Demand side management programs in smart grid through cloud computing

Deep learning based real time Demand Side Management controller for smart building integrated with renewable energy and Energy Storage System

Over the last few years, smart (power) grids course has become a part of the electrical and computer engineering curriculum. Due to the multidisciplinary nature of the smart grids, teaching methods and materials vary significantly across universities. In this paper, we present our experience in teaching smart power grids course at the division sustainable development at Hamad Bin Khalifa University. The class was offered during Fall 2016 and 2017 semesters and the class roster was composed of graduate students with diverse engineering backgrounds such as electrical, mechanical, chemical, petroleum, and materials science. Special attention has been given to smart grid applications that require end-user involvement such as distributed renewable integration, electric vehicles, and demand-side management programs. This paper further presents sample assignments and projects.

For the past three years (1989, 1990, and 1991), all US electric utilities that sell more than 120 GWh/year have been required to report to the Energy Information Administration data on their demand-side management (DSM) programs. These data provide a rich and uniquely comprehensive picture of electric-utility DSM programs in the United States. Altogether, 890 utilities (of about 3250 in the United States) ran DSM programs in 1991; of these, 439 sold more than 120 GWh and reported details on their DSM programs. These 439 utilities represent more than 80% of total US electricity sales and revenues. Altogether, these utilities spent almost $1.8 billion on DSM programs in 1991, equal to 1.0% of total utility revenues that year. In return for these (and prior-year) expenditures, utility DSM programs cut potential peak demand by 26,700 MW (4.8% of the national total) and cut annual electricity use by 23,300 GWh (0.9% of the national total). These 1991 numbers represent substantial increases over the 1989 and 1990 numbers on utility DSM programs. Specifically, utility DSM expenditures doubled, energy savings increased by almost 50%, and demand reductions increased by one-third between 1989 and 1991. Utilities differed enormously in their DSM-program expenditures and effects. Almostmore » 12% of the reporting utilities spent more than 2% of total revenues on DSM programs in 1991, while almost 60% spent less than 0.5% of revenues on DSM. Utility estimates of future DSM-program expenditures and benefits show continuing growth. By the year 2001, US utilities expect to spend 1.2% of revenues on DSM and to cut demand by 8.8% and annual sales by 2.7%. Here, too, expectations vary by region. Utilities in the West and Northwest plan to spend more than 2% of revenues on DSM that year, while utilities in the Mid-Atlantic, Midwest, Southwest, Central, and North Central regions plan to spend less than 1% of revenues on DSM.« less

Due to environmental concerns, economic issues, and emerging new loads, such as electrical vehicles (EVs), the importance of demand side management (DSM) programs has increased in recent years. DSM programs using a dynamic real-time pricing (RTP) method can help to adaptively control the electricity consumption. However, the existing RTP methods, particularly when they consider the EVs and the power system constraints, have many limitations, such as computational complexity and the need for centralized control. Therefore, a new transactive DSM program is proposed in this paper using an imperfect competition model with high EV penetration levels. In particular, a heuristic two-stage iterative method, considering the influence of decisions made independently by customers to minimize their own costs, is developed to find the market equilibrium quickly in a distributed manner. Simulations in the IEEE 37-bus system with 1141 customers and 670 EVs are performed to demonstrate the effectiveness of the proposed method. The results show that the proposed method can better manage the EVs and elastic appliances than the existing methods in terms of power constraints and cost. Also, the proposed method can solve the optimization problem quick enough to run in real-time.

PurposeThe purpose of this paper is to examine potential demand side management (DSM) programs in terms of their impacts to the overall economy in Thailand.Design/methodology/approachA multi‐sector computable general equilibrium (CGE) model of Thailand has been developed to accomplish the objectives of this study. The potential DSM program considered refers to replacement of less efficient electrical appliances with their efficient counterparts in the household sector in Thailand.FindingsThe study finds that the economy‐wide impacts of the DSM program (e.g., economic welfare, GDP, international trade) depend on three key factors: the project economics of the DSM option or the ratio of unit cost of electricity savings to price of electricity (CPR); the implementation strategy of the DSM option; and scale or size of the DSM option. This paper shows that the welfare impacts of the DSM programs would improve along with the project economics of the DSM programs. If the DSM program is implemented under the CDM, the welfare impacts would increase along with the price for certified emission reductions units. On the other hand, the welfare impacts would increase up to the optimal size or scale of the program, but would start to deteriorate if the size is increased further.Research limitations/implicationsThe welfare function considered in this paper does not account for benefits of local air pollution reductions. The study provides crucial insights on designing DSM projects in Thailand to ensure that DSM programs are beneficial for the economy as a whole.Originality/valueAnalyses of DSM options under the CDM using CGE models are not available in the literature. This is the first paper in this area.

