Nokhum Markushevich and Aleksandr Berman
Abstract. The impact of customer-side voltage on load and energy consumption gained attention in the 1970s and 1980s [1]-[9]. The interest in and spread of Optimal Volt/Var Control (OVVC) as a near-real-time application of Distribution Management Systems (DMS) primarily began in the 1990s [10]-[20]. They grew exponentially with advances in Modern Distribution Systems (MDS). The current state of development on the OVVC can be assessed based on many review publications, e.g., [21]-[25]. Such high attention to this application requires adequate understanding of the landscape in which the OVVC operates.
Recent advances in communication technology, analytical capabilities, and the integration of smart devices into electric power distribution systems present opportunities and necessitate a more comprehensive management approach for active distribution systems, based on a holistic view of the entire power system. OVVC influences operations across generation, transmission, distribution, and customer domains. Some aspects of such an approach were presented in [26]-[31].
This paper presents a comprehensive overview of the real-world conditions for OVVC, defining the scope of operational areas and presenting a conceptual framework that addresses modular objective functions and multiple constraints.
Keywords: Distribution Management Systems (DMS), Optimal Voltage and VAR Control (OVVC), Distributed Generation, Voltage Quality, Smart Distribution Systems, Active Distribution Systems, Modern Distribution Systems.
1. Introduction.
Optimal Volt/Var Control (OVVC) in active distribution systems requires the near-real-time coordination of voltage and reactive power, as well as sometimes active power, across multiple controllable devices. This coordination serves several purposes, including:
• Maintaining voltage quality
• Energy conservation and peak demand reduction
• Supporting transmission operations
• Enabling preventive and self-healing adjustments
Comprehensive OVVC applications utilize multi-mode objective functions, combining several operational goals with distinct constraints. The transition between optimization states may be automatically triggered by power system conditions or DMS signals, or manually initiated by a Distribution System Operator (DSO). OVVC objectives vary based on the state of the power system, including:
1. Normal: System parameters remain within acceptable limits; supply meets demand.
2. Alert: A potential risk of instability (e.g., equipment outage or insufficient contingency reserve).
3. Emergency: Voltage violations, overloads, or generation deficits requiring immediate corrective actions.
4. In Extremis (Blackout): Full or partial system de-energization, necessitating rapid recovery measures.
5. Restorative: Efforts are underway to restore full operation after a disturbance
OVVC may have different objectives even within the same state of the power system, and the OVVC objectives may vary for other portions of the same distribution system.
This paper is organized as follows: First, we will briefly discuss the objectives of the OVVC in the distribution system under different states of the power system, from the perspective of a holistic approach to power system operations. Second, we will list and characterize the state variables that, in our opinion, should be considered by OVVC in modern distribution systems. Third, we will address the controllable variables and some of their specifics. Fourth, we will list and characterize the dependent variables of the OVVC, and then address the conditions of uncertainty for OVVC.
2. OPERATION OF OVVC IN DIFFERENT STATES OF THE POWER SYSTEM
2.1 OVVC IN THE NORMAL STATE OF THE POWER SYSTEM
In the normal state, the primary goal of Optimal Volt/Var Control (OVVC) is to maintain customer terminal voltages within acceptable limits. Additional objectives may be pursued, even when voltages remain within standard levels, including:
:
• Energy conservation: Some utilities aim to keep voltage close to the minimum standard limit, assuming that lower voltage always conserves energy. However, this is not always the case. See section 3.2.1 below.
• Peak load reduction: Achieved under normal conditions to provide economic benefits to both the power system and customers.
.
• Loss reduction: This objective of OVVC cannot be considered independent of load change due to voltage change. The results of such optimization may differ in terms of both the absolute value of loss reduction (in kWh) and the relative loss reduction (as a percentage of load).
• Reactive power support for transmission: OVVC may maximize or minimize reactive power drawn from or supplied to transmission systems [31].
• Maintaining specific power factor levels: Ensuring certain buses operate within prescribed power factor limits.
• Premium power service for specific customers: Tailoring voltage quality with tighter tolerances, reduced voltage imbalance, or minimized fluctuations [32].
• Mitigating voltage fluctuations due to Distributed Energy Resources (DERs): Countering intermittent variations from renewables [33]-[35].
These objectives must be executed in near-real-time, meaning control actions should remain relevant throughout the interval between consecutive control adjustments. Certain time-constrained devices, such as ULTC transformers and capacitor banks, require strategic activation schedules to optimize voltage conditions while minimizing unnecessary operations.
For instance, the voltage controller of the ULTC has a time delay, and the ULTC should not operate, say, more than 24 times daily. This does not mean that it should operate by equal time intervals. During the uphill change of the load, whether approaching the load peak or going downhill, the ULTC should operate more frequently. Similar constraints apply to the capacitor banks, although with a different preferred number of operations and time intervals between consecutive switching operations. The controls of smart inverters for DERs and/or microgrid control systems may be continuous or discrete, depending on the type of control: remote controls of voltage and/or VAR settings, the mode of operation, or autonomous control based on local settings.
The appropriate optimization state of the system must be chosen to achieve the best results over the most prolonged time interval between consecutive operations of the time-constrained devices. This means that the state of the power system to which the optimization will be applied must be predicted.
2.2 OVVC IN ALLERT STATE OF POWER SYSTEM
In the alert state, OVVC must adapt to the heightened risks by incorporating additional objectives:
• Load reduction integrated with demand response:
o Helps mitigate potential overloads and reduces strain on system components [36]-[37].
