system estimates occupancy with an accuracy of 80%. Using data collected from this system, we construct multivariate Gaussian and agent based models for predicting user mobil-ity patterns in buildings. Using these models, we can predict room usage thereby enabling us to control the HVAC systems in an adaptive manner. Our simulations indicate a 14% re-duction in HVAC energy usage by having an optimal control strategy based on occupancy estimates and usage patterns.
Knowing the presence or the actual number of occupants in a space at any given time is essential for the effective management of various building operation functions such as security and environmental control (e.g., lighting, HVAC). As occupants "interact" with the indoor environment, they will affect environmental conditions through the emission of CO2, heat and sound, and relatively little effort has been reported in the literature on utilizing this environmental sensing data for occupancy detection. This paper presents the findings of a study to address this question by exploring the most effective environmental features for occupancy level detection. A sensor network with robust, non-intrusive sensors such as CO2, temperature, relative humidity, and acoustics is deployed in an open-plan office space. Using information theory, the physical correlation between the number of occupants and various combination of features extracted from sensor data has been studied. The results show significant correlation between features extracted from humidity, acoustics, and CO2, while little correlation with temperature data. Using features from multiple sensors increases correlation further, and nearly 90% information gain is acquired when nine of the most informative features are combined.
This paper provides a viable and practical solution to the challenge of real-time estimation of the number of people in areas
of a building, during an emergency egress situation. Such estimates would be extremely valuable to first responders to aid
in egress management, search-and-rescue, and other emergency response tactics. The approach of this paper uses an extended
Kalman filter, which combines sensor readings and a dynamic stochastic model of people movement. The approach is demonstrated
using two types of sensors: video with real-time signal processing to detect number of people moving in each direction across
a threshold such as an entrance/exit, and passive infra-red motion sensors that detect people occupancy within its field of
view. The people movement model uses the key idea that each room has a “high-density” and “low-density” area, where high-density
corresponds to a queue of people at a bottleneck exit doorway, and low-density represents unconstrained flow of people. Another
key feature of the approach is that constraints on occupancy levels and people flow rates are used to improve the estimation
accuracy. The approach is tested using a stochastic discrete-time simulation model of a 1500 square meter office building
with occupancy up to 100 people, having a video camera at each of the three exits, and motion sensors in each of the 42 office
rooms. The simulation includes stochastic models of video sensors having a probability of detection of 98%, and motion sensors
with probability of detection of80%. Averaged over 100 simulation runs and averaged over the evacuation time, the sensor-only
approach produced a mean estimation error per room of 0.35 people, the Kalman filter with cameras only had a mean error of
0.14 people, and the Kalman filter with all sensors produced a mean error of 0.09 people. These results show that an effective
combination of models and sensors greatly improves estimation accuracy compared to the state-of-the-art practice of using
sensors only.
Heating, cooling and ventilation accounts for 35% energy usage in the United States. Currently, most modern buildings still condition rooms assuming maximum occupancy rather than actual usage. As a result, rooms are often over-conditioned needlessly. Thus, in order to achieve efficient conditioning, we require knowledge of occupancy. This paper shows how real time occupancy data from a wireless sensor network can be used to create occupancy models which in turn can be integrated into building conditioning system for usage based demand control conditioning strategies. Using strategies based on sensor network occupancy model predictions, we show that it is possible to achieve 42% annual energy savings while still maintaining American Society of Heating, Refrigerating and Air-Conditioning (ASHRAE) comfort standards.
We introduce the sensor-utility-network (SUN) method for occupancy estimation in buildings. Based on inputs from a variety of sensor measurements, along with historical data regarding building utilization, the SUN estimator produces occupancy estimates through the solution of a receding-horizon convex optimization problem. State-of-the-art on-line occupancy algorithms rely on indirect measurements, such as CO2 levels, or people counting sensors which are subject to significant errors and cost. The newly proposed method was evaluated via experiments in an office building environment. Estimation accuracy is shown to improve significantly when all available data is incorporated in the estimator. In particular, it is found that the average estimation error at the building level is reduced from 70% to 11% using the SUN estimator, when compared to the naive approach that relies solely on flow measurements.
