Power Management in Microgrids

As microgrids are becoming more common, the issues linked with micro-grids need to be addressed to effectively improve the quality of delivered power. This article discusses the architecture of micro- grids, control components and systems and integration strategies to achieve higher efficiency, resilience and sustainability…

March 5, 2018

powermanagement-microgrids-microgrid-control

Microgrids provide a new infrastructure for more efficient, resilient and cost-effective power systems. This architecture works like a power nest with scattered conventional and non-conventional energy sources throughout the distribution network. Micro-power sources combined with their interfaces are categorized as distributed generators (DG). Distributed generators work according to load demand and their converters adjust voltage and frequency and harmonize it with the network such that the load requirements of the end users are met. Distributed generators are considered as small power generating units and are usually rated at < 100kW. Solar photovoltaics, fuel cells, wind turbines, micro-turbines, flywheels and super capacitors can all be used as micro-power sources. Power electronic converters are utilized to connect these sources to the supply network through a common grid bus.

The concept of practical microgrid originated from the Consortium of Electric Reliability Technology Solutions (CERTS) as “a system consisting micro-power resources providing electric as well as heat power to the distributed loads”. The energy conversion, control and storage is taken care of by power electronic devices and components. It is capable of operating as a singly controlled element or with the main grid simultaneously. CERTS’s micro-grid presents a systematic architecture, design, control, and operation in regulating, managing and providing power. This microgrid is shown in figure 1 with connections to micro-power sources at nodes 8, 11, 16 and 22. It is interfaced with the main grid through a point of common coupling (PCC). The specific architecture varies with different types of load, communication, control and monitoring technologies. Structural variants are based on the distributed nature of the supply network and have high efficiency and reliability. The control of microgrid is a hierarchical strategy with primary (local) and secondary (supervisory) controllers. The job of primary control is to regulate active and reactive power based on local voltage and frequency information. This information is used by distributed generators to adjust their output so as to maintain the balance of power. The secondary control provides supervisory actions, monitoring and protection through various ICT.

Under the recent foundational framework of IEEE 519- 2014, the supplier of electricity is responsible for the quality of power supplied. The end-user is responsible for limiting harmonic current injections based on the size of the end-use load relative to the capacity of the system. Large-scale use of DG units may lead to transients, voltage instability, power fluctuations, and harmonics. The unpredictable and unstable nature of renewable energy resources leads to output power fluctuations. Connection and disconnection of micro-sources from the main grid causes unbalanced voltage and small voltage drop due to heavy loading conditions causes partial voltage instability. As micro-grids are becoming more common, the issues linked with micro-grids need to be addressed to effectively improve the quality of delivered power. This paper discusses the architecture of micro- grids, control components and systems and integration strategies to achieve higher efficiency, resilience and sustainability.

Figure 1: High level diagram of CERTS micro-grid [2]

Power Management Strategies

The role of different control components and power conditioning units is very important to manage power flow during different modes and transitions. CERTS recommends that the addition of new micro-power sources without damaging the existing system, the connection to or isolation from grid seamlessly, active and reactive power processing corresponding to load variations and independence of choosing operation points should all be possible. It can be categorized into different approaches depending on the architecture of micro-grid systems, inter dependency of micro-grid components and the nature of load demand.

A. Power Flow Control

To understand how power flow is controlled in micro-grids, let’s analyze the model of micro-sources first. There are two fundamental components of a micro-source model. A DC interface and a voltage source inverter (VSI). The magnitude and phase of the output voltage is controlled by the voltage source inverter. Output voltage and frequency values depends on real and reactive power of the micro-source. The active and reactive power injected by the DG units to the node can be given by an inductor at the point of common coupling (PCC).

Figure 2: Active Power-Frequency control

B. Active and Reactive Power Regulation (P/Q)

In grid-connected mode of operation, the main grid controls the micro-grid load power and fluctuations in voltage and frequency to meet load requirements. The micro-grid does not play any role in regulating the voltage and frequency. The controller is configured to operate in cascaded double loop control for feed-forward compensation in DC voltage. The outer loop is designed to decouple active and reactive power while the inner loop is designed for current feedback regulation.

C. Voltage and Frequency Regulation (V/f)

In islanded mode of operation, adequate voltage and frequency support is needed for stable standalone operation. This can be provided by making the use of inverter voltage and frequency regulation. Local voltage and frequency is used as feedback. The requirement of supervisory control for global communication is waived off. Hence, this strategy is typically useful for islanded mode of operation.

D. Load Frequency Control

There are two different strategies to regulate voltage and frequency as a result of load unbalance during islanded operation. Local secondary control (through a controlled DG unit) and micro-grid central control (MGCC). The anticipated value of the reactive power of the prime mover of the two cases can be determined as per frequency deviation. Figure 2 shows characteristics curve determined by the MGCC known as droop curve during islanded operation for frequency restoration satisfying the balance of power while importing from or exporting to the grid.

where, V is grid voltage, E is inverter output voltage and L is inductive reactance. Hence, the relationship between the inverter voltage, system voltage and the inductive reactance determines the flow of real and reactive power from the system. The micro-source couples to the power system using an inductor at the point of common coupling (PCC). PCC is a node that connects micro-grid DG units to the main grid.