• Enhanced reactive power support for transmission:
o Transmission contingency analysis may forecast voltage drops leading to load collapse in parts of the distribution system or disruptions for sensitive customers (e.g., violating the ITIC curve) used in power quality assessments.
o OVVC should proactively change voltage limits in vulnerable areas to counteract these risks. In that case, OVVC should aim for higher voltage limits in these areas and provide greater reactive power support. However, these actions will result in increased loads in these parts of the distribution system, which may lead to even lower contingency voltages. To choose the appropriate OVVC solutions, the distribution and transmission control systems should be coordinated [30].
2.3 OVVC IN EMERGENCY STATE OF POWER SYSTEM
Under these conditions, load-shedding schemes are deployed, including:
• Under-Voltage Load Shedding (UVLS): Automatically disconnects loads to prevent system collapse.
• DER & Microgrid Ride-Through Operations: Distributed Energy Resources (DERs) and microgrids adjust their operating modes to mitigate disturbances.
The rapidly changing nature of an emergency event often renders general OVVC operations too slow to respond to all fluctuations effectively. As a result:
• Certain OVVC functions may need to be suspended during an emergency.
• If OVVC remains operational, it must adapt by prioritizing stability and fast-response mechanisms over standard optimization objectives.
• Pre-programmed responses and modified control strategies can override regular volt/var adjustments to align with grid emergency protocols.
2.4 OVVC IN IN EXTREMIS AND RESTORATIVE STATES OF POWER SYSTEM
During a blackout (In Extremis) or system restoration phase, the focus of OVVC shifts from optimization to rapid recovery and stability. The key objectives in these states include:
1. Managing Voltage Profiles & Reactive Power:
• As sections of the grid are systematically re-energized, OVVC ensures voltage levels remain within safe operating limits to prevent over-voltage or under-voltage conditions.
2. Gradual Re-Coordination of Voltage Regulation Devices:
• Each restored section must be assessed independently before integrating with previously re-energized areas.
• OVVC must re-coordinate voltage regulators, capacitor banks, smart inverters, and microgrid controllers to maintain balanced restoration across the network.
OVVC should operate in a highly adaptive mode, dynamically adjusting its functions as more sections of the system come online.
.
3. State Variables of OVVC
3.1 DISTRIBUTION DOMAIN STATE PARAMETERS
• Network Parameters:
o Substation configurations, transformer connectivity, and capacitor settings
o Voltage regulators and their operational constraints
o Real-time state of voltage controllers and associated devices
o Constraints on the operations of substation devices
These state variables are typically known to the DMS and EMS.
• Primary distribution network topology and parameters:
o Changes in feeder configurations due to maintenance, service restoration, or reconfiguration. Typically, well-documented by Distribution Management Systems (DMS)
• Primary distribution network parameters (line characteristics, step-up/down transformers, voltage regulators, feeder capacitors, and their controllers’ parameters, availability for remote control and constraints, etc. Distribution solid-state transformers (SSTs), their parameters, availability for remote control, and constraints.
These parameters are well-known by DMS models from the feeding Transmission-to-distribution substation to the distribution transformers’ kVA. The impedances and tap positions of the distribution transformers are typically not known to the DMS.
• Secondary distribution circuits’ topology and parameters
o Typically, not known to the DMS. However, some improvements to the estimates of the secondary circuit equivalent can be achieved by using an analytical analysis of the AMI data and near-real-time distribution power flow results [38]. Data collected from IoT-enabled smart meters and advanced sensors can supplement the assessment of the secondary distribution network
• Advanced Sensors: Deployed on transmission lines, transformers, and distribution 38feeders to monitor voltage, current, temperature, and power quality.
• Power Quality and Security state variables
o Voltage tolerances at customer terminals under normal and emergency conditions
These tolerances may be different for different groups of customers, such as primary service customers with or without voltage regulators, and microgrids
The voltage tolerances may depend on other state variables. They should be determined before or during the OVVC run; for example, the low voltage limit should be high enough. The high voltage limit should be low enough not to allow the expected voltage dips/swells to violate the ITIC curve. The emergency low voltage limits should be high enough to prevent load collapse during a probable power system contingency and its consequences in the distribution grid.
o Emergency loading limits of circuit elements
o Load collapse voltage levels
o Probable voltage dips/swells during a transient event
o Voltage variations due to DERs' intermittency
This variable requires certain constraints on other variables, e.g., reactive power of DERs should be below its maximum capability to allow room for reaction to the voltage variations [33]-[35].
• Economic values of the distribution system, such as cost of power losses, cost of replacement/upgrade of circuit elements (this is a benefit to the distribution system in case of load reduction and deferral of upgrade due to OVVC, and is a cost in the case of load increase).
• Distributed Energy Resources and Microgrids
o DERs and Microgrids comprise many state variables and may belong to both the distribution and the customers’ domains. It must be noted that in many cases, the information representing the DERs and Microgrids may be available only at the PCCs. For individual prosumers, it may be the Smart Meter and/or the Smart Inverter, and for a Microgrid, its control system. However, in other cases, the utility may have access to some information behind the PCC. The points of available information must be known by the OVVC application to predict and control operations accurately.
o The state variables from DERs and Microgrids include, but are not limited to, the following: The kinds and placement of DERs, their category and type according to IEEE1547-2018, their parameters, controllers and their parameters, available and current modes of operation and settings, ride-through settings and ranges, availability for remote control, and constraints. The state variables at the PCCs of the advanced microgrid will be presented below in the section for the customers’ domain.
