Fig 20 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
We pose and study a scheduling problem for an electric load to develop an Internet of Things (IoT) control system for power appliances, which takes advantage of real-time dynamic energy pricing. Using historical pricing data from a large U.S. power supplier, we study and compare several dynamic scheduling policies, which can be implemented in a sma...
Similar publications
A blockchain is an advanced technology that can power over a decentralized network. The authors bring it up to design the autonomous transaction system for e-commerce applications; because of the dramatic increase in IoT devices, communication between physical things is enabled. This brings more efficiency and accuracy, which benefits the outsiders...
Citations
... The use of IoT (Internet of Things) and machine learning techniques in healthcare has revolutionized various aspects of medical diagnosis, including cancer detection and diagnosis. IoT in Healthcare: The Internet of Things (IoT) has emerged as a disruptive technology in healthcare, enabling the collection and analysis of real-time data from various sources, including wearable devices and medical sensors [1], [2]. IoT-based systems have been successfully employed in remote patient monitoring, disease management, and early detection of health abnormalities. ...
... These algorithms are trained on large datasets and can learn patterns and relationships that may be difficult for humans to discern. In the context of cancer diagnosis, machine learning models can analyze complex protein datasets, genetic data, and medical images to predict the presence of cancer, determine its stage, and assist in treatment planning [1], [3], [4]. ...
... The peak load interval, the maximum cost of power at peak time and normal time, and the power availability are taken from the smart grid. This information is communicated to the smart home to take decisions about power consumption [26]. The expected bill information is also evaluated by using this information. ...
... The field of home automation has witnessed significant advancements in recent years, with a growing emphasis on energy efficiency and sustainability. The field of home automation has witnessed significant advancements in recent years, with a growing emphasis on energy efficiency and sustainability [1]- [3]. Numerous studies have focused on the integration of IoT technologies in home automation systems. ...
This research article presents the development and analysis of an IoT-enabled smart home automation system aimed at enhancing energy efficiency. The proposed system integrates various sensors, a Raspberry Pi as a central controller, and utilizes the Message Queuing Telemetry Transport (MQTT) protocol for efficient data transmission. The system architecture comprises multiple layers, including the device layer, broker layer, service layer, application layer, and cloud layer, each serving specific functions such as sensor data collection, communication, data management, and user interface. Real-time monitoring and control are facilitated through the application layer, which enables users to interact with the system using Node-Red Authentication. The system demonstrates fast response times of less than 15 ms, ensuring prompt and accurate actions in controlling home appliances. Moreover, the system effectively reduces energy consumption by optimizing the operation of appliances based on sensor inputs and user preferences. The proposed system showcases significant energy savings compared to manual systems, with power consumption reduced by up to 40%. The results indicate that the system achieves efficient energy management and contributes to cost savings for homeowners. Additionally, the system's user-friendly interface and real-time monitoring capabilities enhance the overall user experience. The research findings emphasize the potential of the proposed system in promoting energy-efficient practices in smart homes. Further advancements and research in this area can lead to more intelligent and sustainable home automation systems, ultimately benefiting both homeowners and the environment.
... Kaur et al. [13] express the task scheduling as supervised machine learning (ML) classifier issues that activate the load in either chosen time bin. The feature utilized in the ML issue is spot and day forward price, delayed label of the preceding day, and hourly market. ...
... The combination of Energy Cycle and Energy Management along with profile constraints has enabled to develop new algorithms from the existing data. The Energy Scheduling using internet of things (IoT) and machine learning (ML) incorporating time-based control for cost reduction is proposed in [8,9]. Various control schemes developed from existing data results in an intelligent enhancement of the system. ...
The advantages of real-time dataset with artificial intelligence-based machine learning algorithms can be utilized for improving energy management in Electric Vehicles (E-Vehicle). Here the machine learning algorithms such as classification, regression, reinforcement, and clustering are incorporated to identify the suitable profile pattern. Electric vehicle modelling can be developed from profile, outline, and formulation for energy efficiency can be done from vehicle state, specifications from acceleration and deceleration with necessary constraints. Each vehicle energy management stages are realized as kinetic energy, mechanical energy, and electrical energy for minimizing the losses within a boundary except proving the constraints. With a problem formulated in each stage for classifying the input profile, it is found that regression analysis matches the energy optimization, reinforcement technique fixes the boundary at every profile input and clustering of data ensures formation of data set. The proposed scheme is validated with MATLAB/Simulink environment and the results shows that the new methods improved the energy efficiency in every stage at the level of 10 to 30%. Also, among the energy efficient techniques, classification and regression schemes adds an error loss of 0.2 to 0.3%, the reinforcement method includes the data handling error of 0.3 to 0.4% and finally, the clustering method increases the error loss in the level of 0.4 to 0.5%.
... The combination of Energy Cycle and Energy Management along with profile constraints has enabled to develop new algorithms from the existing data. The Energy Scheduling using Internet of Things (IoT) and Machine Learning (ML) incorporating time-based control for cost reduction is proposed in [8,9]. Various control schemes developed from existing data results in an intelligent enhancement of the system. ...
