Design of machine learning models with domain experts for automated sensor selection for energy fault detection
طراحی مدلهای یادگیری ماشینی با کارشناسان دامنه برای انتخاب سنسور خودکار برای تشخیص خطای انرژی-2019
Data-driven techniques that extract insights from sensor data reduce the cost of improving system energy performance through fault detection and system health monitoring. To lower cost barriers to widespread deployment, a methodology is proposed that takes advantage of existing sensor data, encodes expert knowledge about the application system to create ‘virtual sensors’, and applies statistical and mathematical methods to reduce the time required for manual configurations. The approach combines sensor data points with encoded expert knowledge that is generic to the application system but independent of a particular deployment, thereby reducing the need to tailor to individual deployments. This paper not only presents a method that detects faults from measured energy data, but also (1) describes an engagement method with experts in the energy system domain to identify data, (2) integrates domain knowledge with the data, (3) automatically selects from among the large pool of potential input data, and (4) uses machine learning to automatically build a data-driven fault detection model. Demonstration on a commercial building chiller plant shows that only a small number of virtual sensors is necessary for fault detection with high accuracy rates. This corresponds to the use of only five out of 52 original sensor data points features. With as few as four features, classification F1 scores exceed 90% on the training set and 80% on the testing set. The results are implementable and realizable using off-the-shelf tools. The goal is to design with domain experts an energy monitoring system that can be configured once and then widely deployed with little additional cost or effort
Keywords: Machine learning | Domain knowledge | Time series | Fault detection | Anomaly detection | Energy savings | Energy efficiency
Efficient jobs scheduling approach for big data applications
رویکرد برنامه ریزی شغلی کارآمد برای داده های بزرگ-2018
The MapReduce framework has become a leading scheme for processing large-scale data applications in recent years. However, big data applications executed on computer clusters require a large amount of energy, which costs a considerable fraction of the data center’s overall costs. Therefore, for a data center, how to reduce the energy consumption becomes a critical issue. Although Hadoop YARN adopts fine-grained resource management schemes for job scheduling, it doesn’t consider the energy saving problem. In this paper, an Energy-aware Fair Scheduling framework based on YARN (denoted as EFS) is proposed, which can effectively reduce energy consumption while meet the required Service Level Agreements (SLAs). EFS not only can schedule jobs to en ergy-efficiency nodes, but also can power on or off the nodes. To do so, the energy-aware dynamic capacity management with deadline-driven policy is used to allocate the resources for MapReduce tasks in terms of the average execution time of containers and users request resources. And then, Energy-aware fair based scheduling problem is modeled as multi-dimensional knapsack problem (MKP) and the energy-aware greedy algorithm (EAGA) is proposed to realize tasks fine-grained placement on energy-efficient nodes. Finally, the nodes which have been kept in idle state for the threshold duration are turned off to reduce energy costs. We perform ex tensive experiments on the Hadoop YARN clusters to compare the energy consumption and executing time of EFS with some state-of-the-art policies. The experimental results show that EFS can not only keep the proper number of nodes in on states to meet the computing requirements but also achieve the goal of energy savings.
Keywords: Big data ، Dynamic scheduling ، Energy efficiency ، MapReduce ، Resource allocation
Selecting new product designs and processing technologies under uncertainty: Two-stage stochastic model and application to a food supply chain
انتخاب طراحی های جدید محصول و فناوری های فرآوری تحت عدم قطعیت: مدل تصادفی دو مرحله ای و کاربرد برای یک زنجیره تامین غذا-2018
New product introduction frequently requires new processing technologies, and the development of new processing technologies also allows for the introduction of new products. An assessment of these new products and technologies must account for changes in the whole supply chain. This paper presents a two-stage stochastic mixed integer linear programming model that integrates the selection of new product designs and processing technologies in a supply chain context. Special attention is given to the demand uncertainties with regard to product specifications and volumes. The first stage of the model selects the processing technologies that determine the set of feasible product designs, leaving the detailed product designs and the production volumes as recourse actions to the second stage. We apply the developed approach to product designs and processing technologies in the dairy sector. Here, the substitution of milk powders through milk concentrates is currently being considered, which may lead to extensive energy savings in production. In an interdisciplinary effort, we first derive the design space encompassing the feasible dairy technologies and product designs for concentrates. Through numerical investigation we then show that flexible technologies are selected that can be used to produce different product designs. We also show that the selection of technologies is highly dependent on the uncertain demand characteristics of the new concentrate products.