It is clear that the learning speed of feedforward neural networks is in general far slower than required and it has been a major bottleneck in their applications for past decades. Two key reasons behind may be: (1) the slow gradient-based learning algorithms are extensively used to train neural networks, and (2) all the parameters of the networks are tuned iteratively by using such learning algorithms. Unlike these conventional implementations, this paper proposes a new learning algorithm called extreme learning machine (ELM) for single-hidden layer feedforward neural networks (SLFNs) which randomly chooses hidden nodes and analytically determines the output weights of SLFNs. In theory, this algorithm tends to provide good generalization performance at extremely fast learning speed. The experimental results based on a few artificial and real benchmark function approximation and classification problems including very large complex applications show that the new algorithm can produce good generalization performance in most cases and can learn thousands of times faster than conventional popular learning algorithms for feedforward neural networks.1
In modern buildings which are becoming smart day by day, indoor temperature can be forecasted with the data obtained from outfitted sensors. Predictive models based on data can accurately forecast temperature which further saves energy by optimisation of resources and technologies such as heating, ventilation and air conditioners for making the atmosphere conducive and comfortable in the ambient environment. This paper discusses an experiment from such house fitted with sensors for different parameter affecting indoor temperature. We propose a hybrid model which is based on particle swarm optimisation and emerging extreme learning machine to forecast the temperature to optimise the use of energy further. The proposed hybrid model also includes the variant online sequential extreme learning machine(OSELM) which accepts data online and is adaptive to the changing environmental conditions. We perform experiment based on many sensors combinations affecting temperature using particle swarm optimization and statistical tools to determine their relevance and correlation with temperature and compare results with conventional methods. Proposed methods improved the accuracy of the forecasting and also generalisation performance over other methods on the same dataset.
In the present day environment, smart buildings require optimization of energy consumption through monitoring, consumption prediction and making policy decisions accordingly. Attributes related to building design and structure play a vital role in heating load(HL) and cooling load(CL) of the building which directly affects the energy performance of the buildings. For prediction of HL and CL, emerging machine learning approaches can help in improving accuracy and efficiency in real time. This paper provides improvements in energy load assessment of the buildings. It is the first is the in-depth study and analysis of design and structural attributes and their correlation with HL and CL, the novel methods based on ELM and its variants online sequential ELM(OSELM) to predict HL and CL. This study also proposes OSELM based online/real-time prediction when data is coming in stream The total 24 models have been developed including 12 models based on ELM and 12 models based on OSELM with different feature sets and activation functions. Models have been compared on the basis of accuracy, computational performance and efficiency with few existing models. The experimental results show that the proposed models learn better and outperform other popular machine learning approaches such as the artificial neural network(ANNs), support vector machine(SVM), radial basis function network(RBFN), random forest(RF) and existing work in the energy and building domain.
Optimization of energy performance of the buildings has been a great interest of research. Energy performance monitoring and maintenance depend on the several parameters such as temperature, humidity, sunlight, roof area, wall area and mostly on heating ventilation and air conditioning(HVAC). Each building has its own patterns and settings obtained through monitoring which depends on many parameters such as users activities, environmental related attributes and building design structural parameters etc. In this respect, an estimation of energy load of the building in real-time efficiently for optimization becomes an important task for cost-effective energy management. The paper proposes an Intra ensemble model based on variants of emerging machine learning approach which includes extreme learning machine(ELM), Online Sequential ELM(OSELM) and Bidirectional ELM(B-ELM). The energy performance estimation requires the model to be real-time and efficient. This asks for use of highly correlated parameters and a very efficient model. For this, OSELM based model for real-time prediction of energy performance has been used. ELM variants are used because of their fast computation and efficiency of prediction over conventional machine learning models. The proposed model has been compared with few state of art methods on accuracy and efficiency criteria and proposed models outperformed existing methods.
Drowsiness during driving is a major cause of accidents of drivers which has socio-economic and psychological impact on the affected person. In Intelligent Transportation Systems (ITS), the detection of the drowsy and alert state of the driver is an interesting research problem. This paper proposed a novel method to detect the drowsy state of the driver based on three parameters, namely physiological, environmental and vehicular. The undertaken model proposes a simplistic approach and achieves comparable results to the state of the art with an ROC score of 81.28 and also elaborates on the specificity and sensitivity metrics.