The OVVC should consider different types of DERs and their various control options, as well as the availability of these controls. This includes:
• Solar Photovoltaics (PV)
o Residential and commercial rooftop PV systems. They are typically connected to local distribution networks.
o Small-Scale Solar Farms: These installations are often situated on brownfields or other local sites, interfacing directly with distribution feeders
• Battery Storage Systems
o Residential Batteries: These systems, often paired with solar PV, are used for home energy management and connected to local distribution lines.
o Community or Commercial Storage: Larger batteries can serve as grid resources at the distribution level, providing services such as peak shaving or backup power.
• Wind Turbines
o Small Wind Turbines: These are used in residential or community settings and are directly connected to the distribution portion of the grid.
o Medium-Scale Wind Farms: Medium-Scale Wind Farms are less common than large commercial wind farms, but medium-scale installations can serve local needs via distribution systems.
• Combined Heat and Power (CHP) Units. The additional state variables include parameters of supplemental heat sources, if any, as well as the incremental cost of heat replacement.
o Industrial and Commercial CHP: Often used in industrial parks, campuses, or commercial buildings, these units provide localized power and heat and are connected through distribution infrastructure.
• Microturbines
o Common applications include use in commercial buildings, small utilities, remote sites, industrial processes, and facilities with significant heating requirements. When installed as distributed energy resources (DERs), microturbines are often connected to local distribution systems and can be integrated with other DERs in microgrid configurations.
• Microgrids:
o Microgrids are comprised of various distributed resources and loads. They interface with distribution systems and can operate independently or in parallel with the grid. Microgrids present many additional state variables attributed to their PCC [39]-[41]. These may include, but are not limited to, the following:
Dispatchable active and reactive loads under normal and emergency conditions
Normal and emergency voltage tolerances
Load shedding schemes’ parameters
Load control and/or Demand Response schemes parameters, like the following
Amount of available real and reactive load reduction
Time needed for activation of the load reduction
Available duration of the load reduction
Cost of the load reduction
Steadiness of the load reduction during the time of engagement
Probability of execution
Contractual or operational agreements between the Microgrid owner/operator and EPS/DSO
• Fuel Cells:
o Typically installed in commercial buildings or industrial sites, fuel cells that generate electricity and heat locally connect to distribution networks.
• Biomass and Biogas Systems:
o These systems, especially those focused on small-scale generation and localized energy production, connect to distribution systems to serve nearby communities.
• EV-to-grid
o Distributed at significant charging places and multiple customers’ premises
• Demand Response Systems:
o Although not generating resources, these technologies primarily interact at the distribution system level, affecting local consumption patterns and sometimes interfacing with smart appliances.
• Virtual Power Plants
o VPPs are systems that aggregate multiple distributed energy resources (DERs) like solar panels, wind turbines, battery storage, and demand response units, to operate as a single power plant.
As we are discussing OVVC, we have added reactive power sources to the list of distributed energy resources (DERs).
• Capacitor Banks:
o These are widely used at distribution substations and along feeder lines
• STATCOMs and other Power Electronic D-Var devices:
o STATCOMs are preferred for fast, flexible voltage support around DERs and renewable interfaces,
o Can be installed near a PCC of primary service customers
• Static Var Compensators (SVCs):
o Typically installed in transmission systems, industrial installations, and larger commercial sites, as well as in critical points of distribution systems
• Synchronous Condensers (SC):
o These are typically installed in transmission systems and at voltage-critical points within the distribution network
• Power Electronic-based Controllable Series Compensators
3.2 CUSTOMER DOMAIN STATE PARAMETERS
3.2.1 CUSTOMER DEMAND AND CONSUMPTION.
Customer demand, energy consumption, and reactive power load are both state and dependent variables. They define the power flow in the entire power system and change in response to changes in voltage.
As state variables, accurate real-time knowledge of nodal loads is essential for the optimization function to estimate the distribution power system state effectively. Concurrently, these loads function as dependent variables due to their inherent voltage dependence.
The large number of nodal loads within a distribution system, coupled with their significant and dynamic variations, traditionally forces model-based OVVC algorithms to rely on simplifying assumptions. Consequently, distribution state estimation and resulting optimization outcomes often become approximate, diverging from actual system conditions and optimal states. For example, with thermostat-controlled loads, lower voltages can increase aggregate energy usage due to a higher load coincidence factor [42]. In such cases, the Conservation Voltage Reduction (CVR) and Load-to-Voltage (LTV) factors may exhibit negative correlations and differ significantly across customers based on thermostat-controlled equipment size, the duration of voltage changes, and ambient temperature. Loads that adapt to prolonged voltage reductions behave similarly, with LTV and CVR changing across distribution locations and times. Therefore, effective OVVC requires highly accurate, time- and location-specific LTV and CVR factors; otherwise, suboptimal or even counterproductive solutions may result.
The primary challenge in determining LTV and CVR factors is distinguishing "noise" in load data from actual voltage-dependent responses. Such noise often overshadows the effects of voltage. Legacy methodologies aggregate LTV and CVR factors at substation buses or feeder heads, which have been used by utilities [13, 18, 43]. This approach leverages auxiliary information correlated with noise, but high DER penetration complicates its execution because net power at substations and feeder heads becomes more variable [44]. Recent advancements in AMI, AI, IoT, cloud computing, digital twins, smart buildings and homes, smart appliances, and advanced analytics may essentially eliminate these limitations, providing the necessary auxiliary data for models to improve their estimates adaptively. Some ideas on these issues were presented in [45]. This information should be valid for the time interval of each step of optimization. A short-term predictive procedure is needed to choose the appropriate input values.
