The grid-connected microgrid: local network of the future?

October 15th, 2015, Published in Articles: Energize

 

The microgrid had its origin in off-grid applications which comprised of several generation sources, having all the features of smart grid was able to balance supply and demand. The development of advanced microgrid controllers has facilitated connection of microgrids to the main grid and lead to an increase in possible applications of grid tied microgrids.  

There are three classic examples of microgrid (MG) communities experienced in the past , each experiencing its own set of challenges to sourcing quality power, which were met by implementing the customised technology-based solution. These are:

  • Isolated autonomous: which can be found on islands or remote mainland areas with no connection to the main grid. There are still many communities using off-grid power, and the availability of advanced control equipment improves availability and quality of power in isolated grids.
  • Semi-autonomous: which occur in remote locations on the mainland such as remote communities, research stations, defence bases and industrial sites or mines.
  • Weakly-connected: which appear at the ends of the lines of larger traditional grids, and may also appear in facilities that can go ‘off-grid’ when desired.

In addition to these traditional models, a new MG, normally connected to a strong grid, is emerging. These applications are made up of large consumers or groups of consumers that have installed renewable energy (RE) on site, for reasons of security and energy savings, but who still draw a major portion of their energy from the grid. The MG is configured so the RE system functions both under grid connected conditions, to save energy, and under islanded condition, to save fuel.

Fig. 1: The IREC microgrid control system (Colet-Subarachs [5]).

Fig. 1: The IREC microgrid control system (Colet-Subarachs [5]).

A further development is the resilient microgrid, which comprises groups of consumers in a distribution network where renewable energy or embedded generation has been installed. These distribution networks are normally connected to a strong grid but are designed to function autonomously under disaster conditions such as extreme weather, load shedding and  natural and manmade disasters. The resilient grid has been developed in response to the situation created by hurricane Katrina and other natural disasters, and has been aided by the rise in distributed generation and smart grid technologies. Resilient grid policy is being implemented widely in the USA.

Some developers are putting microgrids into new construction. In Sacramento, California, a 34-unit residential complex is being purpose-built with an integrated microgrid designed by Sunverge Energy. The system will automatically switch residents to the cheapest power source, whether solar or conventional, while storing backup power for use if the grid goes down [1].

Today’s microgrid market is extensive and covers the following areas:

  • Institutional and campus microgrids
  • Commercial and industrial microgrids
  • Military microgrids
  • Community and utility microgrids
  • Resilient microgrids.

Microgrid developments

A modern MG will include renewable and fossil fuel generation, energy storage facilities and load control. The microgrid is also furthering the development of the smart grid, as a first stage of development, where smart grid technologies and principles can be applied while remaining connected to a stable main grid. The development of renewable energy systems using synchronous converters has made the concept of an islanding MG more possible, as such converters can easily be adapted to centralised frequency control or generation by a controller in the MG, and frequency is not dependant on loading, as it is with standby generators. This is important in an MG where the majority of the energy is from renewables, which require a reference frequency to operate correctly.

An extreme case would be a MG which normally operates in island mode and uses the grid for a back-up or supplementary energy source. There are several university campuses operating in this mode, although these must be considered experimental in nature.

Industrial and mining grid connected microgrids

These are essentially hybrid RE/diesel systems which were developed from existing standby plant installed on industrial or commercial properties. The addition of rooftop solar to such properties in recent years led to a need for a system to coordinate and control the combined operation of these sources, as well as the load, under both grid-fail and grid-connected conditions. The use of microgrid controllers added a number of functions which were not available with either system and which has greatly improved the energy management capabilities of the installation.