The Indian health system is becoming very complex and large and faces a lack of developed policy, time bound and real-time solutions, and tracking and absence of advanced data collection and data analysis technologies. The problem becomes more complex when collected data are merged and analysed. These merged data result in the categories of big data having more dimensions which need more sophisticated approaches and intelligent systems for getting useful information to be used further in policy and decision making. An intelligent health system would result in better health policy making, execution, and faster modification if something is not right. The objective of this chapter is to present the new framework in the health care sector and enhance the idea of innovative discussions on how government schemes approach big data analytics to develop a public health system. This chapter first discusses the health sector problem in India and analyses the solution with the integration of machine learning and big data analytics approaches. It also proposes an intelligent architecture of a machine learning framework for developing accurate, effective, decentralised, and dynamic health insights for policy decision making to resolve health-related issues.
Following in the footsteps of the renowned report “Control in an Information Rich World,” Report of the Panel on “Future Directions in Control, Dynamics, and Systems” chaired by Richard Murray (2002), this paper aims to demonstrate that Systems & Control is at the heart of the Information and Communication Technologies to most application domains. As such, Systems & Control should be acknowledged as a priority by funding agencies and supported at the levels necessary to enable technologies addressing critical societal challenges. A second intention of this paper is to present to the industrials and the young research generation, a global picture of the societal and research challenges where the discipline of Systems & Control will play a key role. Throughout, this paper demonstrates the extremely rich, current and future, cross-fertilization between five critical societal challenges and seven key research and innovation Systems & Control scientific challenges. This paper is authored by members of the IFAC Task Road Map Committee, established following the 19th IFAC World Congress in Cape Town. Other experts who authored specific parts are listed below.
At present, most of the buildings are using process of heating, ventilation and air conditioning (HVAC)-systems. HVAC systems are also responsible for consumption of huge amount of energy. Home automation techniques are being used to reduce the waste of resources especially energy that is available to us in the form of temperature, electricity, water, sunlight, etc. Forecasting and predicting the future demand of the energy can help us to maintain and to reduce the cost of energy in the buildings. In this paper, we use the experiments the small medium large system (SML system) which is the house built at the university of CEU cardinal Herrera (CEU-UCH) for competition named Solar Decathlon 2013. With the data available from this experiments, we try to predict and forecast the future temperature condition intelligently for energy conservation with the model based on Extreme Learning Machine (ELM). This will help in determining the energy needs of the buildings and further will help in efficient utilization and conservation of energy.
We present results from a week-long experimental evaluation of a scalable control algorithm for a commercial building heating, ventilation, and air-conditioning (HVAC) system. The experiments showed that the controller resulted in 37% energy savings without sacrificing indoor climate. In contrast to prior work that reports energy savings without a careful measure of the effect on indoor climate, we verify that the controller achieves the energy efficiency improvements without any adverse effect on the indoor climate compared to the building's baseline controller. This is established from measurements of a host of environmental variables and analysis of before-after occupant survey results. We present a complete system to retrofit existing buildings including the control algorithm and the supporting execution platform which includes the deployment of a wireless sensor network. Results show that there is a large variation in energy savings from zone to zone, which indicates that estimating energy savings potential of novel HVAC control systems is not trivial even from experiments-something that prior work with uniformly positive messages did not emphasize.
Extreme learning machines (ELMs) basically give answers to two fundamental learning problems: (1) Can fundamentals of learning (i.e., feature learning, clustering, regression and classification) be made without tuning hidden neurons (including biological neurons) even when the output shapes and function modeling of these neurons are unknown? (2) Does there exist unified framework for feedforward neural networks and feature space methods? ELMs that have built some tangible links between machine learning techniques and biological learning mechanisms have recently attracted increasing attention of researchers in widespread research areas. This paper provides an insight into ELMs in three aspects, viz: random neurons, random features and kernels. This paper also shows that in theory ELMs (with the same kernels) tend to outperform support vector machine and its variants in both regression and classification applications with much easier implementation.
Heating, cooling and ventilation accounts for 35% energy usage in the United States. Currently,mostmodern buildings still condition rooms assuming maximum occupancy rather than actual usage. As a result, rooms are often over-conditioned needlessly. Thus, in order to achieve efficient conditioning, we require knowledge of occupancy. This article shows how real time occupancy data from a wireless sensor network can be used to create occupancy models, which in turn can be integrated into building conditioning system for usage-based demand control conditioning strategies. Using strategies based on sensor network occupancy model predictions, we show that it is possible to achieve 42% annual energy savings while still maintaining American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) comfort standards.