Another component of customer load relates to payment plans for energy, demand, power factor, and demand response, such as Time-of-Use, Real-time pricing, Load Control, and Demand Response contracts.
Regarding the Load Control and Demand Response, the estimates of the following state variables are needed for integrating the OVVC with load management, which can increase the benefits of OVVC in the alarm and contingency states of the power system [36], [37]:
• Amount of available real and reactive load reduction
• Time needed for activation of the load reduction
• Available duration of the load reduction
• Cost of the load reduction
• Steadiness of the load reduction during the time of engagement
• Probability of execution
• Contractual or operational agreements between the customer and EPS/DSO. These agreements encompass the systems and equipment involved (smart meter-based customer, smart home, smart appliances, and smart building), the amount of load reduction, duration, number of reductions per given time interval, and compensation for participation in the program.
3.2.2 ADVANCED CUSTOMERS’ MICROGRIDS
At the PCC of advanced microgrids, relevant state variables related to Volt/Var control during connected operation include [41] :
• Microgrid Volt/Var Control Mode: Current operating mode as managed by the μEMS (e.g., voltage regulation, power factor control, reactive power priority).
• Volt/Var Setpoints: Target setpoints for voltage, reactive power exchange (export/import), or power factor at the PCC.
• μEMS Settings/Parameters: Key operational settings within the microgrid EMS influencing Volt/Var management. For example, active droop settings or control thresholds related to reactive power dispatch.
• Power Factor at the PCC: Real-time indicator of power factor, as applicable
In some cases, the DMS and μEMS objectives for volt/var control may conflict, or may involve a cost for one party to meet the objective of another party.
State variables at the PCC relevant to coordinating emergency actions during bulk power system contingencies include:
• Microgrid Connection Status: islanded or grid-connected.
• Emergency Operating Mode: Active operating mode of the microgrid in response to a contingency (e.g., islanded, grid-supporting, load shedding).
• Ride-Through Settings: Specific configuration for voltage and frequency ride-through capabilities (e.g., voltage and frequency thresholds, duration settings).
• Load Shedding Status: Percentage of total microgrid load currently shed and minimum load shedding capacity.
• Emergency DER Dispatch Capacity: Incremental output changes enabled for DERs within the microgrid when acting as dispatchable units.
Other state variables include:
• Aggregated Load and Generation Voltage Dependencies: (LTV/CVR) for both real and reactive power, as observed at the PCC. These reflect the collective response of the microgrid's loads and DERs to voltage changes. These dependencies are changing due to the influence of numerous factors, such as the composition of different load categories, the load-generation balance in microgrids, the modes and settings of DER volt/var controls, and the operations of other sources of reactive power and energy storage in the microgrids, among others. It implies that the μEMS performs corresponding analyses of the microgrid's operations under different voltages at the PCC
• Microgrid Power Capability Curves: Real and reactive power capability curves, as observed at the PCC. The capability curves include these components:
o Dispatchable real and reactive power range available from DERs.
o Available demand response capacity, under specified notice, duration, and cost.
o LTV and CVR coefficients for dispatchable loads.
Load participation across microgrid management programs can overlap, where activating one may implicitly constrain the participation of others.
In the absence of accurate state estimates that reflect the time scale of advanced microgrids and environmental effects, the uncertainty of OVVC is increased, and its performance and reliability can be impacted.
3.2.3 OTHER CUSTOMERS' PARAMETERS
A large number of state variables characterizing customer properties have a significant impact on the outcome of OVVC. They include the following:
• Customer category
• Basic load/process type
• Contractual/operational agreements
• Demand response
• Microgrid
• Energy/demand pricing type
• Prosumer with battery storage
• Prosumer with/without smart inverter
• Smart building (including local control system)
• Smart house with/without smart appliances with smart sensors
• Voltage quality requirements
o Standard consumer (standard voltage quality requirements, or contractually broader voltage quality requirements)
o Primary service customer with/without local voltage control
Tighter/relaxed voltage tolerances
Tighter/relaxed voltage imbalance tolerances
Tighter/relaxed sags/swells tolerances
• With/without backup resource.
3.3 TRANSMISSION/GENERATION DOMAINS STATE PARAMETERS
The transmission/generation state variables for the distribution OVVC can be obtained from the power system EMS and/or from the transmission system operator (TSO) and can comprise, but not be limited to, the following:
• Transmission System Equivalent Impedance: The equivalent impedance of the transmission system as seen from the distribution substation.
• Real-Time Marginal Loss Factor: Incremental cost of real power losses on the transmission network due to reactive and active load changes.
• Contingency Reserve Status: Real and reactive power reserves available at the transmission level, along with costs associated with violating these limits.
• Transmission Congestion Metrics: Real-time measurements for overloads and transmission contingencies.
• Generation Dispatch Costs: Real-time incremental costs or production data.
• TSO/EMS Directives: Specific commands for load curtailment or reactive power support from the Transmission System Operator (TSO) or Energy Management System (EMS).
• Contractual/operational agreements between Transmission and Distribution domains.