Elements of a microgrid

In addition to generation sources and loads, modern MGs comprise several additional elements:

Microgid controllers

There are a number of MG controller units available on the market. The functions of a typical controller are [1]:

  • Grid interactive control functions: Area EPS control, spot market, DMS, transmission SCADA, and connection to adjacent microgrids.
  • Supervisory control functions:  Forecasting, data management and visualisation, optimisation (e.g.
    volt/var, economic dispatch), dispatch, emergency handling, generation smoothing, spinning reserve, black start, and protection coordination.
  • Local area control functions: Status control, load management, energy management, plant controller, AGC, fast load shedding, resynchronization, and disturbance recording.
  • Device level control functions: Voltage/frequency control, reactive power control, energy storage control, load control, generation control, and fault protection.

Energy storage systems

Studies have shown that some form of storage is necessary for islanded operation of the microgrid, as well as stabilisation during grid connected operation. Battery storage is the most common form used.

Load and generator monitoring and control systems

These are required to gather data from the components of the grid and execute control functions under the control of the MG controller.

Controllable generation and loads

An important function of the controller when operating in island mode is the balancing of generation with load. When the load does not match the available generation, as may occur in a MG with a high percentage of RE sources or when all sources are not available, some form of load control or generation curtailment or localised demand side management (DSM) is needed.

To achieve this on a centralised basis, the following will be required:

  • All generation sources must be monitored, and information on the generation capacity must be available to the MG controller. It must be possible to alter the power output of generators, or shut down generators by the controller.
  • All load connection points must be monitored and information on power consumption in real time must be available to the controller
  • It must be possible to reduce, increase or disconnect the load from the controller.

To achieve this aim all loads or load connection points must be fitted with monitoring and control devices. This requirement has been taken to the extreme in the case of several single facility MGs, where a large number of loads are connected via socket outlets [4]. A system deployed in a stand-alone farm in Girona, Spain, NoBaDis, is based on intensive use of information to maximise the penetration of renewable energy through two main strategies: anticipation and opportunity. The loads are classified in four different types according to priority and are controlled by intelligent sockets interconnected in a mesh network. Each socket measures the main electric parameters, communicates to the local computer bi-directionality by using ZigBee communications, and is able to switch on/off the loads. The power system is composed by low-density energy sources (renewable) and high-density cogeneration engine [4].

Ore advanced system developed by IREC (Catalonia Institute for Energy Research) [5] uses a two level hierarchy of microgrid controller (called the i-node) and intelligent socket (called the i-socket) which controls either a generation source or a load as shown in Fig. 1. This is implemented as part of the smart city project located in Malaga, Spain.

The communication architecture of such a system is composed of a hierarchical layer system. The bottom layers are embodied by these two elements. The iNode (intelligent node) develops the global management of MG tasks and connects supervising and control systems(through a gateway) to the terminal equipment (iSocket). Its functions are managing the data received from iSockets and setting overall operation of the MG, developing its own algorithms.

Its main tasks include:

  • Regulation: control of energy generation and consuming entities
  • Billing: energy measurement and real-time pricing
  • Management: asset management and condition based maintenance
  • Metering: full system monitoring
  • Security: securing the microgrid’s electrical system

The operational requests of this controller are aggregation and coordination of iSockets and electrical safety guarantee.

The iSocket is an element located in the lowest hierarchy layer of the communication system. It handles the device connected to it (generation, storage or load), based on the instructions received from the iNode. The operational requests of this controller are local regulation and electrical safety. The system can operate in a centralised or distributed mode. In centralised mode the sockets report to and are controlled by the MG controller. In distributed mode each socket operates independently according to a set points.

In grid connected mode the overall objective is to optimise operating performance and cost while ensuring that the system is capable of meeting the performance requirements in stand-alone mode. One very appealing technology essentially allows the MG to behave as an aggregated power entity that can be made dispatchable by the utility. Particularly beneficial to the utility is the fact that this feature can be designed to compensate for intermittency associated with renewable energy resources such as wind energy and solar energy, essentially pushing the management burden inside the MG.

