Chapter 6

Monitoring and energy management of the microgrid

Abstract

Monitoring and energy management of the microgrid introduces the composition of the microgrid monitoring system, energy management, and optimized control methods.

Keywords

microgrid monitoring
PV monitoring
wind power monitoring
microturbine monitoring
ES monitoring
load monitoring
comprehensive monitoring
design of monitoring system
energy management
forecast of DG
load forecast
frequency response characteristics
power balance control
load shedding
generator tripping
important loads
optimization control
power exchange
The monitoring and energy management system of the microgrid serves real-time, extensive monitoring of distributed generation (DG), energy storage (ES), and loads within the microgrid. In grid-connected operation, islanded operation, and during transition between operation modes, it controls and optimizes the DG, ES, and loads, thereby ensuring secure and stable operation of the microgrid at the maximum energy efficiency.

6.1. Monitoring

6.1.1. Structure of the monitoring system

The monitoring system coordinates with the local control and protection and remote distribution dispatch, and has the following functions:
1. Real-time monitoring of supervisory control, data acquisition, and DG;
2. Service management: forecast of power flow (including tie line power flow, DG node power flow, and load flow) and DG output; DG output control and power balance control;
3. Smart analysis and decision-making: optimized dispatch of energy.
Figure 6.1 shows the structure of the monitoring system.
image
Figure 6.1 Structure of the microgrid monitoring system.
By collecting information of DGs, lines, the distribution network, and loads in real time, the monitoring system monitors the power flow across the microgrid and adjusts the operation of the microgrid in real time based on operation constraints and energy balance constraints. In the monitoring system, energy management is the core that integrates the DG, load, ES, and the point of common coupling (PCC). Figure 6.2 shows the functional architecture of the energy management software.
image
Figure 6.2 Functional architecture of the energy management software of the microgrid monitoring system.

6.1.2. Composition of the monitoring system

The microgrid monitors the DGs, ESs, loads, and control devices in the system. The monitoring system comprises photovoltaics (PV) monitoring, wind power monitoring, microturbine monitoring, monitoring of other power sources, ES monitoring, and load monitoring.

6.1.2.1. PV monitoring

The operation and alarm information of PV units is monitored in real time for comprehensive statistics, analysis, and control of PV electrification, as shown in Figure 6.3.
image
Figure 6.3 PV monitoring and statistics.
PV monitoring can achieve the following functions:
1. Real-time display of total power, daily total power output, aggregate power output, aggregate CO2 emissions, and daily power output curve.
2. View of inverters’ operation parameters, including DC voltage, direct current, DC power, AC voltage, alternating current, frequency, present power, power factor, daily power output, aggregate power output, aggregate CO2 emissions, inside temperature, and 24 h power output curve.
3. Monitoring of inverter operation, warning of equipment failure with audio and visual alarms, and identification of failure cause and occurrence time. Failure information includes excessively high grid voltage, low grid voltage, high grid frequency, low grid frequency, high DC voltage, low DC voltage, overload of inverter, overheating of inverter, short circuit of inverter, overheating of radiator, islanding of inverter, and communication failure.
4. Forecast of short-term and super-short-term power output, providing the basis for optimized dispatch of energy.
5. Adjustment of power output and control of start and stop of inverters.

6.1.2.2. Wind power monitoring

The operation and alarm information of wind turbine generators are monitored in real time for overall statistics, analysis, and control of wind power, as shown in Figure 6.4.
image
Figure 6.4 Wind power monitoring and statistics.
Wind power monitoring is mainly intended for the following:
1. Real-time display of total power, total daily power output, aggregate power output, and 24 h power output curve.
2. Collection of operation data of wind turbine generator, including three-phase voltage, three-phase current, grid frequency, power factor, output power, generator speed, rotor speed, temperature of generator windings, oil temperature in gearbox, ambient temperature, temperature of control board, wear and temperature of mechanical brake lining, cable twisting, nacelle vibration, anemometer, and wind vane.
3. Forecast of short-term and super-short-term power output, providing the basis for optimized dispatch of energy.
4. Adjustment of power output and control of start and stop of inverters.