There are hundreds of utility demand-side management (DSM) programs that have operated in the United States since the early 1980s. Conservation and load management programs have enrolled hundreds of thousands of residential utility customers and thousands of commercial/industrial customers. According to the Energy Information Administration, electric utilities in the United States spent $2.8 billion on DSM programs in 1993. Many issues have arisen as a result of the implementation of these programs, and this paper examines four key factors: supplier and customer perspectives; DSM implementation problems; the role of kW demand savings; and engineering estimates, political estimates, and reality. Each of these factors can have a critical role in the success or failure of DSM programs. Due to communication barriers between departments at some utilities, many of these issues can go unresolved until third parties step into the picture and point out discrepancies. As daunting as some of these challenges appear, there are practical solutions that can make DSM programs more cost-effective for the utility and more attractive to customers. These solutions should help progressive and proactive utility companies in the future deregulated market.

Over the past five years, the Energy Information Administration (EIA) has been collecting data annually from US electric utilities on their demand-side management (DSM) programs, both current and projected. The latest data cover activities for 1993 and projections for 1994 and 1998. In 1993, 991 utilities operated DSM programs. That year, they spent $2.8 billion, a 13% increase over 1992 expenditures. These and earlier DSM programs saved 44,000 GWh of energy and reduced potential peak demand by 40,000 MW, 30% and 22% increases over the 1992 values, respectively. While some people predict the demise of electric-utility DSM programs, the data do not paint so bleak a picture. In most parts of the country, DSM programs grew in 1993 and utilities (as of Spring 1994) projected continued growth through 1998. Expenditures grew from 1.3% of revenues in 1992 to 1.5% in 1993, and are expected to grow 2.5% per year faster than inflation, which is equivalent to revenue growth. Thus, DSM spending is expected to stay constant at 1.5% of revenues through 1998. Because of the cumulative effect of DSM programs, energy savings are expected to grow from 1.2% of sales in 1992 to 1.6% in 1993 and 3.0% in 1998. Potential-peak reductions are expected to increase from 5.9% of peak demand in 1992 to 6.8% in 1993 and 8.9% in 1998. However, the growth in spending is not as rapid as the 8% annual real growth projected a year earlier. Actual expenditures in 1993 were 6.5% lower than projected early that year. Energy savings, on the other hand, were the same as projected earlier. Potential peak reductions were actually 9% higher than previously projected.

Over the past five years, the Energy Information Administration (EIA) has been collecting data annually from U.S. electric utilities on their demand-side management (DSM) programs, both current and projected. The latest data cover activities for 1993 and projections for 1994 and 1998. In 1993, 991 utilities operated DSM programs. That year, they spent $2.8 billion, a 13% increase over 1992 expenditures. These and earlier DSM programs saved 44,000 GWh of energy and reduced potential peak demand by 40,000 MW, 30% and 22% increases over the 1992 values, respectively. While some people predict the demise of electric-utility DSM programs, the data do not paint so bleak a picture. In most parts of the country, DSM programs grew in 1993 and utilities (as of Spring 1994) projected continued growth through 1998. Expenditures grew from 1.3% of revenues in 1992 to 1.5% in 1993, and are expected to grow 2.5% per year faster than inflation, which is equivalent to revenue growth. Thus, DSM spending is expected to stay constant at 1.5% of revenues through 1998. Because of the cumulative effect of DSM programs, energy savings are expected to grow from 1.2% of sales in 1992 to 1.6% in 1993 and 3.0% in 1998. Potential-peak reductions are expected to increase from 5.9% of peak demand in 1992 to 6.8% in 1993 and 8.9% in 1998. However, the growth in spending is not as rapid as the 8% annual real growth projected a year earlier. Actual expenditures in 1993 were 6.5% lower than projected early that year. Energy savings, on the other hand, were the same as projected earlier. Potential peak reductions were actually 9% higher than previously projected.

Electric vehicles (EVs) are gaining more interest recently due to the various challenges and opportunities they can provide to the utility operator. For electric utilities that have demand side management (DSM) program, EVs can represent a burden or a help depending on the control strategy and the DSM program. In this paper, an automated fuzzy-based controller is proposed to successfully integrate and coordinate the EVs charging in a decentralized and fair way. The controller takes into consideration the owner requirements, the voltage at the point of connection and the pricing signal coming from the utility. The controller is tested under different DSM programs that exist in the literature and a new DSM program that is compatible with the current infrastructure is proposed to accommodate EVs as prosumers. The controller performance is validated through Matlab simulations. The results showed that the controller can successfully coordinate the charging of EVs in a fair manner, avoid under-voltages and shave the system peaks without any rebound effect.