keywords: Product design |Technology selection |Supply chain |Dairy industry |Two-stage stochastic programming
Development of building energy saving advisory: A data mining approach
توسعه مشاوره صرفه جویی در انرژی ساختمان: رویکرد داده کاوی-2018
Occupants’ behavior and their interaction with home appliances are crucial for assessing building energy consumption. This study proposes a new methodology for monitoring the energy consumed in building end-use loads to build an advisory system. The built system alerts occupants to take certain measures (prioritized recommendations) to reduce energy consumption of end-use loads. The quantification of po tential savings is also provided upon following said measures. The proposed methodology is also capable of evaluating the energy savings performed by the occupants. The system works based on the analysis of historical data generated by occupants using data mining techniques to output highly feasible recom mendations. For demonstration purposes, the methodology was tested on the real dataset of a building in Japan. The dataset includes detailed energy consumption of end-use loads, categorized as hot water supply, lighting, kitchen, refrigerator, entertainment & information, housework & sanitary, and others. Re sults suggest that the developed models are accurate, and that it is possible to save up to 21% of total energy consumption by only changing occupants’ energy use habits.
Keywords: Occupant behavior ، Data mining ، Building energy
The kingdom of the bicycle: what Wuhan can learn from Amsterdam
پادشاهی دوچرخه: آنچه ووهان می تواند از آمستردام یاد بگیرد-2017
China used to be called “the Kingdom of the Bicycle,” but this is no longer the case. Bicycle use in China has been marginalized over the past 30 years. In contrast, the Netherlands has seen bicycle use grow since the 1970s. This paper—through a comparative analysis of data from Wuhan and Amsterdam—explores the reasons why the two countries have gone in different directions. Although these cities have different socio-demographics they experienced similar issues. This paper suggests lessons that Wuhan can learn from Amsterdam. However, these are one-way as it is considered that Amsterdam has little to learn from the decline of bicycle use in Wuhan.
Keywords: Transport | energy savings | bicycle | Wuhan | China | Amsterdam | the Netherlands
Environmental benefits from ridesharing: A case of Beijing
مزایای زیست محیطی از ridesharing: یک مورد از پکن-2017
Emerging ridesharing travel could be an effective way in China to reduce travel demand by cars, which can further seek to shift personal transportation choices from an owned asset to a service used on demand and lessen the traffic jam and emissions. Drawing on the raw observed ridesharing trip data pro vided by DiDi Chuxing company, this study evaluated the direct environmental benefits of ridesharing resulted from the travel mode shift and the indirect environmental benefits resulted from the attitude change towards car purchase behavior. The megacity Beijing is taken as the empirical context given its more serious situation of traffic congestion and difficulties for car purchase. Estimation results show that direct annual energy savings made by ridesharing are approximately 26.6 thousand tce, and annual emis sion reductions of CO2 and NOx are approximately 46.2 thousand tons and 253.7 tons, respectively. Besides, using ridesharing service will lead to substantial energy savings and emission reductions from the long-term perspective attributing to the weakening willingness on purchasing new cars. Promoting EVs among ridesharing vehicles and switching to clean electricity generation as well as improving vehicle efficiency can further enhance the environmental benefits of ridesharing, with maximum effects amount ing to 67% of energy savings and 57% of CO2 emission reductions compared to 2016 level of the fuel related energy consumption and emissions made by ridesharing.
Keywords: Ridesharing | Direct impacts | Indirect impacts | Energy saving | Emission reduction | Beijing
Achieving energy savings by intelligent transportation systems investments in the context of smart cities
دستیابی به صرفه جویی انرژی توسط سرمایه گذاری سیستم های حمل و نقل هوشمند در زمینه شهرهای هوشمند-2017
Investments in intelligent transportation systems (ITS) are beginning to take place in the context of smart city initiatives in many cities. Energy efficiency and emissions reduction are becoming essential rationales for such investments. It is important, therefore, to under stand under what conditions investments in ITS in the context of smart cities produce energy savings. We reviewed existing literature, conducted case studies and interviews, and found that the smart cities context has transformed traditional ITS into ‘‘smart mobil ity” with three major characteristics: people-centric, data-driven, and powered by bottom up innovations. We argue that there are four main steps for smart mobility solutions to achieve energy savings and that several institutional, technical, and physical conditions are required at each step. Energy savings are achieved when users change their behavior and result in less travel, modal shift, and reduction of per-km energy consumption in the short term. Smart mobility solutions also enable other energy saving policies or initiatives, which would otherwise not be feasible. In the long term, users’ lifestyles could change and lead to fur ther energy savings. For cities in developing countries with lower motorization, less-developed infrastruc ture, less financial resources, and less institutional and technical capacity, our recommen dations to achieve benefits from smart mobility investments are: (1) involve all public and private players in a collaborative and transparent setting; (2) develop the technical capac ity to procure and monitor information services; and (3) focus on basic infrastructure, including a coherent road network and basic traffic management measures.