6.1.2.3. Microturbine monitoring

The operation and alarm information of microturbines is monitored in real time for overall statistics, analysis, and control.
Microturbine monitoring is mainly intended for the following:
1. Monitoring of major operation parameters, including speed, gas inflow, gas pressure, exhaust pressure, exhaust temperature, knock intensity, and oxygen content;
2. Monitoring of voltage, current, frequency, phase, and power factor before and after being connected to the grid;
3. Analysis, management, and adjustment of operation status.

6.1.2.4. Monitoring of other power sources

Similar to generation monitoring mentioned earlier, for other power sources, present output voltage of DG, working current, input power, grid-connected current, grid-connected power, grid voltage, current power output, aggregate power output, 24 h power output curve, and 24 h grid-connected power curve need to be monitored, for the purpose of ensuring secure and stable operation of the system.

6.1.2.5. ES monitoring

The operation and alarm information of batteries and power control system (PCS) is monitored in real time for overall statistics, analysis, and control of ES, as shown in Figure 6.5.
image
Figure 6.5 ES monitoring.
ES monitoring is mainly intended for the following:
1. Real-time display of energy that can be discharged or charged, maximum discharge power, present discharge power, discharge duration, total energy charged, and total energy discharged.
2. Remote communication of operation status, protection information, and alarm information of AC/DC bidirectional converter. Protection information includes under-voltage protection, over-voltage protection, open phase protection, under-frequency protection, over-frequency protection, over-current protection, component failure protection, battery failure protection, and overheating protection.
3. Remote measurement of battery voltage, battery charge, and discharge currents, AC voltage, and input and output power of the AC/DC bidirectional converter.
4. Remote regulation of battery charge and discharge duration and currents, and battery protective voltage, allowing for remote regulation of related parameters of the AC/DC bidirectional converter.
5. Remote control of battery charge and discharge of the AC/DC bidirectional converter.

6.1.2.6. Load monitoring

The operation and alarm information of loads is monitored for overall statistics, analysis, and control of loads, as shown in Figure 6.6.
image
Figure 6.6 Load monitoring.
Load monitoring is mainly intended for the following:
1. Monitoring of load voltage, current, active power, reactive power, and apparent power;
2. Recording of maximum load and occurrence time, maximum three-phase voltage and occurrence time, maximum three-phase power factor and occurrence time, and statistics and monitoring of voltage eligibility rate and black-out occurrence time;
3. Warning of overload, and query of historical curve, reports, and events.

6.1.2.7. Overall monitoring

The overall operation information of the microgrid is monitored, including the frequency of the microgrid, voltage at the PCC, and power exchange with the distribution network; the total power output, state of charge (SOC) of the ES, total active loads, total reactive loads, total active power of sensible loads, total active power of controllable loads, and total active power of loads that can be completely shed are collected in real time; and the status of all circuit breakers in the microgrid, active power and reactive power of all branches, and alarm information of all equipment are monitored in real time, thus realizing real-time monitoring and statistics of the entire microgrid (see Figure 6.7).
image
Figure 6.7 Overall microgrid monitoring and statistics.