Today's power systems are based on Tesla's design principles developed in the 1880s and have evolved over time to become the present state. Although communication technology is developing very fast, the development of power systems has not been able to keep up with it. Because the structure of the power system used is often far behind and is unable to meet the needs of the 21st century. With the rapid development of today's technology, it has become possible to make the electricity network better by utilizing the computer and network technologies in the electricity networks. Thus, the electricity networks will provide a safe and uninterrupted energy to the consumers by providing bi-directional information and electricity flow. The grids that can do this are called smart grids. One of the most important features of smart grids is; in the event of a possible interruption or failure, continue to improve the self-healing energy flow. The main goal in self-healing is; to be effective against network breakdowns and at the same time to take security against network breakdowns. To be able to achieve this, the smart grid needs to do the following: a) Quick and accurate detection of mains faults. b) Redistribution of network resources to protect the system from harmful effects. c) To ensure continuity of service in any positive or negative situation. d) The service is the most reduced of the self-renewal period. Various solutions have been proposed for the self-healing of the transmission network. This solution is recommended: optimal voltage control with genetic algorithm base, unified power flow controller (UPFC) and islanding process. Several solutions have been proposed for the self-healing of the distribution network. This solution is recommended: a new smart distribution grid based on the propulsion system, ant colony algorithm, a new multi-stakeholder control system (MACS) for intelligent distribution networks, fault location, isolation and service restoration (FLISR). Some methods have been developed to provide transient stability when self-healing is performed in the smart grids . These methods include; staking is the real-time monitoring and load-balancing method of the network using the phaser unit (PMU) following the generator rotor angles. In this study, the self-healing methods mentioned above are explained in detail in smart grids. Today's power systems are based on Tesla's design principles developed in the 1880s and have evolved over time to become the present state. Although communication technology is developing very fast, the development of power systems has not been able to keep up with it. Because the structure of the power system used is often far behind and is unable to meet the needs of the 21st century. With the rapid development of today's technology, it has become possible to make the electricity network better by utilizing the computer and network technologies in the electricity networks. Thus, the electricity networks will provide a safe and uninterrupted energy to the consumers by providing bi-directional information and electricity flow. The grids that can do this are called smart grids. One of the most important features of smart grids is; in the event of a possible interruption or failure, continue to improve the self-healing energy flow. The main goal in self-healing is; to be effective against network breakdowns and at the same time to take security against network breakdowns. To be able to achieve this, the smart grid needs to do the following: a) Quick and accurate detection of mains faults. b) Redistribution of network resources to protect the system from harmful effects. c) To ensure continuity of service in any positive or negative situation. d) The service is the most reduced of the self-renewal period. Various solutions have been proposed for the self-healing of the transmission network. This solution is recommended: optimal voltage control with genetic algorithm base, unified power flow controller (UPFC) and islanding process. Several solutions have been proposed for the self-healing of the distribution network. This solution is recommended: a new smart distribution grid based on the propulsion system, ant colony algorithm, a new multi-stakeholder control system (MACS) for intelligent distribution networks, fault location, isolation and service restoration (FLISR). Some methods have been developed to provide transient stability when self-healing is performed in the smart grids . These methods include; staking is the real-time monitoring and load-balancing method of the network using the phaser unit (PMU) following the generator rotor angles. In this study, the self- healing methods mentioned above are explained in detail in smart grids.

The rapid development of various power electronics applications facilitates the integration of many smart grid applications in recent years. However, integration of intermittent renewable energy sources, highly stochastic electric vehicles (EVs) activities on the grid and time-varying smart loads have increased the level of grid vulnerability to unusual and high complexity and quality-related problems. Among these problems is to accurately estimate the real contribution and consumption of household loads, in the era of smart appliances and interoperability operation, and its relative impact to the grid’s operation. Specifically, household loads represent a significant percentage of electrical energy consumption and, therefore, could offer great prosperity to the rise of the demand-side management (DSM) programs, which subsequently improve the stability of the grid’s operation. As a result, our main focus in this dissertation is to develop DSM strategies based on Artificial Intelligence (AI) techniques to properly model and estimate the amount of support smart homes could offer to the smart grids and microgrid’s operation. Throughout the way to achieve our goals, we develop an energy management framework for smart homes that operate in efficient and reliable microgrids with multiple energy sources and energy storage applications to meet the demands at a stable voltage and frequency limits. Furthermore, we develop a precise short-term load forecasting (STLF), which is a critical tool needed to manage a DSM program for residential loads that have very high uncertainty and volatility in load consumption. We also develop an energy exchange portal with communication sources, demands, and connectivity information between each consumer and the local power utility at the distribution level. Finally, creative AI methodologies have been developed throughout the way to facilitate the integration, control, and management of the DSM programs taking into account the consumers’ own privacy and security. The security of the DSM is provided by preserving the indoor privacy of the smart homes by sharing limited and encoded data among household appliances controllers.

