Duck curve leveling in renewable energy integrated grids using internet of relays

https://doi.org/10.1016/j.jclepro.2021.126294Get rights and content

Abstract

Power grids are undergoing deregulation, privatization, and decentralization worldwide. The smart grid allows the integration of clean power renewable energy technologies to the utility grid through mini, micro, and nano grids. Conventional centralized power grids used to rely on fossil fuel-fired power plants which are being replaced worldwide with solar, wind, small hydro, geothermal, biomass, wave, and alternative energy technologies. Installed solar and wind power generation capacity has surpassed 1300 GW in 2020. Solar and wind powers rise during the day and fall during the evening. Consumer demand increases steadily during the evening and peaks after sunset when solar generation becomes zero and wind power falls substantially. Decentralized smart grid faces duck curve limitation on the integration of renewable energy technologies as centralized utilities used to face the peak-hours issue. Smart grid operators have no effective solution of load peaking after sundown except keeping fossil fuel-fired plants on standby to ride through the duck curve. We present a solution to the duck curve problem by integrating smart load shedding devices in micro and nano grids to adjust their demand frugally during peak hours to support the national grid. Information and communication (ICT) technologies that enabled the internet of things (IoT) have been attempted to facilitate smart load shedding in nano grids. ICT enabled voltage transformers (VT) and current transformers (CT) supply signals to heuristic rules based multifunction smart relaying system. Smart meters, instrument transformers, and status monitoring field devices generate a massive amount of data. The research designed an IoT and LabVIEW based software platform to carry out demand-side management. Under frequency (UF) and under-voltage (UV) relaying functions were used to conduct load shedding in nano grids. This nanoscale demand-side management experiment paves the way to ride through utility-scale duck curve setback. This PT/CT data based economic solution performs at nanoscale same as ABB PML 630 load shedding controller. This UF and UV relaying based multifunction relay starts shedding loads when frequency falls below <1% and voltage below 10% of rated values. Our load shedding scheme has successfully functioned on nanoscale 5 kW system. The future study section proposes to integrate big data technologies in-home energy management system (HEMS) through extract, transform and load (ETL) pipeline to import data from PT/CT and circuit breakers (CB) integrated smart meter to transform it according to requirement by loading data into Hadoop “big data” processing platforms for utility-scale load management with help of day-ahead load and weather-dependent renewable power generation forecasting techniques.

Introduction

Hydrocarbons have played a key role in economic growth, the food supply chain, and prosperity worldwide. However, the rampant rise in greenhouse gas (GHG) emissions has forced the world community to undertake a grand energy transition (GET) from fossil fuels to renewable and alternative energy sources. Global carbon dioxide emissions due to the consumption of fossil fuels account for 90% of all emissions. Anthropogenic and natural GHG emissions drive climate change, destroy wildlife habitats, pollute the atmosphere and oceans. Nature set aside excess carbon by converting buried dead organic matter with anaerobic digestion under high temperature and pressure conditions in millions of years. British oil and coal are 150 and 300 million years old. Fossil fuels companies discovered the zillions years old hydrocarbons to produce heat and electric energies. Fossil fuels include coal, gas, oil, and shale. Coal was recognized 3000 years ago, but Thomas Walker started the first commercial mining in 1750 in Kentucky, USA. Global coal demand peaked in 2014 at more than 5600 MTCE in 2014 and nosedived to less than 5000 MTCE in 2020 (California Independent System Operator, 2012). Global 1.14 trillion tons (750 GTOE) coal reserves, at declining rates due to energy transition, may last for the next 150 years. The history of natural gas is 2500 years old, but William Hart dug the first natural gas well in 1821 for street lightning in Fredonia, New York. Natural gas is used for heating, cooling, and power generation. Global natural gas demand was 4000 BCM in 2019 and forecast to be 4080 BCM in 2020 but fell to 3850 BCM in 2020 due to the covid-19 pandemic (Gao et al., 2019). Russia, Iran, Qatar, Turkmenistan, and America hold 127 TCM natural gas reserves. Global natural gas reserves are 6923 TCF which might end by 2080. The shale boom has sustained the fossil fuel industry. Crude was used as caulking material in 4000 BC in Mesopotamia but Erwin Drake discovered the first oil field in 1859 at Oil Creek near Titusville, Pennsylvania. Crude oil demand and production peaked at 100.1 and 95.192 million barrels per day in 2019 (John et al., Austin). Oil demand peaked at $10/barrel due to the underlying energy transition from fossil fuels to renewables. Global crude oil reserves at a rate of 4 billion tonnes a year will deplete by 2051 (Ecotricity UK). Global energy consumption is 14.5 GTOE out of which 11 GTOE comes from fossil fuels. Fossil fuel-driven global warming hypothesis might not be true as global shut down of industry during Covid-19 pandemic in 2020 resulted in 7% decline in global GHG emissions which may increase later (Khahro et al., 2013; Ethan). Permafrost collapse in the North Pole is releasing carbon dioxide, methane, nitrous oxide, and microorganisms causing a risk of climate change and ancient epidemics (Chelsea). Paris Accord is an excellent initiative for a cleaner and greener future. Energy transition to solar and wind power sources is an on-time effort being pursued worldwide. Solar and wind power integrated power grids have started suffering from the duck curve limitation due to the fall of solar and wind powers after sunset.