6.1.3. Design of the monitoring system

The monitoring system of a microgrid should be designed to enable management and control of the distribution network dispatch layer, central control layer, and local control layer. The distribution network dispatch layer coordinates multiple microgrids (microgrid acts as a single controllable entity with respect to the macrogrid) to maintain security and economy of the distribution network, and the microgrid is regulated and controlled by the distribution network. The central control layer centrally manages the DG sources (including ES) and various types of loads, maximizes utilization and optimizes operation of the microgrid in grid-connected operation, and regulates the output of DG sources and load demands in islanded operation, thereby maintaining stability and security of the microgrid. The DG controller and load controller at the local control layer maintain transient power balance and under-frequency load shedding, ensuring transient security of the microgrid.
The monitoring system, a core link in the microgrid that integrates the DG, load, ES, and the PCC, generates control and regulation strategies based on fixed power balance algorithms to ensure stability of the microgrid in grid-connected or islanded operation and during transfer between various modes. Figure 6.8 exhibits the model of the energy management controller.
image
Figure 6.8 Model of the energy management controller.
The coordination of microgrid local control and protection, microgrid central monitoring and management, and remote control of distribution and dispatch can achieve power balance and control within the microgrid by control of the tie-line power flow. Figure 6.9 presents the coordinated control of an entire distribution system interconnected with the microgrid.
image
Figure 6.9 Coordinated control of distribution system interconnected with microgrid.
In addition to data acquisition, the following problems should be considered in designing the microgrid monitoring system to facilitate control, management, and operation of the microgrid:
1. Protection of the microgrid: Solutions should be recommended based on the rational settings of various protections of the microgrid and online check of rationality of protection settings, with a view to avoiding blackout due to unintentional operation of the protection under some circumstances.
2. Integration of DG: There are various types of DG sources in a microgrid that are distributed across the system and produce power intermittently. As such, the solution should provide ways for reasonable integration and coordination of DG sources to maintain stability of the microgrid both in grid-connected and islanded operations.
3. Forecast of DG output: The output of wind power and PV in a super-short-term is forecast based on the weather information released by the bureau of meteorology, historical meteorological information and historical records of power generation, thus enabling forecast and control of the microgrid.
4. Microgrid voltage/reactive power balance and control: The microgrid, as a single controllable power system, should meet the requirements placed by the distribution network on its power factor or absorption of reactive power to avoid long-distance transmission of reactive power, and should maintain a high voltage quality through voltage/reactive power balance and control when operating in parallel with the distribution network.
5. Microgrid load control: In islanded operation or when the distribution network has special requirements on loads or power output of the microgrid while the output of DG sources is fixed, the loads need to be shed, reconnected, or regulated based on their importance, so as to keep reliable power supply to important loads and maintain security of the microgrid.
6. Microgrid power output control: In islanded operation or when the distribution network has special requirements on loads or power output of the microgrid, the security and economy of the microgrid can be maintained by reasonably regulating the output of various DG sources, and in particular, by charging and discharging of batteries, and coordinating microturbines between cooling, heat, and power.
7. Multilevel optimized dispatch of microgrid: Load control and generation control are coordinated in various operation modes (grid-connected mode and islanded mode) and at various levels (DG, microgrid, and dispatch) to maintain the security and economy of the entire microgrid and provide support for optimized dispatch of the distribution network.
8. Coordination between the microgrid and utility grid: A microgrid can be deemed as a load or power source for the utility grid. Coordination between the two is conducive to reducing the loss of the distribution network and shifting loads. And when a serious fault occurs on the utility grid, the reasonable output of the microgrid will speed up the recovery of the grid.

6.2. Energy management

Microgrid energy management is intended to maintain security and stability and improve energy efficiency of the microgrid by forecasting the DG, ES, and loads within the microgrid and optimizing control over the DG, ES, and loads based on their characteristics in grid-connected operation, islanded operation, and during transfer between different operation modes.

6.2.1. Forecast of DG

Forecast of DG, part of the microgrid energy management, is to forecast the short-term and super-short-term output of DG (wind power and PV) to provide a basis for optimized energy dispatch. It improves the use of DG sources, economic and social benefits, and increases reliability and economy of the microgrid.
The output of DG can be forecast by statistical methods or physical methods. The former is to find out the inherent law by collecting and analyzing historical data; and the latter to calculate using physical equations with meteorological data as the input.
Currently, continuous predication method, Kalman filtering method, random sequential method, artificial neural network method, fuzzy logic method, spatial correlation method, and support vector machine method are mainly used. Studies have been carried out on using these methods to forecast PV and wind power. In practice, high-precision ones should be selected from these methods considering their advantages and disadvantages.
With a high precision, the similar day and least squares support vector machine-based method can maintain the economy of the microgrid and meet the demands of control mode switching of the main power source for forecast of DG. This method involves two steps, selecting a similar day and forecasting the output of DG on the forecast day based on DG output on the similar day and weather data of the forecast day.
The similar day can be selected by level of correlation. The weather information released by the meteorological bureau includes the weather type, temperature, humidity, and wind strength. The days that have the same weather type (sunny, rainy, or cloudy) as the forecast day can be preliminarily selected as the similar day. The factors affecting PV power output are mainly irradiance and temperature, and those affecting wind power output mainly wind strength. The resemblance with the forecast day is then calculated from the most recent historical day, and the day most similar to the forecast day is used as the similar day. Finally, based on DG output on the similar day and the weather information of the forecast day, the DG output of the forecast day is calculated.
In the forecast of super-short-term DG output, after the data of DG output on the similar day is obtained, the DG output for the next hour can be forecast based on the weighted real-time weather data (irradiance, temperature, and wind strength) of the present hour.