E. Active and Reactive Power Regulation (P/Q)

F. Voltage and Frequency Regulation (V/f)

G. Load Frequency Control

H. Decentralized Control

In decentralized control, disjoint control models can be improvised to regulate active and reactive power. It maintains the balance of power effectively and adds the frequency recovery mechanism.

I. Harmonic Compensation

In the distribution feeders, distributed generators act like a non-linear loads while the inverters generate higher order harmonics. Non-linear current flowing through the load causes voltage distortions that appear at the point of common coupling. The type and severity of harmonics depend on the power converter technology and interconnection configuration. The percentage of harmonics relative to fundamental and total harmonic distortion (THD) allowed by IEEE 519-2014 can be seen from table 1.

Table 1: Percentages of allowed harmonics (IEEE 519-2014)

Harmonic mitigation approaches are based on making the distributed generators of a distributed power system behave as a resistance at harmonic frequencies. This useful, low cost and compact design improves power quality effectively without the need of separate controllers in d-q (synchronous) and αβ (stationary) reference frames and additional phase-locked loops (PLL) and provides smooth output voltages and currents in grid-connected and islanded modes of operation. A capacitive virtual impedance loop can also be used to damp the voltage harmonics at the point of connection with load by introducing a capacitive component and effectively distorting the output voltage of inverter. Hence, harmonic voltage is obtained via virtual impedance to the inner loops that provide compensation against drop due to inductive grid side impedance.

Figure 3. State transitions diagram for power sources in (a) Grid-connected mode (b) Islanded mode

Intelligent Control

The evolution of power grids has led to the usage of ICT to meet the needs distributed power infrastructure. The development of intelligent technologies and systems is a major step forward to handle the complexity of highly integrated and communicative architecture in distributed energy systems. To cope with the challenges of efficiency, reliability and security, some intelligent control design strategies are proposed as follows:

A. CPS based Design

The change in hierarchy of electrical power industry needs efficient and economic utilization of energy. The deployment of advanced technology such as smart grid in modern power industry to minimize continuous interruptions demands the use of information and communication systems. The inter-compatibility and absorbing these distributed energy resources (DERs) into this competitive market from economic and reliability point of view is a modern challenge. The use of Cyber-Physical Systems (CPS) which uses intelligent electronic devices (IEDs) to measure physical parameters, actuators to control and finally communicate the processed information is a proposed turnkey solution for this problem. The cyber physical approach in energy systems can be utilized to help achieve optimized system performance in a safe real time environment. Practical renewable energy sources need to have maximum reliability and affordability. A comprehensive technique for testing and evaluation of the system performance serves as a supplement for valid execution of the real-time control based on cyber-physical modeling in order for the system to operate reliably in the condition of unwanted events.

Figure 4: Hybrid automaton of multi-source gridconnected micro-grid

A key approach to observe the real-time behavior of a system which has a physical plant that is operated by information and actuation layers has been presented. A set of simulations show that different physical parts of a smart grid act on the information provided by the operator to meet varying load demands from different types of sources such as large power generators and micro-sources and their synchronization in case of extraordinary physical and environmental conditions. The transitions among different modes of operation of main grid with two micro-grids are shown in figure 3. 1 denotes ON and 0 denotes OFF. The transitions are dictated by the supervisory control. For functional verification purposes, the model has been subjected to a grid-side fault event where it has been demonstrated that it is capable of maintaining the power supply by switching to alternately available power resources every time the system is exposed to faults. This enables a unified control design to simulate not only the electrical behavior of the grid but also the interactions with the real world.

B. Hybrid Control

The major challenges of a smart power network are the control and communication of decentralized power generating units and loads during distribution, the uncontrollable and unstable nature of distributed energy resources (DERs) integrated within the grid and power quality during transitory modes and states. In order to solve these issues, an integrated control-communication strategy capitalizing on power processing ability of a smart grid to yield smooth, safe and reliably operation is discussed. This is done by developing a real-time model of the system which is capable of performing operations and making control decisions as per utility needs. For grid transition management and supervisory control issues, loads are effectively supplied from the main grid as well as distributed energy resources in case of grid absence to improve overall system reliability, efficiency and power density.

Smart grid is operated in two basic modes; grid-connected (with-grid) and islanded (off-grid) mode. The control process of combination of micro-sources in different modes is discrete. For each micro-source and the main grid itself, the control process is continuous. Therefore, a smart grid is clearly a hybrid system and hybrid automaton strategy can be used for an operating model of a distributed grid. “An automation of a system predefines the normal and transitory operating states and predetermines the transition routes that the system follows when a transition happens due to change in operating condition”. The control strategy in the form of hybrid automaton of two source grid-connected micro-grid system is shown in figure 4. Eight combinations of three available power sources are possible. Each of these sources have ON and OFF states such as G=0 or G=1, and similarly A=0 or A=1 and B=0 or B=1.. In order to study the transient behavior to see the recovery time, a fault condition is simulated which shows that system maintains maximum reliability with varying dynamics and load demand.