The advantages of real-time dataset with artificial intelligence-based machine learning algorithms can be utilized for improving energy management in Electric Vehicles (E-Vehicle). Here the machine learning algorithms such as classification, regression, reinforcement, and clustering are incorporated to identify the suitable profile pattern. Electric vehicle modelling can be developed from profile, outline, and formulation for energy efficiency can be done from vehicle state, specifications from acceleration and deceleration with necessary constraints. Each vehicle energy management stages are realized as kinetic energy, mechanical energy, and electrical energy for minimizing the losses within a boundary except proving the constraints. With a problem formulated in each stage for classifying the input profile, it is found that regression analysis matches the energy optimization, reinforcement technique fixes the boundary at every profile input and clustering of data ensures formation of data set. The proposed scheme is validated with MATLAB/Simulink environment and the results shows that the new methods improved the energy efficiency in every stage at the level of 10 to 30%. Also, among the energy efficient techniques, classification and regression schemes adds an error loss of 0.2 to 0.3%, the reinforcement method includes the data handling error of 0.3 to 0.4% and finally, the clustering method increases the error loss in the level of 0.4 to 0.5%.
... In several scheduling applications, if scheduling an item is interpreted as an action, then the constraints on the sequence of actions arise naturally. For example, in real-time scheduling of home appliances, the dryer cannot be scheduled before the washer [3], [4]. In this paper, we are interested in a method for obtaining optimal control policies for such action-constrained scenarios. ...
In this paper, we use concepts from supervisory control theory of discrete event systems to propose a method to learn optimal control policies for a finite-state Markov Decision Process (MDP) in which (only) certain sequences of actions are deemed unsafe (respectively safe). We assume that the set of action sequences that are deemed unsafe and/or safe are given in terms of a finite-state automaton; and propose a supervisor that disables a subset of actions at every state of the MDP so that the constraints on action sequence are satisfied. Then we present a version of the Q-learning algorithm for learning optimal policies in the presence of non-Markovian action-sequence and state constraints, where we use the development of reward machines to handle the state constraints. We illustrate the method using an example that captures the utility of automata-based methods for non-Markovian state and action specifications for reinforcement learning and show the results of simulations in this setting.
... Performance: overall performance improvements is the most common challenge inside this group [20,30,31,33,34,37,54,56,61,62,68,72,77]. Some edge devices, such as single board computers or microcontrollers, are constrained in resources and are not capable of handling large workloads. ...
Serverless computing, or Function as a Service (FaaS), represents a research trend where applications are built and deployed as a group of stateless functions. Although initially proposed for the cloud, serverless computing has also found its place on Internet of Things (IoT) while bringing functions closer to the devices, in order to reduce latency and avoid unnecessary energy and resource consumption. It is interesting that solutions can work in an integrated manner on edge, fog, and cloud layers. Mission-critical functions can be executed on edge and fog in order to benefit from low-latency responses, while heavy functions can be executed on the cloud to process huge amount of data produced by IoT sensors, as long as Internet connection is available. Existing surveys focus on serverless computing for specific layers and do not address a broad, integrated, and systematic vision regarding how IoT benefits from serverless on edge, fog, and cloud. With this in mind, this paper provides a comprehensive Systematic Literature Review that, after the selection process, covers 60 papers on the field of serverless computing for IoT on the three layers. This gives us insights about how functions are offloaded to different devices and how they interact with each other. We bring main components employed to incubate and execute functions, as well as the main challenges and open questions for this subject. Protocols, programming languages, and storage services related to the solutions are also presented. Finally, we show a rich taxonomy summarizing all characteristics in a single figure, along with a discussion about the overall architecture of serverless applications for IoT. We conclude that serverless computing is a promising technology for IoT applications, but several improvements still need to be made to popularize this concept and make it easier to use.
... Donghan et al. [30] show a broad range of applications which use artificial intelligence (AI) methods in energy research. Within energy systems analysis, the first AI applications focus on building analysis and management systems [2,15,31,39,59,65] , local integrated energy systems [5,14,41] , smart charging [20] , demand prediction [4,7,8,40,51] , big data analysis including data mining [42,45] and investment decisions [21] . Algorithms therefore need to be enabled to understand and interpret this large variety of data sources from heterogeneous domains. ...
... time stamp indicates either the start, middle, or ending of the time step). Energy system models use many different kinds of time series from several dis- 31 ciplines, e.g. meteorological data, time series of power generation and consumption or economic time series, like energy prices. ...
Heterogeneous data, different definitions and incompatible models are a huge problem in many domains, with no exception for the field of energy systems analysis. Hence, it is hard to re-use results, compare model results or couple models at all. Ontologies provide a precisely defined vocabulary to build a common and shared conceptualisation of the energy domain. Here, we present the Open Energy Ontology (OEO) developed for the domain of energy systems analysis. Using the OEO provides several benefits for the community. First, it enables consistent annotation of large amounts of data from various research projects. One example is the Open Energy Platform (OEP). Adding such annotations makes data semantically searchable, exchangeable, re-usable and interoperable. Second, computational model coupling becomes much easier. The advantages of using an ontology such as the OEO are demonstrated with three use cases: data representation, data annotation and interface homogenisation. We also describe how the ontology can be used for linked open data (LOD).