6.2.2. Load forecast

Load forecast is to forecast future loads for analysis of demands, so that the operators can learn in time about the operation status of the system in the future. It is a major basis for the forecast of future operation of electrical power systems. Load forecast plays an important role in control, operation, and planning of the microgrid. Therefore, improving the forecast precision can contribute to a higher security and a better economy.
Load forecast methods used today are classified into traditional methods and modern methods based on their time of appearance. Traditional methods mainly include regression analysis method and sequential method, while modern methods mainly include expert system theory, neural network theory, wavelet analysis, gray system theory, fuzzy theory, and combinational method.

6.2.3. Frequency response characteristics of DG and loads

6.2.3.1. Response speed of DG

The DGs in a microgrid can be grouped as follows based on their frequency response capability and time:
1. PV and wind: Their output is affected by the weather, but not affected by the change in the system and hence, they can be deemed as constant-power sources.
2. Microturbines and fuel cells: Their response time ranges from 10 s to 30 s. In the case of a significant power deficiency in the microgrid system and the system has strict requirements on frequency. Instantly after islanding, microturbines and fuel cells cannot respond quickly enough to increase the output, and therefore, they are ignored in maintaining power balance at the instant of islanding.
3. The response time of ESs is generally 20 ms or even shorter. Therefore, it can be deemed that they can fill up the power deficiency with their maximum capacity in no time. The maximum capacity of ESs is roughly equal to the increment of power output that all DG sources can contribute to at the instant of islanding.

6.2.3.2. Frequency response characteristics of loads

The relationship between the active power of loads in an electrical power system and frequency of the system varies with the type of loads, as detailed next:
1. Loads whose active power does not vary with the frequency, such as lamp, electric furnace, and rectification loads;
2. Loads whose active power is in direct proportion to the frequency, such as ball mill, winch, compressor, and cutting machine;
3. Loads whose active power is in direct proportion to the square of frequency, such as the eddy current loss in the transformer core and feeder loss in the grid;
4. Loads whose active power is in direct proportion to the cube of frequency, such as ventilation fan and circulating pump with a small static head;
5. Loads whose active power is in direct proportion to the high degree of frequency, such as the feedwater pump with a large static head.
Without considering voltage fluctuation in the system, the relationship between system frequency f and the active power of load PL can be expressed as

PL=PLN(a0+a1f*+a2f*2++aif*i++anf*n)

image(6.1)
where f*=f/fNimage, N refers to the rated condition, * is the per unit value, PLN is the active power of load at the rated frequency, and ai is the scaling factor.
In the simplified system frequency response model, loads whose active power is in direct proportion to a high degree of frequency are not taken into account. Converting Eq. (6.1) to the differential equation of frequency, the frequency response factor of loads can be derived:

KL*=a1*=PL*f*

image(6.2)
Let ∆P be the power surplus and ∆f be frequency increment, then

PL*=PPL=PPLif*=ffN=f(1)f(0)f(0)

image(6.3)
where f (0) refers to the present frequency and f (1) the target frequency. In the case of power deficiency Pqe due to sudden change of the power output (e.g., tripping of generators) (if Pqe < 0, it indicates that more generators are switched in and power surplus occurs), part of loads will be shed to regulate the frequency, and then