3.4 EXTERNAL STATE PARAMETERS
OVVC is a real-time application that also utilizes short-term predictive models, meaning it must continuously adjust to expected external environmental and social factors that cause instantaneous variations in load demand, its composition, and the output of distributed energy resources (DERs). Many of the state parameters listed above depend on environmental conditions and social events
1. Real-Time Environmental Influences
• Ambient temperature - changes demand and its sensitivity to voltage
• Wind Speed – changes the DER output
• Solar Irradiation – changes demand and PV output
2. Social Events Driving Instantaneous Demand Surges
• Major Sporting Events (e.g., Super Bowl, Olympics)
• Concerts & Festivals
• Cultural & Holiday Energy Consumption Patterns
4. CONTROL VARIABLES
The control variables of OVVC are predominantly located in the Distribution domain and partially in the Customer domain. There are multiple control variables of different natures. They may differ considerably, even for the same type of device, due to various designs of the control system and/or different regulations and/or DSO requirements governing the control application (See e.g., IEEE1547-2018). OVVC should include the ability to recognize and utilize all types of relevant control capabilities.
4.1 CONTROLS AT THE T/D SUBSTATION
• ULTC Transformers /Voltage regulators (AVR) tap position
o Constrained by available regulation range; number of operations per given time
o Suspended/operational depending on state conditions
o Associated with the cost of operations
• ULTC/AVR voltage controller settings
o Voltage setpoints (the execution of this control has a degree of uncertainty due to the deadband and time delay [46].
o Voltage deadband (as a control variable, can be used to change the sensitivity of the voltage controller and the number of ULTC operations)
o Time delay (as a control variable, can be used to change the sensitivity of the voltage controller and the number of ULTC operations)
o Line-drop compensation control (LDC)
Modern systems incorporate flow direction detection and dynamic gain adjustment
o Suspended/operational depending on state conditions
This control is constrained by the available regulation range of the ULTC; the number of operations per given time, and is associated with the cost of ULTC operations. The availability of this central control must be known to the OVVC in near-real-time [46]
• Substation capacitors control
Constrained by an available number of capacitors, the number of operations per given time, the interval between consecutive operations, and the available reactive power output dependent on bus voltage.
Suspended/operational depending on state conditions
Associated with the cost of operations and active losses in the capacitors
The availability of this control is known to OVVC from SCADA
4.2 CONTROLS IN THE PRIMARY DISTRIBUTION NETWORK
• Direct feeder capacitors control
o Constrained by the available number of capacitors, the number of operations per given time, the interval between consecutive operations, and the available reactive power output dependent on bus voltage.
o Suspended/operational depending on state conditions
o Associated with the cost of operations and active losses in the capacitors
• Control of the settings of the capacitor controller (this control introduces a degree of uncertainty in execution)
o Voltage settings
o Current settings
o Reactive power settings
o Time of use settings
o Combination of some of the above
o Time delay
o Deadband
o Suspended/operational depending on state conditions
Constrained by available capacitors, the number of operations per given time, the interval between consecutive operations, and the available reactive power output, dependent on bus voltage.
Associated with the cost of operations and active losses in the capacitors
• Control of feeder AVRs (three-phase, single-phase)
o Similar to the controls of substation ULTC voltage controllers
• Control of DERs available for generating /absorbing reactive power
o Depending on the available communication and DER’s controller, the OVVC/DSO issues commands or requests to install particular settings of the controller or mode of operation, like (see IEEE1547-2018)
Maximum generation of reactive power
Maximum absorption of reactive power
Constant given power factor
Constant given voltage
Volt-Var mode of operation with corresponding settings
Volt-Watt mode of operation with corresponding settings
Watt-Var mode of operation with corresponding settings
Var-priority mode of operation with corresponding settings [47]-[49]
It is worth noting that the maximum current, which in some inverters is the same as the nominal current, limits the capacity of a DER/Inverter installation, not just the apparent power. This means that the available reactive power depends on the voltage and the real power output, and these dependencies are not piece-wise linear [50]-[51]. See Fig.1.
Figure 1. Available reactive power for the Volt/Var mode of a smart inverter with fixed active power
Figure 2 shows a sample diagram for the volt/var control mode of a smart inverter similar to that presented in IEEE1547-2018. As seen in this example, to respect the volt/var settings of this mode, the active power should be reduced under lower voltages. Actually, in this case, it is a var-priority mode.
Figure 2. Sample Diagram for Volt-Var control mode of smart inverter. Rated apparent power=1, maximum current=rated current
Smart inverters from different manufacturers significantly differ in their specifications regarding maximum current, voltage tolerances, apparent and active power limits under various temperatures, rated power factors, voltage, reactive power, and active power settings for different modes of operation, as well as limits on export active power. The various constraints of the smart inverters may significantly impact the expected effect of the inverters’ control.
• Control of Microgrids
o Microgrids may have their own decentralized energy management systems (EMS). Depending on the contractual/operational agreements between the Microgrid owner/operator, OVVC can control or request a particular mode of operation or settings similar to the controls of smart inverters, and, in addition, control of net active power, depending on the state of the power system. For instance, in alert or contingency states of the power system, OVVC can request a reduction in net active load or an increase in net generation [39]-[40]. This kind of control of microgrids is associated with uncertainty in the execution. The owner/operator of an advanced microgrid may have operational objectives that are acceptable to the EPS DMS/operator or may have conflicting objectives. In this case, the OVVC objective function should select one alternative or a compromise, and include the cost of lost benefits [40].
• Control of demand-side resources according to existing Demand Response conditions for each resource or a group of resources under the same conditions to increase the combined effect of load reduction by OVVC and Demand Response [36], [37].
• Control the Reactive power of Combined Heat and Power (CHP) Controllers
o If necessary, consider controlling the active power of CHP, considering the operational constraints and incremental cost of the change, including the cost of heat replacement, if any.