Keywords: Energy savings | ITS | Smart cities | Smart mobility
Categories and functionality of smart home technology for energy management
دسته بندی ها و قابلیت های تکنولوژی خانه هوشمند برای مدیریت انرژی-2017
Technologies providing opportunities for home energy management have been on the rise in recent years, however, its not clear how well the technology - as its currently being developed - will be able to deliver energy saving or demand shifting benefits. The current study undertakes an analysis of 308 home energy management (HEM) products to identify key differences in terms of functionality and quality. Findings identified opportunities for energy savings (both behavioural and operational) as well as load shifting across most product categories, however, in many instances other potential benefits related to convenience, comfort, or security may limit the realisation of savings. This is due to lack of information related to energy being collected and presented to users, as well as lack of understanding of how users may interact with the additional information and control provided. While the current study goes some way to identify the technical capabilities and potential for HEM products to deliver savings, it is rec ommended that further work expand on this to identify how users interact with these technologies in their home, in both a standalone and fully integrated smart home environment to deliver benefits to both homes and the grid.
Keywords: Home energy management | Energy efficiency | Smart home | Home automation | Internet of things
Energy-aware and quality-driven sensor management for green mobile crowd sensing
مدیریت سنسور انرژی آگاه و کیفیت محور برای سنجش سبز جمعیت سیار-2016
Mobile Crowd Sensing (MCS) is a novel class of Internet of Things applications which exploits the inherent mobility of wearable sensors and mobile devices to observe phenomena of common interest, typically over large geographical areas (e.g. trafﬁc conditions, air pollution, noise in urban areas). Since MCS applications generate large amounts of sensed data which is collected and preprocessed by devices with limited energy supply, challenges arise with respect to sensor management to ensure an energy- aware and quality-driven data acquisition process. In this paper we present a framework for Green Mobile Crowd Sensing (G-MCS) which utilizes a quality-driven sensor management function to continuously select the k-best sensors for a predeﬁned sensing task. Our G-MCS solution utilizes a cloud-based architecture centered around a publish/subscribe communication model to enable the interaction of mobile devices with the cloud for energy-aware MCS. In particular, it obviates redundant sensor activity while satisfying sensing coverage requirements and sensing quality, and consequently reduces the overall energy consumption of an MCS application. We present a model for G-MCS and evaluate its energy savings for different application requirements and geographical sensor distribution scenarios. Furthermore, our model evaluation on a real data set shows that in certain identiﬁed cases, signiﬁcant energy consumption reductions can be achieved by utilizing the proposed framework, which opens the door for green solutions within the area of MCS applications.& 2015 Elsevier Ltd. All rights reserved.
Keywords: Mobile crowd sensing | Internet of Things | Energy-aware system
Scaling up the energy service company business: market status and company feedback in the Russian Federation
مقیاس گذاری تجارت خدمات انرژی شرکت: وضعیت بازار و فیدبک شرکت در فدراسیون روسیه-2016
Many energy efﬁciency professionals have proposed using Energy Performance Contracts (EPCs) as a mechanism to improve public sector energy efﬁciency in countries with restrictive government budgets. However, in practice, most middle-income countries have used this mechanism only in a limited way. Russia offers an interesting case study because of its huge energy savings opportunities, increasing en- ergy prices, robust political backing for public sector energy efﬁciency, and evolving legislation that supports EPCs. In 2009, the Russian Federation initiated reductions in the countrys energy intensity, including of the large public sector, which accounts for 9 percent of Russias total energy consumption. To achieve energy efﬁciency goals in the public sector, Russia experimented with its public procurement rules, adjusting them to encourage EPCs. We reviewed the Russian governments ofﬁcial public database, conducted structured interviews with Energy Service Companies (ESCOs) in Russia and supplemented them with online research. Even though this process might not have captured all of the EPCs signed in Russia as of mid-2013, we estimate that nearly 50 ESCOs signed about 150 contracts in public facilities. Most ESCO contracts in Russia are for 5 years, and they generally are small (under $100,000). ESCOs in Russia face a challenging environment, which leads to smaller projects. ESCOs also are concerned about costly and risky tender procedures, uncertainty regarding repayment from public facilities, the inability to expand projects, and ﬁnancing. We discuss these challenges and propose potential solutions at policy and company levels. The ESCOs feedback regarding Russias experimental model can inform the countrys program for public sector energy efﬁciency and offer lessons for other countries attempting to develop the EPC mechanism.© 2015 Elsevier Ltd. All rights reserved.
Keywords: Energy performance contract | Energy service company | Energy saving | Public sector energy efficiency | Procurement | Russian Federation