KL*=PL*f*=PqePjhPLΣPjhf(1)f(0)f(0)

image(6.4)
where Pjh refers to the active power of loads to be shed. To reach the target frequency f (1) by load shedding, Pjh shall be

Pjh=PqeKL*(f(1)f(0))(PLPqe)f(0)KL*(f(1)f(0))

image(6.5)
In the case of power surplus Pyy due to sudden change of loads (e.g., load shedding) (if Pyy < 0, it indicates that the number of loads rises and power deficiency occurs), some generators will be tripped and then

KL*=PyyPqjPLPyyf(1)f(0)f(0)

image(6.6)
According to Eq. (6.6), to reach the target frequency f(1), the active power of generators to be tripped Pqj is

Pqj=PyyKL*(f(1)f(0))f(0)(PLPyy)

image(6.7)

6.2.4. Power balance

In grid-connected operation, generation and consumption in the microgrid are normally not limited, and only when needed, the macrogrid sends generation or consumption orders to the microgrid to control power exchange between the two. Specifically, in grid-connected operation, the macrogrid, based on economic analysis, sends power exchange setting values to the microgrid to maintain optimized operation. Based on the setting values, the energy management system of the microgrid will exercise control over the output of DG sources and charge and discharge of ESs, so that the microgrid operates on the specified power exchange rate in a secure and economic manner. In determining the output of various DG sources based on the power exchange setting values, the characteristics of the DG sources and control response characteristics should be considered for the energy management system.

6.2.4.1. Power balance in grid-connected operation

When the microgrid is grid-connected, the grid provides rigid voltage and frequency support, and normally exercises no special control over the microgrid.
In some cases, the grid specifies the amount of power exchange between the grid and microgrid, thus necessitating monitoring of power flow through the PCC.
When the actual power exchange deviates significantly from the setting value given by the grid, the microgrid control center (MGCC) needs to disconnect some loads or generators from the microgrid, or reconnect the loads or generators previously rejected to the microgrid, to minimize the deviation. The deviation of actual power exchange from the setting value is calculated as follows:

P(t)=PPCC(t)Pplan(t)

image(6.8)
where Pplan(t)image means the active power exchange setting value sent from the grid to the microgrid at the time of t, and PPCC(t)image the active power flowing through the PCC at the time of t.
If ∆P(t) > ɛ, there is a power deficiency in the microgrid, and the MGCC needs to reconnect the generators previously tripped to the microgrid, or disconnect some less important loads from the microgrid; if ∆P(t) < –ɛ, there is a power surplus in the microgrid, the MGCC needs to reconnect the loads previously shed to the microgrid, or trip some DG sources that produce electricity at a higher cost.

6.2.4.2. Power balance during transition from grid-connected mode to islanded mode

At the instant of transition from grid-connected mode to islanded mode, the power flowing through the PCC is suddenly cut. If this power flows to the microgrid before the transition, a power deficiency of such an amount will occur in the microgrid after transiting to islanded mode; otherwise, a power surplus of such an amount will occur in the microgrid after the transition. The microgrid usually suffers a significant deficiency due to the sudden loss of power from the grid.
If, at the very beginning of islanded operation, emergency control measures are not taken, the microgrid will experience a dramatic frequency decline, causing protective outage of some DG sources, followed by a greater deficiency and further frequency decline, then protective tripping of other DG sources, and finally collapse of the microgrid. As such, to keep the microgrid in islanded operation for a long time, it is necessary to take control measures at the instant of the microgrid being separated from the grid to maintain power balance.
In the event of power deficiency at the very beginning of islanded operation, it is necessary to immediately shed all or some less important loads (or even some important loads) and increase the output of ES; in the event of power surplus, it is necessary to immediately reduce the output of ES or even trip some of the DG sources. This will restore the microgrid to power balance quickly.
The instant deficiency (or surplus, as the case may be) in the microgrid is equal to the power flowing through the PCC before the separation.