4.3 CONTROL OF VIRTUAL POWER PLANT (VPP)
o VPP aggregates multiple DERs and load management means to operate as a single power plant, optimizing dispatch and providing grid services. The efficiency of OVVC in controlling VPPs depends greatly on contractual and operational agreements that allow the VPP operator or utility to manage DER outputs and/or controllable loads. Each region’s regulatory environment might specify different permissions and operational boundaries for VPP control. Due to the geographical distribution of the VPP components, aggregated demands or requests to the VPP, even with indications of the required control in each geographical area, will result in a suboptimal solution and increased uncertainty if OVVC is used.
• Control of Demand Response Systems
o OVVC can use this control by the existing contractual agreements to enhance the efficiency of Volt/t/Var [37]
5. DEPENDENT VARIABLES OF OVVC
5.1 IN THE CUSTOMERS’ DOMAIN
The following are the dependent variables in the customer domain:
• Voltage quality metrics at the customer terminals and their relation to voltage quality constraints. They include voltage deviations from nominal voltage, voltage imbalance, voltage flickers, and voltage dips and swells. If “soft” limits on these variables are used, the “cost” of violations is also a dependent variable and is attributed to the “guilty” entity. The constraints of the acceptable voltage deviation at the customers’ PCC are not always the standard values. The permissible voltage tolerance for primary service customers may be tighter or looser than the standard tolerance, depending on whether local voltage conditioning devices are used at the point of common coupling (PCC). There may be contractual wider voltage tolerances for some customers who agree to them in exchange for compensation. These agreements may define the voltage ranges, specific times of day (e.g., peak load times), duration, frequency, and compensation mechanism for voltage adjustments.
• Changes in the active and reactive loads of the customer due to the voltage change by the OVVC and activated by OVVC demand management [36], [37]. These variables are associated with the incremental cost of energy to the customer due to load-to-voltage dependencies, as per the customer’s energy schedule, and the activation of demand response as per the agreement between the customer and EPS. For demand-charged customers, the cost associated with changing the power factor is also a dependent variable. In the case of EV control, a different “cost” may be associated, considering transportation-related changes.
• In smart buildings/homes, the changes in the states of innovative (smart) components/appliances that OVVC can control through IoT. These changes are associated with a composite cost determined by agreements with EPS for permitting such control.
• Microgrids' controllable DERs’ dependent variables include
o Voltages at the PCC that are subject to constraints depending on the microgrids’/DERs’ voltage regulation range, with corresponding “cost” of violations. In case of conflicting objectives between the microgrid/DER owners and EPS, the cost of a compromise is associated with the cost to the microgrid/DER and or to the EPS
o Changes in net active and reactive loads due to OVVC actions in different power system states and by the contractual agreements between the EPS and the microgrid. These variables are associated with the incremental cost of energy to the microgrid or distributed energy resource (DER) owner. In the case of using replacement energy, the cost is incurred by the entity providing the replacement energy.
5.2 IN THE DISTRIBUTION DOMAIN
The dependent variables in the distribution domain include the following:
• Change of loading circuit elements, which are associated with
o Power losses in distribution, and their incremental cost due to the OVVC (which ultimately becomes the cost to the customers)
o The adequacy and security of the power system, and consequently, the benefits of deferral of upgrades or the cost of additional construction in case of load increase.
• The cost of operations of controllable devices and the cost of losses in them, if any
• Cost of voltage quality violations, if it is the EPS's responsibility. This includes the cost of breaches of the agreed-upon voltage deviation tolerances, voltage imbalance limit (where applicable), ITIC curve violations, and flicker requirements violations.
• Cost of contractual obligations with customers and transmission domains related to OVVC, e.g., if there is a contract between the distribution and transmission domains of reactive power support, and the distribution and/or customers' systems experience an increase in cost due to meeting the transmission request, then these are the costs of the agreement, which presumably will be compensated by the transmission domain. However, if the contractual request is not provided, the cost of violating the contract is borne by the distribution.
5.3 IN THE TRANSMISSION DOMAIN
The dependent variables in the Transmission domain include
• Change of active and reactive loading of transmission elements due to OVVC in distribution, which are associated with
o Change of power losses in transmission and their incremental cost due to the OVVC
o Change of bus voltages and the cost of deviation from operationally safe levels
o The adequacy and security of the transmission system, and consequently, the benefits of deferral of upgrades or the cost of additional construction in case of load increase.
• Reactive power contingency reserve that is supported by reactive power resources in the transmission domain. If the reserve drops below a safe level, the system becomes vulnerable to insecurity. Reactive support from the distribution domain may reduce the reactive power deficit and eliminate the cost of insufficient reserve
• Cost of contractual obligations with distribution domain related to OVVC, e.g., if there is a contract between the distribution and transmission domains of reactive power support, and the distribution and/or customers' systems experience an increase in cost due to meeting the transmission request, then these are the costs of the agreement to the transmission domain. However, if the distribution does not fulfill the contractual request, then the cost of violating the contract is borne by the distribution, which is paid to the transmission.
5.4 IN THE GENERATION DOMAIN
The dependent variables in the Generation domain include incremental benefits of deferring new generation due to reduced loads in distribution or incremental capital expenses in cases of increased loads in distribution due to OVVC.
6. UNCERTAINTY OF OVVC
The voltage tolerances for volt/var optimization under normal conditions are narrow. Hence, the accuracy of voltage and VAR control is critical for preventing violations of required voltage quality. As with any other complex control process, volt/var optimization in active distribution networks operates under conditions of significant uncertainty [46]. Measures to reduce the degree of uncertainty should be considered when choosing the methodology and the properties of the control process to ensure that power quality remains within acceptable limits and to obtain the expected benefits.