In order to achieve sustainability, steps must be taken to reduce energy consumption. In particular, heating, cooling, and ventilation systems, which account for 42% of the energy consumed by US buildings in 2010 [8], must be made more efficient. In this paper, we demonstrate ThermoSense, a new system for estimating occupancy. Using this system we are able to condition rooms based on usage. Rather than fully conditioning empty or partially filled spaces, we can control ventilation based on near real-time estimates of occupancy and temperature using conditioning schedules learned from occupant usage patterns. ThermoSense uses a novel multisensor node that utilizes a low-cost, low-power thermal sensor array along with a passive infrared sensor. By using a novel processing pipeline and sensor fusion, we show that our system is able measure occupancy with a RMSE of only ≈0.35 persons. By conditioning spaces based on occupancy, we show that we can save 25% energy annually while maintaining room temperature effectiveness.
Detecting when a household is occupied by its residents is fundamental to enable a number of home automation applications. Current systems for occupancy detection usually require the installation of dedicated sensors, like passive infrared sensors, magnetic reed switches, or cameras. In this paper, we investigate the suitability of digital electricity meters -- which are already available in millions of households worldwide -- to be used as occupancy sensors. To this end, we have collected fine-grained electricity consumption data along with ground-truth occupancy information for 5 households during a period of about 8 months. Our results show that using common classification methods it is possible to achieve occupancy detection accuracies of more than 80%.
With ever-rising energy demand and diminishing sources of inexpensive energy resources, energy conservation has become an increasingly important topic. Building heating, ventilation, and air conditioning (HVAC) systems are consid-ered to be a prime target for energy conservation due to their significant contribution to commercial buildings' energy consumption in the US. Knowing a building's occupancy plays a crucial role in implementing demand-response HVAC controls, with a corresponding potential for reduction of HVAC energy consumption, especially in office buildings. This paper evaluates occupancy modeling (both binary detection and multi-class estimation) using twelve ambient sensor vari-ables. Performance of six machine-learning techniques is evaluated in both single-occupancy and multi-occupancy offices. Of the six, the decision-tree technique yielded the best overall accuracy (i.e. 96.0% to 98.2%) and root mean square error (RMSE) (i.e. 0.109 to 0.156). The contribution of each individual ambient sensor variable is evaluated via informa-tion gain. It is found that CO 2 , door status, and light variables have important contributions to the final modeling results. It is observed that the overall accuracy generally increases as the number of sensors increases. This paper also examines the possibility of building a global occupancy model, and explores the reasons for low performance of global occupancy estimation. Lastly, the occupancy model is used to estimate and visualize the accumulative room and thermal zone usage in an office test-bed building for three months. The results reveal that the effective vacancy accounts for a substantial portion of the operational hours, varying from 19.8% to 29.8% with an average of 23.3%, which bears significant potential for energy savings. Furthermore, the authors simulated HVAC energy consumption of the test-bed building for three months in DesignBuilder and EnergyPlus, and compared energy consumption of occupancy-based demand-response HVAC controls using the authors' occupancy-modeling results to the conventional HVAC controls currently implemen-ted in the test-bed building. The results demonstrate that 20% of gas and 18% of electricity could be effectively saved if occupancy-based demand-response HVAC control is implemented.
Occupant presence and behaviour in buildings has been shown to have large impact on heating, cooling and ventilation demand, energy consumption of lighting and appliances, and building controls. Energy-unaware behaviour can add one-third to a building's designed energy performance. Consequently, user activity and behaviour is considered as a key element and has long been used for control of various devices such as artificial light, heating, ventilation, and air conditioning. However, how are user activity and behaviour taken into account? What are the most valuable activities or behaviours and what is their impact on energy saving potential? In order to answer these questions, we provide a novel survey of prominent international intelligent buildings research efforts with the theme of energy saving and user activity recognition. We devise new metrics to compare the existing studies. Through the survey, we determine the most valuable activities and behaviours and their impact on energy saving potential for each of the three main subsystems, i.e., HVAC, light, and plug loads. The most promising and appropriate activity recognition technologies and approaches are discussed thus allowing us to conclude with principles and perspectives for energy intelligent buildings based on user activity.
Aim (1) To increase awareness of the challenges induced by imperfect detection, which is a fundamental issue in species distribution modelling; (2) to emphasize the value of replicate observations for species distribution modelling; and (3) to show how ‘cheap’ checklist data in faunal/floral databases may be used for the rigorous modelling of distributions by site-occupancy models.
Location Switzerland.