Pqe=PPCC

image(6.9)
PPCC is expressed as a positive number if the power flows from the grid to the microgrid and vice versa. A Pqe greater than 0 indicates power deficiency in the microgrid at the instant of separation; and a Pqe smaller than 0 indicates power surplus.
As the ESs are intended to provide uninterrupted power supply to important loads for a certain period in islanded mode, such principles for power balance at the instant of separation apply that less important loads are shed first under the assumption that the output of all ESs is 0, then the output of the ESs is adjusted, and finally some important loads are shed if necessary.

6.2.4.3. Power balance in islanded operation

The microgrid is capable of operating in both grid-connected mode and islanded mode. When the microgrid disconnects from the grid after a fault, by adjusting the output of DG sources, output of ESs, and loads to achieve power balance and control, the microgrid can maintain stable operation. This ensures uninterrupted supply to important loads with DGs, thereby contributing to a higher efficiency of DG sources and supply reliability.
During islanded operation, the output of DG sources in the microgrid may vary with the environment (such as the irradiance, wind strength, and weather condition), leading to significant fluctuations of voltage and frequency. Therefore, it is necessary to monitor the voltage and frequency of the microgrid in real time, so that measures can be taken in time to deal with sudden change of sources and load that may impair the security and stability of the microgrid.
Supposing that the power deficiency at a moment during islanded operation is Pqe, then PL=Pqe/PLΣimage. It can be inferred from Eq. (6.2) that

Pqe=f(0)f(1)f(0)×KL*PL

image(6.10)
If, in islanded operation, the frequency at a moment f(1) is lower than fmin, there occurs a power deficiency on the microgrid, requiring the MGCC to reconnect the generators previously tripped, or shed part of the less important loads. While if f(1) is higher than fmax, there appears a remarkable power surplus on the microgrid, requiring the MGCC to reconnect loads previously shed or trip some DG sources.
1. Load-shedding control in the case of power deficiency
In the case of power deficiency (Pqe > 0), the following control strategies apply:
a. Calculate the current active output PS∑ and maximum active output PSM of ESs.

PSΣ=PSiPSM=PSmaxi

image(6.11)
where PSi is the active output of the ES i, which is positive during discharge and negative during charge.
b. If Pqe + PS∑ ≤ 0, it indicates the ES is being charged, and if the charging power is greater than the power deficiency, reduce the charging power until PSΣ'=PSΣ+Pqeimage, and stop the control. Otherwise, set the active output of the ES to 0 and recalculate the power deficiency P'image.
Pqe'=Pqe+PSPS=0
image(6.12)
According to Eq. (6.5), the allowable forward and reverse deviations of power deficiency can be calculated based on the maximum frequency fmax and minimum frequency fmin:

Pqe+=KL*(fmaxf(0))(P1Pqe)f(0)KL*(fmaxf(0))Pqe=KL*(f(0)fmin)(P1Pqe)f(0)+KL*(f(0)fmin)

image(6.13)
c. Determine the amount of less important loads to be shed

Pjhmin(1)=PqePqePjhmax(1)=Pqe+Pqe+

image(6.14)
d. Shed less important loads. Shed loads in an ascending order of importance. For loads of the same importance, shed them in a descending order of power. If PLi (power of a load) > Pjhmax(1)image, do not shed this load and proceed to check the next one; if PLi < Pjhmin(1)image, shed this load and proceed to check the next one. If Pjhmin(1)PLiPjhmax(1)image, shed this load and stop checking other loads. After shedding the load i, recalculate the power deficiency based on Eq. (6.15), and the amount of less important loads needing to be shed based on Eq. (6.14), and then proceed to check the next load.