The many sources of uncertainty for OVVC include, but are not limited to, the following:
• Direct measurements’ inaccuracy and latency
• Determination of voltage critical points
• Load//energy sensitivity to voltage
• Sensitivity of components of the objective function to controls
• The power system model
• The nodal load models
• The models of secondary equivalent
• The DER/microgrid models
• The control equipment properties, such as the uncertainty of the impact of LTCs due to deadband and time delay, the uncertainty of the effect of capacitors, and the uncertainty of the impact of DERs.
• The lack of information from microgrids, VPPs, and other composite customers.
• The contractual/operational agreements.
7. CONCLUSIONS
1 The modern landscape of Optimal Volt/Var Control (OVVC) extends far beyond the traditional scope of optimization. The technical, operational, regulatory, and economic conditions across the generation, transmission, distribution, and customer domains of the power system shape it.
2 To function effectively, OVVC must adapt dynamically to varying power system states—from normal operations to emergency response and grid restoration. The presence of distributed energy resources (DERs), microgrid architectures, and sensor-driven analytics adds layers of complexity that challenge conventional control models.
3 The optimization of volt/var may involve control of active power by a war-priority mode of smart inverters and/or by using demand response.
4 OVVC in distribution operates under conditions of uncertainty, which should be accounted for.
Acknowledgement. The authors used the assistance of AI Microsoft Copilot and Open GPT-4o in editing the text of the paper.
References:
1. Wheeler, P.L., Dickenson, A.H. and Gibbs, T. J., "The Effect of Voltage Reduction on Distribution System Loads, IEEE Conference Paper A-78-542-3, Presented at the Summer PES Meeting, Los Angeles, July 16-21, 1978.
2. Priess, R.F. and Warnock, V.J., "Impact of Voltage Reduction on Energy and Demand", IEEE Transactions on Power Apparatus and Systems, Vol. PAS-97,No. 5, Sept./Oct. 1978, pp. 1665-1671.
3. Erickson, J.C. and Gilligan, S.R., "The Effects of Voltage Reduction on Distribution Circuit Loads", IEEE Transactions on Power Apparatus and Systems, Vol. PAS-101, No. 7, July 1989, pp. 2014-2018.
4. Warnock, V.J. and Kirkpatrick, T.L., "Impact of Voltage Reduction on Energy and Demand", IEEE Transactions on Power Systems, vol. PWRS-1, No., May 1986, pp. 92-97.
5. Chen, M.S., Shoults, R. and Fitzer, J., "The Effects of Reduced Voltages on the Efficiency of Electric Loads", IEEE Transactions on Power Apparatus and Systems, Vol. PAS-101, No. 7, July 1982, pp. 2158-2166.
6. Nokhum S. Markushevich, Voltage Control and Electric Energy Conservation, , Moscow: Energoatomizdat, 1984, pp.1 100. Sofia: Technika, 1987, pp. 1 84.
7. Impact of Voltage Reduction on Energy and Demand, A Thesis Presented to the Faculty of the College of Engineering and Technology, Ohio University, Khalil Matar, November 1990
8. Daniel Kirshner, "Implementation of Conservation Voltage Reduction at Commonwealth Edison", IEEE Transactions of Power Systems, Vol. 5, No.4, pp. 1178-1182, November 1990.
9. B.W. Kennedy and R.H. Fletcher, "Conservation Voltage Reduction (CVR) at Snohomish County PUD", IEEE Transactions of Power Systems, Vol. 6, NO. 3, pp. 986-998, August
10. Nokhum S. Markushevich, Voltage and Var Control in Automated Distribution Systems, DA/DSM Conference, January 1993, Palm Springs, California
11. Cost-Benefit Study for Distribution Automation at B.C. Hydro, Nokhum S. Markushevich, Ivan C. Herejk, Ron E. Nielsen, Utility Consulting International, USA, Distribution 2000 Conference, November 1993, Australia
12. Nokhum S. Markushevich, I.C. Herejk, Ron Nielsen, Functional Requirements and Cost-Benefit Study for Distribution Automation at B.C. Hydro, IEEE Transactions on Power System. Vol. 9, No. 2. May 1994
13. Alf Dwyer, Ron Nielsen, Joerg Stangl, Nokhum S. Markushevich, Load to Voltage Dependency Tests at B.C. Hydro; IEEE/PES 1994 Summer Meeting, July 1994
14. Nokhum S. Markushevich, Martin Delson, Erik Benedictson, Alf Dwyer, Ivan Herejk, Carl Kober, Ron Nielsen, Joerg Stangl, Planning a Distribution Automation Pilot Project at B.C. Hydro, DA/DSM Conference, January 1994, Florida
15. Edward H.P. Chan, Nokhum S. Markushevich, Integration Of Distribution Automation into Power System Operation, DA/DSM Conference, January 1994, Florida
16. Nokhum S. Markushevich, Ron E. Nielsen, Update on DA Pilot Project at B.C. Hydro, Fifth DA/DSM Conference January 1995, San Jose, California
17. I. Roytelman; B.K. Wee; R.L. Lugtu, Volt/var control algorithm for modern distribution management system, IEEE Transactions on Power Systems, Year: 1995 | Volume: 10, Issue: 3
18. Nokhum S. Markushevich, R.E. Nielsen, A.K. Nakamura, J.M. Hall, R.L. Nuelk, Impact of Automated Voltage/Var Control in Distribution on Power System Operations, DA/DSM Conference January 1996, Tampa, Florida.