The duck curve reflects the timing disparity between peak demand and peak yield of solar parks and wind farms. Power control engineers know the electricity demand rises during the day and peaks in the evening. The utility operators ramp up dispatchable power sources as demand peaks and renewable energy sources start ramping down. A similar situation occurs during normal loads when the sun starts eclipsing or the floating clouds shade the solar farms. Peak hours start in the evening and reach peak plateau after the sunset when there is no more solar power and wind power also falls. The silhouette of duck curve formed by dipping generation and steadily rising demand may be transformed from sitting to flying state by under frequency (UFR) and under-voltage (UVR) relaying mediated load shedding. Conventional methods to deal with duck curves include the deployment of dispatchable generation, orienting solar panels to the west to increase solar generation before sunset, energy storage, and energy demand management. Energy methods include pumped storage (Zeng et al, 2019), large battery banks (Omine et al., 2019), electric vehicles (Isomura et al., 2020), hydrogen storage, power-to-X storage (Kulkarni et al., 2018), solar heat, and LNG cold storage techniques. Energy demand management strategies may include the use of energy-efficient lighting, time-of-use (TOU) pricing, real-time pricing, power delivery from west to east for later sundown, and smart grid technologies. A power utility taught the duck how to fly by controlling wastewater, heating, and cooling loads (Mortaji et al., 2017). A typical duck curve for a solar/wind integrated utility is shown in Fig. 1.

Electric grids have now transformed into solar parks and wind farm integrated power grids. UN sustainable development goals (SDG-17) targets and IPCC guidelines pushing utilities for energy transition from fossil fuels to renewable energy resources. Smart grid technologies facilitate distributed energy resources (DER) integration with power grid. A paradigm shift in power utilities has inundated grids with IEDs, IoTs, AI, IT and ICT technologies. Power pundits advise to electrify everything to get rid of GHG emissions of coal and gas power plants. Climate change electric grid digitalization has led power utilities to the duck curve problem. California ISO noted in 2016 that the duck curve is a major limitation in transformation of the electric grid (California Independent System Operator, 2012). Metrological studies show that the sunshine and wind speeds start increasing in morning, reach peak plateau in middle of day and then start decreasing in evenings. Solar electricity nosedives to zero after sunset and wind power reasonably falls as shown in Fig. 2.