Pqe'=PqePLqci

image(6.15)
where PLqc–i means the active power of loads that are shed.
e. After shedding less important loads as appropriate, if –PSMPqePSM, adjust the output of ESs to provide the remaining power deficiency until PS∑ = Pqe, and then stop the control. Otherwise, calculate the amount of important loads needing to be shed, that is

Pjhmin(2)=PqePSMPjhmax(2)=Pqe+PSM

image(6.16)
f. Shed important loads in a descending order of power. If PLi (power of a load) >Pjhmax(2)image, do not shed the load and proceed to check the next one; if PLi < Pjhmin(2)image, shed the load and proceed to check the next one; if Pjhmin(2)PLiPjhmax(2)image, shed the load and stop checking other loads. After shedding the load i, recalculate the power deficiency based on Eq. (6.15) and the amount of important loads needing to be shed based on Eq. (6.16), and then proceed to check the next load.
g. Adjust the output of ESs to provide the remaining power deficiency after appropriate load shedding until PS∑ = Pqe.
2. Generator tripping control in the case of power surplus
In the case of power surplus (Pyy > 0), it is necessary to trip some generators, and the control strategies are similar to those in the case of power deficiency:
a. Calculate the current and maximum active output of the ESs based on Eq. (6.11).
b. If –PSMPyyPS∑PSM, adjust the output of the ESs to absorb the power surplus after a proper number of generators are tripped until PSΣ'=PyyPSΣimage, and then stop the control. Otherwise, proceed to the next step.
c. Calculate the allowable forward and reverse deviations of power surplus based on the allowable upper and lower limit of frequency:

Pyy+=KL*(f(0)fmin)f(0)(PL0Pyy)Pyy=KL*(fmaxf(0))f(0)(PL0Pyy)

image(6.17)
d. If the ES is being discharged (PS∑ > 0), set the discharge power to 0 and recalculate the power surplus:

Pyy=PyyPSPS=0

image(6.18)
e. Calculate the amount of DG sources needing to be tripped:

Pqjmin=PyyPSMPSPyyPqjmax=Pyy+PSMPS+Pyy+

image(6.19)
f. Trip the generators in a descending order of power. If PGi (power of a source) > Pqj–max, do not trip the source and proceed to check the next one; if PGi < Pqj–min, trip the source and proceed to check the next one; if Pqj–minPGiPqj–max, trip the source and stop checking other sources. After tripping the source i, recalculate the power surplus based on Eq. (6.20) and the amount of sources needing to be tripped based on Eq. (6.19), and proceed to check the next source.

Pyy=PyyPGqci

image(6.20)
where PGqc–i means the active power of DG that is tripped.
g. Adjust the output of ESs to absorb the remaining power surplus after a proper number of generators are tripped until PS∑ = –Pyy.

6.2.3.4. Power balance

After the microgrid is reconnected to the grid, the DGs switch to P/Q control, and their power output relies on the dispatch plan of the distribution network. The MGCC needs to gradually put the loads or generators that were automatically disconnected from the microgrid into operation to maintain security and stability of the microgrid.

6.3. Optimized control

Renewable sources, such as wind power and PV power, are connected to the grid via inverters. Normally, the inverters are controlled to follow the frequency and voltage of the system, and keep the maximum output of DG sources rather than adjust their own output when the system frequency or voltage is excessively high or low. To maintain stability, there must be a master power source that can automatically change its output following the change of power frequency and voltage. Either a rotating generator, such as a diesel generator or pumped storage unit, or a large-capacity ES can serve as a master power source. Renewable sources shall be used as practical as possible. When renewable power is insufficient, the master power source can be used.
The economic operation control of the microgrid aims to maximize energy efficiency, minimize operation costs, and ensure the best economy of the microgrid by making best use of renewable energy while ensuring grid stability. Various optimization measures have been developed in view of the characteristics of various sources.