Frequency control

In grid connected mode, distributed generators and storage within the MG will synchronise to the frequency and magnitude of the grid voltage, and will optimise the energy supply as determined by the energy management unit. Grid voltage and frequency stability are maintained by the large generators connected to the grid. In islanded mode however, steady state and dynamic power balance between load, generation and energy storage within the MG must be maintained without any dependence on the grid or communications infrastructure, an important requirement that has been proven in research projects. This is possible with intelligent MG controllers for generators and load management units using the classical “frequency droop control”, voltage droop control and frequency dependant load control principals.

Electronic converters used in distributed generators (DGs) require a frequency reference source for operation. In a grid connected mode the DGs derive their frequency from the main grid, as if they were connected individually to the lines. In the disconnected mode the frequency source will be the MG controller. The studies also show that in order to maintain frequency balance in an islanded MG, there is need for a reference sine wave generator inside the master unit which imitates the main network phase voltages [3].

Stable operation of the MG both in grid connected and islanded mode, seems to favour require the use of electronic controls and inverters in the DGs. This cuts out the use of traditional rotary devices, unless an intermediate stage is applied. There are nonetheless, numerous newer fossil fuel based resources having adequate capacity to supply the needs of a small network. Examples are fuel cells and micro-turbines, both of which use converters.

Case study

ABB South Africa has announced its intention to install an MG at the company’s premises in Longmeadow, Gauteng. The MG solution includes a rooftop solar photovoltaic (PV) field and an energy storage grid stabiliser, PowerStore [6].

A 750 kW rooftop PV plant together with a 1 MVA/380 kWh battery-based PowerStore will be added to the site’s existing back-up diesel generators. This will enhance the use of renewable energy and provide continuity of supply when the regular power supply is disrupted as well as during transitions from grid to island operation. One of the big problems with MGs is the variable output from both wind and solar PV systems. Adding a power stabiliser solves this problem.

Under normal grid-tie conditions, the solar PV will disconnect when the grid goes down, and backup power will only be provided by standby generators. Some more advanced systems will allow the PV to synchronise with the standby generators once they have started, to save fuel. The process will reverse once the grid returns. When the standby plant shuts down the PV system will disconnect and resynchronise with the grid.

The system provides power during the start-up time of the standby plant and may obviate the shutdown and resynchronisation of the PV. In the same way, when the grid returns the system provides power during the shutdown and changeover to the grid.

Furthermore, the system compensates for short term variations in the solar PV output, which can be expected when clouds pass over the plant. This prevents hunting and associated frequency deviations from the standby plant. This application of the system is essentially a hybrid grid-tied PV/diesel system, with power stabilisation. The system can also respond to sudden changes in load, especially dropped loads, and absorb short bursts of excess generation.
The unit being installed will use battery storage, but it may also be equipped with flywheel storage unit [6].

References

[1]    J Reilly: “Microgrid controllers: Standards for specifications and testing”, www.nrel.gov/esi/pdfs/agct_day3_reilly.pdf
[2]    District energy: “Big corporations embracing microgrids: A threat for utilities?”, www.districtenergy.org/blog/2013/10/25/big-corporations-embracing-microgrids-a-threat-for-utilities/
[3]    Siemens: “Microgrid white paper”, www.siemens.com/download?DLA17_8
[4]    P Salas: “Mas Roig mini-grid: A renewable-energy-based rural islanded microgrid”, Energy Conference (ENERGYCON), 2014 IEEE International
[5]    A Colet-Subirachs, et al: “Centralized and distributed active and reactive power control of a utility connected microgrid using IEC61850”, http://ieeexplore.ieee.org/
[6]    ABB: “ABB to install microgrid solution in South Africa integrating multiple energy sources”, www.abb.co.za/cawp/seitp2/04947a6aade4010ac1257eb5003172d4.aspx

Send your comments to: energize@ee.co.za

Related Articles

  • Greater flexibility in electricity generation essential, delegates hear
  • Medupi magic fails Eskom and South Africa – the stuff of nightmares
  • Government urged to expedite IRP
  • Power developments in Africa, October 2019
  • Machine learning solves grinding mill liner monitoring