Methods We used checklist data collected by volunteers during 1999 and 2000 to analyse the distribution of the blue hawker, Aeshna cyanea (Odonata, Aeshnidae), a common dragonfly in Switzerland. We used data from repeated visits to 1-ha pixels to derive ‘detection histories’ and apply site-occupancy models to estimate the ‘true’ species distribution, i.e. corrected for imperfect detection. We modelled blue hawker distribution as a function of elevation and year and its detection probability of elevation, year and season.
Results The best model contained cubic polynomial elevation effects for distribution and quadratic effects of elevation and season for detectability. We compared the site-occupancy model with a conventional distribution model based on a generalized linear model, which assumes perfect detectability (p = 1). The conventional distribution map looked very different from the distribution map obtained using site-occupancy models that accounted for the imperfect detection. The conventional model underestimated the species distribution by 60%, and the slope parameters of the occurrence–elevation relationship were also underestimated when assuming p = 1. Elevation was not only an important predictor of blue hawker occurrence, but also of the detection probability, with a bell-shaped relationship. Furthermore, detectability increased over the season. The average detection probability was estimated at only 0.19 per survey.
Main conclusions Conventional species distribution models do not model species distributions per se but rather the apparent distribution, i.e. an unknown proportion of species distributions. That unknown proportion is equivalent to detectability. Imperfect detection in conventional species distribution models yields underestimates of the extent of distributions and covariate effects that are biased towards zero. In addition, patterns in detectability will erroneously be ascribed to species distributions. In contrast, site-occupancy models applied to replicated detection/non-detection data offer a powerful framework for making inferences about species distributions corrected for imperfect detection. The use of ‘cheap’ checklist data greatly enhances the scope of applications of this useful class of models.
The rapidly growing world energy use has already raised concerns over supply difficulties, exhaustion of energy resources and heavy environmental impacts (ozone layer depletion, global warming, climate change, etc.). The global contribution from buildings towards energy consumption, both residential and commercial, has steadily increased reaching figures between 20% and 40% in developed countries, and has exceeded the other major sectors: industrial and transportation. Growth in population, increasing demand for building services and comfort levels, together with the rise in time spent inside buildings, assure the upward trend in energy demand will continue in the future. For this reason, energy efficiency in buildings is today a prime objective for energy policy at regional, national and international levels. Among building services, the growth in HVAC systems energy use is particularly significant (50% of building consumption and 20% of total consumption in the USA). This paper analyses available information concerning energy consumption in buildings, and particularly related to HVAC systems. Many questions arise: Is the necessary information available? Which are the main building types? What end uses should be considered in the breakdown? Comparisons between different countries are presented specially for commercial buildings. The case of offices is analysed in deeper detail.
Currently it is difficult to determine when and where people occupy a commercial building. Part of the difficulty arises from shortcomings in available sensor technology, but an even greater deficiency is the lack of analysis methods appropriate to the determination of occupancy. This paper describes a pilot study describing new sensing and data analysis techniques, applied to the determination of space occupancy. The central premise of the paper is that improved building operation with respect to energy management, security, and indoor environmental quality will be possible with better detection of building occupancy resolved in space and time. We developed and deployed a network of passive infrared occupancy sensors in two private offices, and applied analysis tools based on Bayesian probability theory to determine occupancy. Specifically, a class of graphical probability models, called belief networks, was applied to the occupancy data generated by the sensor network. The inference of primary importance is a probability distribution over the number of occupants and their locations in a building, given past and present sensor measurements. Inferences were computed for occupancy and its temporal persistence in individual offices as well as the persistence of sensor status. The raw sensor data were also used to calibrate the sensor belief network, including the occupancy transition matrix used in the Markov model, sensor sensitivity, and sensor failure models. This study shows that the belief network framework can be applied to the analysis of data streams from sensor networks, offering significant benefits to building operation compared to current practice.
Liang et al. [A fast and accurate online sequential learning algorithm for feedforward networks, IEEE Transactions on Neural Networks 17 (6) (2006), 1411–1423] has proposed an online sequential learning algorithm called online sequential extreme learning machine (OS-ELM), which can learn the data one-by-one or chunk-by-chunk with fixed or varying chunk size. It has been shown [Liang et al., A fast and accurate online sequential learning algorithm for feedforward networks, IEEE Transactions on Neural Networks 17 (6) (2006) 1411–1423] that OS-ELM runs much faster and provides better generalization performance than other popular sequential learning algorithms. However, we find that the stability of OS-ELM can be further improved. In this paper, we propose an ensemble of online sequential extreme learning machine (EOS-ELM) based on OS-ELM. The results show that EOS-ELM is more stable and accurate than the original OS-ELM.