C. Droop Control

To integration large numbers of micro-sources into an interconnected distributed power system, voltage regulation is necessary for local reliability and stability. Power sharing in a multi-source multi-load distributed generation system is handled by droop control. This method tunes the frequency of the power electronics inverter similar to the traditional power system frequency adjustment. For instance, as the reactive power generated by the micro-source tends to be more capacitive, the local voltage set-point is reduced. Conversely, as it tends to be more inductive, the voltage set-point is increased. Figure 5 shows micro-grid control strategy using droops. The real time values of active power, reactive power and the voltage magnitude are calculated. Desired voltage magnitude and angle at the inverter terminals are generated by the control signal. The gate pulse generator is responsible for appropriate firing pulses to the inverter to track the control’s requests. The locus where the steady state points are constrained to come to rest is called the droop.

The frequency and magnitude of the voltage is dependent on the voltage source inverter controlled by the droops. The inverter can be controlled with two types of configurations. One of them is active power-frequency (P-ω) and reactive power-voltage (Q-V). Another one is active power-voltage (P- V) and reactive power-frequency (Q-ω). Basic voltage and frequency droop equations can be given as

Where, f0, V0, P0 and Q0 are the nominal values of frequency, voltage, active and reactive power respectively. K is the droop coefficient.

Micro-Grid Protection

One of the potential advantages of DG is that it allows standalone or islanded mode of operation which provides uninterruptible supply of power. “Government regulations must guarantee that distribution generation sources are capable of supplying sufficient short circuit current”. Medium and high voltage DG systems must have fault-ride through capability. These considerations are equally applicable to low voltage DG systems. Few types of protection methods are deemed fit for micro-grids as described in the following subsections.

A. Overcurrent Protection

Overcurrent protection may utilize a communication assisted protection selectivity strategy that has different levels that are applied with voltage-restrained directional overcurrent protection. In overcurrent protection strategies, device selectivity is taken into consideration and the protection is ensured for both on-grid and off-grid modes. One type of instantaneous overcurrent protection strategy which is independent of the location of DG system uses two routines that perform instant protection for local line and remote bus. Another type of overcurrent protection takes into consideration the usage of symmetrical and asymmetrical components. This is applicable to the micro-grids that have a communication channel which is limited to exchanging status information and not electrical measurements.

In differential overcurrent protection, each protection zone has its own relays and current sensors on the secondary side of transformers for every load as well as relays located at the source side. These zone relays detect a fault (as quick as 5ms) when DG source currents exceed the sum of load currents within the protection zone. These protection techniques depend on communication channels which in case of failure may put protection at risk. However, DG sources are often provided with current limiting equipment that strengthens overcurrent protection. In low voltage micro-grids, protection is done using microprocessors controlling overcurrent relays which neither requires communication nor the dependency on fault current magnitude.

Figure 5: Micro-grid control with droop functions

B. Voltage based Protection

In this type of micro-grid protection, output voltages of DG sources are measured and then transformed into DC quantities using the d-q reference frame for in and out of zone faults. A communication link is deployed in the scheme to differentiate between in-zone and out-of-zone faults using pilot wires, ethernet or optical fibers. The faulted zone is identified by transmitting the measured voltage through communication links between any two relays in the micro-grid and comparing it with the mean average value of the two relays. If the fault voltage goes higher than the threshold (that corresponds to the fault type), the zone is tripped.

C. Analytical Protection

In this type of protection, a database of calculated relay settings for different setups is developed. This relay database is with the new settings every time a configuration changes. Action and event tables are maintained for circuit breaker statuses and those settings are then applied for many other micro-grid configurations. MGCC monitors the micro-grid’s state of operation and uses the information from the action and event tables for configuring the relays. Occurrence and direction of a fault is detected during the real-time operation by comparing measured current values with the settings of the relay. However, this protection is not very useful for large micro-grids due to memory restrictions.

Conclusions

Micro-grids are becoming increasingly popular due to their flexible design, resilient operation and by providing reliability, robustness and power quality in the electric power supply network. They can switch between grid-connected and islanded modes of operation which offers a high degree of security. A number of control techniques and algorithms exist for addressing micro-grids issues such as power quality, active and reactive power sharing, harmonic mitigation, frequency mismatch, voltage unbalance, voltage distortions etc. Active power control strategy based on droop control can be used to correct frequency deviation. To reduce the harmonic content of the system, co-control strategy can be utilized which also improves power sharing by tuning micro-grid inverter parameters online. A few intelligent architectures are provided including cyber-physical systems and hybrid automata based control. These techniques govern the operation of a distributed grid with efficiency, reliability and safety. The safe operation does not only depend on the control architecture but also the protection technique used. The best types of protection for micro-grids have the capability to communicate and provide real-time monitoring during all modes of operation.