Sunshine remains high in equatorial countries and varies in tropics. Northern hemispheres have high sun in summer that decreases in autumn and spring. Sunshine decreases to low levels in winter. Situation remains same in southern hemisphere but timings of seasons reverse. When northern hemisphere faces cold waves there are heat waves in Australia. Unlike sunshine, wind speeds decrease in morning and are still present after sunset. In northern hemisphere the wind speed falls from average 4.5 m/s to 3 m/s in spring, 4.3 m/s to 3.5 m/s in winter, 3.7 m/s to 1.4 m/s in autumn and 3.5 m/s to 2 m/s in summer. However, wind speed variations in coastal regions are lesser than plain regions, nevertheless daily, monthly and seasonal variations may occur (Khahro et al., 2013). Wind may be defined as movement of air caused by irregular heating of the Earth’s surface by the sunshine and the Earth’s rotation. There are five zones of winds worldwide. Average wind speed is usually 6 miles/hour in doldrums, 39–50 miles/hour in tropics and 6–13 miles/hour in polar regions. Wind potential is higher in south pole than north pole. Sea and land breeze wind speeds are 6 miles/hour in within 25–30 and 15–20 km ranges. Average wind speeds vary from 5 to 9 m/s on surface of various seas. In polar regions winds are high and sun is low and in doldrums sun is high and winds are low. However, in tropics both solar and wind power potentials are optimum. Metrologist’s claim the sunlight and wind speeds are rising due to climate change (Ethan; Chelsea; Zeng et al., 2019). Sun and wind generally fall in evening so the duck curve problem needs smart solution. Solar heat and hydrogen storage technologies are under extensive research worldwide.

The duck curve has exposed a dark side of solar and murky side of wind energies requiring intellectual solution. A possible within impossible might be to store high time solar/wind energies to provide ride through during peak hours to flatten the duck curve. Electricity, heat, and hydrogen storage options are expensive requiring lot of capital investment. Smart grid technologies may include demand response (Omine et al., 2019), automatic load shedding (Isomura et al., 2020), IoT and AI based systems (Kulkarni et al., 2018). Smart grid technologies may handle the duck curve issue using internet of things (IoT) (Kulkarni et al., 2018), information & communication technologies (ICT) and artificial intelligence (AI) based devices (Mortaji et al., 2017), (Alarbi et al., 2019). Smart grid technologies consist of voltage and current transformers. Synchrophasors or phase measuring units (PMU) provide real-time voltage and current phasors for monitoring, control and protection (Aminifar et al., 2014). These PMUs collect hundreds times faster data than power line communication based SACADA system (Tsado et al., 2015). A resilient communication for smart sensor networks is essential as internet of things (IoT) devices may also be attacked by hackers (Ronen and Shamir, 2016). Exposure of IED and IoT devices to internet have posed serious security and reliability threats (Meneghello et al., 2019), (Munshi et al., 2020). Energy deficient countries usually have modern technologies limitations; therefore, it is more secure and reliable to use smart IoT devices and artificial intelligence (AI) based algorithms in nanogrids to reduce their own loads using under voltage and under frequency relaying mediated smart load shedding. This live load shedding may be initiated with fall of voltage and frequency during peak hours to flatten the duck curve. Empirical relations to level solar and wind curves will be discussed later (eqs. (71), (72))).

Microgrids need novel control and energy management systems to increase resilience and sustainability in changing climate (Pourbabak et al., 2017). Innovative energy management systems require smart IoT devices to impart artificial intelligence to microgrids (Su and Wang, 2012). Microgrids require large scale IoT devices, high data rate Wi-Fi systems for software and hardware based solutions (Manur et al., 2018), (Soliman et al., 2013). IoT based load limiters, energy saving loads, demand response compliant appliances play greater role in microgrids. Smart telemetry, data acquisition and SCADA systems allow utilities to control generation according to the demand. Utilities have always used integrated control for load management. DSM skills have enabled consumers to integrate local control for even more efficient energy management in microgrid (Kul and Şen, 2017)– (Shuvra and Chowdhury, 2019). Changing climate may inflict cyclone, hurricane, heatwave, or flood anywhere. Puerto Rico decided to go for milligrid architecture to increase grid resilience. National grid was changed into several miligrids each composed of multiple microgrids. Rollout of IT, IoT and ICT devices in microgrids gives strength and stability to grid by information exchange and local control capabilities. DC microgrids hold key to solution power problems using PV rooftop systems (Fadel et al., 2019). IoT devices work equally well in AC, DC and hybrid AC/DC microgrids (Bo et al., 2010). IoT based DSM has been demonstrated Arduino and Matlab (Raju et al., 2020).