6.3.1. Optimized control of PV power

Optimized control of PV power is considered from the characteristics of PV sources and characteristics of power generation equipment.
1. Control of maximum output: PV power is a type of renewable energy that is relatively stable and has the highest priority among all renewable sources. PV units usually work at the maximum power output, except when output exceeds the demand, and the ESs have been fully charged.
2. Control of inverter group: PV inverters have the best efficiency and power quality when working at 30–70% of their rating. The efficiency of a PV inverter is dependent on the input power. When the input power is much less than the rated power (e.g., less than 20% of the rating), the efficiency will drop significantly. Furthermore, the total harmonic distortion (THD) in the output current of a PV inverter decreases with the increase of the input power. When inverters are lightly loaded, the THD will increase significantly. Specifically, the THD will exceed 5% when the input power is less than 20% of the rated power, and may even exceed 20% in a less-than 10% case. However, the efficiency will also decline when the input power is more than 80% of the rated power. As such, the PV inverter group control is used in the microgrid to improve the overall efficiency of the PV system.
The PV inverter group control divides PV arrays into groups, and distributes the current on the DC side of the PV system to more inverters or a centralized inverter through a transfer switch. The control strategy is as follows: in the morning, depending on sunlight availability, one inverter is started first, and when the inverter is almost fully loaded, another one is started. Then, in this way, other inverters are started one by one. In the evening, inverters are shut off one by one according to the output of solar panels. This kind of control requires prediction of sunlight change and rain and detection of the input power throughout the day. When the input power is too low, the transfer switch on the DC side is controlled to collect all DC current to one inverter, so that the efficiency of inverters will not decrease significantly due to the decrease of the input power.

6.3.2. Optimized control of wind power

Owing to the uncertainty of wind resources and operation characteristics of wind turbine generators, the output of wind turbine generators fluctuates remarkably, which often causes voltage deviation, fluctuation, and flicker. Thus, the primary concern for wind power control of a microgrid is to eliminate the influence on the stability due to the output fluctuation.
The following solutions are proposed to solve this problem:
1. The MGCC obtains the global measurement data of the system. When significant frequency or voltage fluctuation is detected, the controller determines and issues the control scheme immediately to compensate for the fluctuation to maintain constant power output.
2. Where the power of ESs is limited (the SOC is too small or too large, for instance), the energy management system of the microgrid may send power output orders to wind power inverters to temporarily control the output of wind power. As long as the designated output power varies slowly, the wind turbine generators can maintain a constant output. This limits the maximum output of wind turbine generators. After the generators resume a constant output, their output is increased to the maximum by control means to fully utilize wind resources. When renewable output exceeds the demand, for stability concerns, wind turbine generators are taken out of service first to maintain maximum PV output.

6.3.3. Optimized control of various types of ESs

ESs play an important role in keeping the stability of a microgrid. Usually, depending on the actual demand, diversified ESs are provided. An optimized dispatch system has to develop different control strategies for different types of ESs according to their respective characteristics.
1. Battery ES system (energy type): This type of ES is featured with a small loss, a long storage period, but a low response speed and a short life cycle, and is used only for storage of a large amount of renewable energy and as backup power sources for loads.
2. Flywheel, super-capacitor, and SMES (power type): This type of ES is featured with a high response speed and output power, but a large self-loss in the storage process, and is unsuitable for long-time storage. Therefore, they are mostly used in such circumstances as emergency power deficiency, mode transfer, and system disturbance.
Optimized control of multiple types of ESs can maintain smooth power output and stable voltage and provide backup in emergency. Whenever using ESs, their SOC should be watched all the time. If the SOC is excessively low or high, no power output orders should be given to them to prevent over-charge or over-discharge.

6.3.4. Optimized dispatch strategies

6.3.4.1. Power exchange management in grid-connected mode

In grid-connected operation, generation and consumption in the microgrid are normally not limited, except that the macrogrid, when necessary, sends specific generation or consumption orders to the microgrid through power exchange control. That is, in grid-connected operation, the macrogrid sends power exchange settings to the microgrid for best economy according to analysis results. According to the setting, the energy management system of the microgrid controls the output of DG sources and charge or discharge of ESs to maintain power exchange as instructed while ensuring economic and secure operation of the microgrid. In determining the output of each DG source, the energy management system should take the characteristics and control response characteristics of various DG sources into account.

6.3.4.2. Energy balance control in islanded operation

When the microgrid switches to islanded operation following a fault on the macrogrid, it should be able to maintain stability by energy balance control. In islanded mode, energy balance control, by adjusting the output of DG sources, energy release of ESs, and power consumption of loads, can maintain stability of the microgrid and ensure continuous power supply to important loads while fully utilizing renewable power, thereby improving efficiency of DG sources and supply reliability.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.117.99.152