Demand-side management (DSM) programs aiming to both reduce and render household consumption more flexible are becoming increasingly essential due to ongoing energy crises and the growing integration of renewable energy into energy production. The active involvement of households and energy users is crucial to fully unlock the potential of DSM programs. As this paper demonstrates, despite more than thirty years of feminist scholarly work focusing on the home as an important site of the production of gender inequality, few of these insights have been taken into account by DSM designers. Additionally, we note a broader pattern concerning gaps in knowledge regarding the diverse perspectives of energy users and their domestic contexts, all of which create obstacles to successful rollout and scalability. This paper uses the concepts of the social license to automate and intersectionality to analyze the existing literature on DSM programs. We find that three primary barriers in household DSM programs have been addressed: 1) there is an unresolved tension between DSM technology being perceived as a masculine domain and the home as a feminine domain; 2) low-income households face challenges in accessing the technology needed to enable both flexibility and savings; and 3) disparities in opportunities for youth and the elderly to participate in DSM programs are insufficiently considered. Based on these findings we argue that user diversity—not only conceived of as separate identity category variables but also as implicating overlapping and possible mutually reinforcing marginalizations– is needed to form a starting point in DSM program design for fair and scalable solutions.

As competition in the US electricity industry grows, utilities (and others) worry more about the increases in electricity prices that demand-side management (DSM) programs often cause. Therefore, several utilities have reduced the scope of their DSM programs or focused these programs more on customer service and less on improving energy efficiency. This study uses the Oak Ridge Financial Model (ORFIN) to calculate the rate impacts of DSM. These simulations suggest that DSM programs, although they reduce electric bills, often increase electricity prices. However, utilities can run DSM programs that cut prices. Reducing DSM-program costs, focusing programs on those areas where large transmission and distribution investments can be deferred, timing DSM programs to match avoided costs, and shifting more of the utility's fixed costs to the monthly customer charge will cut DSM-induced price increases.

Electric utilities face a variety of uncertainties that complicate their long-term resource planning and acquisition. These uncertainties include future economic and load growths, fuel prices, environmental regulations, economic regulations, performance and construction cost of existing power plants, cost and availability of purchased power, and the costs and performance of new demand and supply resources. As utilities increasingly turn to demand-side management (DSM) programs to provide energy and capacity resources, it becomes more important to analyze the interactions between these programs and the uncertainties facing utilities. This report uses a new planning model (DIAMOND, developed at Oak Ridge National Laboratory) to explore quantitatively the uncertainty implications of supply-only vs DSM + supply resource portfolios. The analysis focuses on risks to society, with only limited attention to the allocation of risks among customers, shareholders, and others. Four sets of uncertainties are considered in these analyses: economic growth, fuel prices, the costs to build new power plants, and the costs to operate DSM programs. These four types of uncertainties serve as proxies for the many others that face utilities, including delays in completing power plants (proxied by cost of completing plants) and the energy and load reductions caused by DSM programs (proxied by cost of DSM programs). The two types of resource portfolios are tested against these four sets of uncertainties for the period 1990 to 2010. Sensitivity, scenario, and worst-case analysis methods are used. Results show that it is feasible to analyze the effects on uncertainty of including DSM programs in a utility's resource mix. In light of these results, utilities, which to date have done very little such analysis, should conduct such studies as part of their integrated-resource planning activities.

Demand side management (DSM) programs are designed to encourage users to shift the use of their non-critical appliances to off-peak hours. Autonomous DSM programs have recently been proposed to achieve this goal by coordinating the users' energy consumption, using smart meters. On the other hand, this objective can be achieved only when the users actively contribute in DSM programs. Devising a fair billing mechanism is important to encourage the users to keep their contribution in the programs to achieve system optimality in the sense of minimum cost of the system. Another important issue in implementing DSM programs is protecting the users' privacy which is short addressed in DSM literature. In this paper, we introduce the concept of fairness, optimality and privacy in DSM systems. Next, we introduce a class of optimal billing mechanisms. We propose a subclass of optimal billing mechanisms which is fair in terms of distributing the energy cost across the users based on their contribution in minimizing the total cost of system. We show that fairness axioms which have been previously introduced in resource allocation algorithms are achievable in the proposed billing subclass. Next, we apply the secure sum algorithm to protect the users' privacy in implementing this billing mechanism.