Energy use in the home is a major source of carbon emissions and is highly dependent on the activities of the residents. More specifically, the timing of energy use, particularly electricity, is highly dependent on the timing of the occupants’ activities. Thus, in order to model domestic demand profiles with high temporal resolution, for example, in the context of designing and assessing demand side management systems (including the time-shifting of demand), it is of great benefit to take account of residents’ behaviour in terms of when they are likely to be using household appliances, lighting and heating. This paper presents a thorough and detailed method for generating realistic occupancy data for UK households, based upon surveyed time-use data describing what people do and when. The approach presented generates statistical occupancy time-series data at a ten-minute resolution and takes account of differences between weekdays and weekends. The model also indicates the number of occupants that are active within a house at a given time, which is important for example in order to model the sharing of energy use (shared use of appliances, etc.) The data from the model can be used as input to any domestic energy model that uses occupancy time-series as a base variable, or any other application that requires detailed occupancy data. The model has been implemented in Excel and is available for free download.
This paper describes a large-scale wireless and wired environmental sensor network test-bed and its application to occupancy detection in an open-plan office building. Detection of occupant presence has been used extensively in built environments for applications such as demand-controlled ventilation and security; however, the ability to discern the actual number of people in a room is beyond the scope of current sensing techniques. To address this problem, a complex sensor network is deployed in the Robert L. Preger Intelligent Workplace comprising a wireless ambient-sensing system, a wired carbon dioxide sensing system, and a wired indoor air quality sensing system. A wired camera network is implemented as well for establishing true occupancy levels to be used as ground truth information for deriving algorithmic relationships with the environment conditions. To our knowledge, this extensive and diverse ambient-sensing infrastructure of the ITEST setup as well as the continuous data-collection capability is unprecedented. Final results indicate that there are significant correlations between measured environmental conditions and occupancy status. An average of 73% accuracy on the occupancy number detection was achieved by Hidden Markov Models during testing periods. This paper serves as an exploration to the research of ITEST for occupancy detection in offices. In addition, its utility extends to a wide variety of other building technology research areas such as human-centered environmental control, security, energy efficient and sustainable green buildings.
In this paper, we develop an online sequential learning algorithm for single hidden layer feedforward networks (SLFNs) with additive or radial basis function (RBF) hidden nodes in a unified framework. The algorithm is referred to as online sequential extreme learning machine (OS-ELM) and can learn data one-by-one or chunk-by-chunk (a block of data) with fixed or varying chunk size. The activation functions for additive nodes in OS-ELM can be any bounded nonconstant piecewise continuous functions and the activation functions for RBF nodes can be any integrable piecewise continuous functions. In OS-ELM, the parameters of hidden nodes (the input weights and biases of additive nodes or the centers and impact factors of RBF nodes) are randomly selected and the output weights are analytically determined based on the sequentially arriving data. The algorithm uses the ideas of ELM of Huang et al. developed for batch learning which has been shown to be extremely fast with generalization performance better than other batch training methods. Apart from selecting the number of hidden nodes, no other control parameters have to be manually chosen. Detailed performance comparison of OS-ELM is done with other popular sequential learning algorithms on benchmark problems drawn from the regression, classification and time series prediction areas. The results show that the OS-ELM is faster than the other sequential algorithms and produces better generalization performance.
Energy efficiency data set
- T Athanasios
- X Angeliki
Athanasios T, Angeliki X (2012) Energy efficiency data set. http://
archive.ics.uci.edu/ml/datasets/Energy?efficiency?ref=data
news.io
Occupancy detection data set
- L Candanedo
Candanedo L (2016) Occupancy detection data set. https://archive.ics.
uci.edu/ml/datasets/Occupancy?Detection?
An experimental investigation of occupancy-based energy-efficient control of commercial building indoor climate
- C M Bishop
- S Goyal
- R Subramany
- Y Lin
- T Middelkoop
- L Arpan
- L Carloni
- P Barooah
Bishop CM (2006) Pattern recognition and machine learning.
Springer, New York
Brooks J, Goyal S, Subramany R, Lin Y, Middelkoop T, Arpan L,
Carloni L, Barooah P (2014) An experimental investigation of
occupancy-based energy-efficient control of commercial building indoor climate. In: 53rd IEEE conference on decision and
control. IEEE, pp 5680-5685