The internet of things (IoT) is a system of unified intelligent devices, mechanical transducers, digital technologies, metering devices, sensors, relays or monitors having a unique identifiers (UIDs) and ability to transfer data over network without human or any other agent interactions. IoT devices may connect domestic dumb technologies like TV, microwave, lights, motors, refrigerators, heaters, fans, cars and air conditioners in nano and microgrid environment for home automation, home energy management and remote control. Smart grid is utility application of intelligent devices which with support of consumer appliances embedded smart devices collect data to manage transformer overloading in distribution system (Yadav and Prasad, 2019). IoT interconnects myriad of devices with network allowing interactions between humans and machines. Networks transmit data allowing sharing of data and hardware with other devices. It is network of physical devices – things – having embedded sensors, software, and communication modules capable of connecting and data sharing with other connected devices and system through the internet. We will discuss use of IoTs and big data analytics to mitigate the duck curve at utility scale in future work.

Section snippets

Demand side load management

Live load management philosophies may be implemented by utility integrated control such as brownouts, demand side management, load shedding using under voltage and under frequency relays. ICT enabled VTs and CTs act as IoT devices providing voltage and current signals to multifunction relays having under voltage, over current and under frequency functions.

Emerging grid technologies

Nano, micro and milli grids are integral parts of smart grid. These small scale grids share their individual resources to solve smart grid demand-supply problems. Nano, micro and milli grids supply local power needs and support macro national grids in emergencies.

Results and discussions

Demand side management was initiated by acquiring near real-time data of voltage and frequency in the main master node with myRIO-1900 reconfigurable I/O device. It is compact embedded Zynq-7000 FPGA based device with dual-core ARM Cortex-A9 processors. Like traditional microcontrollers, myRIO provides similar input, output and communication channels. The device itself is connected to the designed energy management system over Wi-Fi. The goal of a Data Acquisition system is to acquire/capture

Future work

Modern trends promote near real time synchronization between cyber-physical systems. Digital twins (DT) are designed to mimic exact physical configurations of equipment in digital space. The same system is simulated or controlled in a digital software. However, simultaneous access to network data for various operations such as forecasting, data analysis, data storage, processing, automation etc. becomes an inescapable problem which requires distributed computing based solutions. Cloud computing

Conclusions

Recent trends in decentralizing and digitizing the power system with renewable energy technologies and information communication technologies in hope to manifest net zero energy buildings and sustainable communities, put huge emphasis on software and hardware development to manage the resources and loads of these buildings. Internet of things (IoT) empowered by information communication technologies play a crucial role in this regard. Widespread proliferation of IoT based smart devices and a

CRediT authorship contribution statement

Ali Raza Kalair: Conceptualization, Methodology, Software, Validation. Naeem Abas: Data curation, Writing - original draft, Supervision. Mehdi Seyedmahmoudian: Visualization, Investigation. Shoaib Rauf: Data curation, Resources. Alex Stojcevski: Visualization, Investigation. Nasrullah Khan: Writing - review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This research was supported by a grant from the Higher Education Commission of Pakistan under Technology Development Fund (TDF) (Project ID No. HEC TDF02-086).

References (56)

  • Farrokh Aminifar

    Synchrophasor measurement technology in power systems: Panorama and State-of-the-Art

    IEEE Access.

    (2014)
  • F. Blaabjerg et al.

    Distributed power-generation systems and protection

    Proceedings of the IEEE

    (2017)
  • D. Bo et al.

    Energy management of hybrid DC and AC bus linked microgrid

  • California Independent System Operator

    What the Duck Curve Tells Us about Managing a Green Grid,” Calif. ISO, Shap. A Renewed Futur.,

    (2012)
  • Chelsea H, “The World’s Winds Are Speeding Up.,” Scientific...
  • M. Derafshian et al.

    “Special protection scheme against voltage collapse,” IET Gener

    Transm. Distrib.

    (2016)
  • Domenico Talia Paolo Trunfio Fabrizio Marozzo, Data Analysis in the Cloud Models, Techniques and Applications, 1st Ed....
  • Eclypse Z7 Hardware Reference Manual [Digilent Documentation]...
  • W.A. Elmore

    Protective relaying theory and applicantions

    Protective Relaying Theory and Applications

    (2003)
  • Ethan S, “Our Sun Is Lighter than Ever, and the Problem Is Getting Worse.,”...
  • A.M. Fadel et al.
  • M. Gao et al.

    Short-term forecasting of power production in a large-scale photovoltaic plant based on LSTM

    Appl. Sci.

    (2019)
  • P. Händel

    Properties of the IEEE-STD-1057 four-parameter sine wave fit algorithm

    IEEE Trans. Instrum. Meas.

    (2000)
  • Y. Hu et al.

    A chance-constrained economic dispatch model in wind-thermal-energy storage system

    Energies

    (2017)
  • R. Isomura et al.

    Unit Commitment using advanced direct load control

  • A.T. Johns et al.

    Digital Protection for Power Systems

    (1995)
  • M. Kezunovic et al.

    The role of big data in improving power system operation and protection

  • S.F. Khahro et al.

    Assessment of wind power potential at Hawksbay, Karachi Sindh, Pakistan

    TELKOMNIKA Indones. J. Electr. Eng.

    (2013)
  • Cited by (13)

    • Two-sided sustainability: Simulating battery degradation in vehicle to grid applications within autonomous electric port transportation

      2023, Journal of Cleaner Production
      Citation Excerpt :

      However, the accompanying decentralization of renewable power generation through individual wind and solar plants and the fact that power output is less predictable due to weather and seasons are seen as major challenges for grid stability (Doleski, 2017; Kroposki et al., 2017). The resulting relevance of system services for grid stability puts concepts such as "demand response" and "demand-side integration" at the center of attention to meet the increased demand for balancing power (Kalair et al., 2021; Rollert, 2022; Spieker et al., 2016). Similarly, in the consumer-side implementation in the so-called Smart Grid, attention is increasingly focused on integrating energy-intensive industries, such as logistics, steel, and cement (Zeng et al., 2015).

    • Understanding the energy consumption of information and communications equipment: A case study of schools in Taiwan

      2022, Energy
      Citation Excerpt :

      Our study is the first to explore the EC of ICT in schools in Taiwan, and the results can be used as a guide by schools and education authorities. ICT EC is usually discussed in the context of large-scale applications, such as worldwide, national, or smart city economic development [26–29]. ICT applications have indeed improved the operational efficiency of different sectors, increasing their performance and reducing unnecessary transportation.

    • A two-stage stochastic Stackelberg model for microgrid operation with chance constraints for renewable energy generation uncertainty

      2021, Applied Energy
      Citation Excerpt :

      Another option is to keep fossil generators on standby to go through the duck curve. Finally, the last option considers the installation of more green generators, a solution studied and proposed by authors in [39,40]. However, this last criterion is unreliable due to the significant variability of renewable resources (a problem covered by this study) and critical levels of curtailments of renewable energy that can emerge in systems without sufficient power-to-X flexible technologies.

    • Optimal design of bifacial PV system to mitigate duck-curve problem of power system with the UC problem

      2021, Energy Reports
      Citation Excerpt :

      The duck curve problem includes three major challenges. First, the Netload, which is the difference between PV power generation and demand, decreases significantly in the daytime, it leads to a lack of stability against the disturbance due to the decrease in the synchronous power of the entire system, due to the shutdown generators (Kalair et al., 2021). Second, it frequently happens which is MW scale power imbalance fluctuations during the day by weather uncertainty (Muzhikyan et al., 2019).

    View all citing articles on Scopus
    View full text