19. Nokhum Markushevich, Alex Berman, Dan Nordell, Craig Halverson, The Impact of Simplification of the Distribution System Model on the Benefits of Voltage and Var Control, DA/DSM Distributech Conference, January 1997, San Diego, California.
20. I. Roytelman, B.K. Wee, R.L. Lugtu, T.M. Kulas, T. Brossart, Pilot project to estimate the centralized Volt/Var control effectiveness, Proceedings of the 20th International Conference on Power Industry Computer Applications, 1997
21. Design and Assessment of Volt/VAR Optimization Systems, Technical update, EPRI 1022004, 2011, Project Manager R. Uluski, Principal Investigators: R. Uluski, W. Sunderman, J. Green, J. Roark, and N. Markushevich
22. Tengku Juhana TENGKU HASHIM1, Azah MOHAMED2, Hussain SHAREEF3, A review on voltage control methods for active distribution networks, IEEE
23. H. Mataifa; S. Krishnamurthy; C. Kriger, Volt/VAR Optimization: A Survey of Classical and Heuristic Optimization Methods, IEEE
24. Hamed Ahmadi, Student Member, IEEE, José R. Mart´ ı, Fellow, IEEE, and Hermann W. Dommel, Life Fellow, A Framework for Volt-VAR Optimization in Distribution Systems
25. Sarah Allahmoradi1, Shahabodin Afrasiabi Xiaodong Liang, Junbo Zhao, and Mohammad Shahidehpour, Data-Driven Volt/VAR Optimization for Modern Distribution Networks: A Review, IEEE Access, Volume 12, 2024
26. Nokhum Markushevich, Cross-cutting Aspects of Smart Distribution Grid Applications, Presented at IEEE PES GM 2011, Detroit
27. Nokhum Markushecih, The Benefits and Challenges of the Integrated Volt/Var Optimization in the Smart Grid Environment, Presented at IEEE PES GM 2011, Detroit
28. Nokhum Markushevich and Aleksandr Berman, New Aspects of IVVO in Active Distribution Networks, 2012
29. Nokhum Markushevich, Information Support of Operations of Power Systems with Active Distribution Networks. Available: Information Support of Operations of Power Systems with Active Distribution Networks, EnergyCentral
30. Nokhum Markushevich, Transmission Bus Load Model for Smart Distribution and Transmission Grids. Available: Transmission Bus Load Model for Smart Distribution and Transmission Grids, EnergyCentral
31. Nokhum Markushevich, Reactive Power Support of Distribution and Transmission Systems by Active Distribution Networks. Available: Reactive Power Support of Distribution and Transmission Systems by Active Distribution Networks
32. W.T.J. Hudhorst, E.M.L. Smeets, J.A. Wolse, Primary Power Quality Contract and Labeling. Available: https://www.slideshare.net
33. Nokhum Markushevich, Mitigating Voltage Fluctuation Caused by Variability of Distributed Energy Resources. Available: Mitigating Voltage Fluctuation Caused by Variability of Distributed Energy Resources, EnergyCentral
34. Reigh Walling and Gao Zhi, Eliminating voltage variation due to distribution-connected renewable generation, DistribuTech 2010
35. Nokhum Markushevich, Mitigation of Voltage Fluctuations by Controlling Reactive Power of DER, EnergyCentral
36. N. Markushevich and A. Berman, Distribution Automation and Demand Response, DistribuTech2008, Tampa, FL, January 2008;
37. Nokhum Markushevich and Edward Chan, Integrated Voltage, Var Control and Demand Response in Distribution Systems, Nokhum Markushevich and Edward Chan, IEEE, March 2009, Seattle
38. Nokhum Markushevich and Wenpeng Luan, Achieving Greater VVO Benefits through AMI Implementation, Presented at IEEE PES GM 2011, Detroit
39. Nokhum Markushevich, What will the Microgrids and EPS Talk about? Parts 1 and 2. Available: What will the Microgrids and EPS talk about? Part 1: What will the Microgrids and EPS Talk about? Part 2., EnergyCentral
40. Nokhum Markushevich, Information Exchange between Advanced Microgrids and Electric Power Systems. Available: https://www.scribd.com/document/376567099/Information-Exchange-between-Advanced-Microgrids-and-Electric-Power-Systems
41. NIST/SGOC_UC_IA-1, Seven Interactive use cases on advanced microgrids in Active Distribution network. Available: http://smartgrid.epri.com/Repository/Repository.aspx/
42. Nokhum Markushevich, Analysis of electric heating load dependency on voltage. Available: Analysis of electric heating load dependency on voltage, EnergyCentral
43. Nokhum Markushevich, Aleksandr Berman, and Ron Nielsen, Methodologies for Assessment of Actual Field Results of Distribution Voltage and Var Optimization, presented at IEEE PES 2012 T and D
44. Nokhum Markushevich, Voltage Reduction Effect in Active Distribution Networks. Available: https://www.scribd.com/document/376755241/Voltage-Reduction-Effect-in-Ac...
45. Nokhum Markushevich, Some Considerations on IoT for Power System Operations. Available: Some Considerations on IoT for Power System Operations, Energycentral
46. Nokhum Markushevich. Uncertainty of Voltage Control in Active Distribution Networks. Available: https://www.scribd.com/document/376554187/Uncertainty-of-Voltage-Control-in-Active-Distribution-NetworksHeartbeats
47. Nokhum Markushevich, Var-priority Mode of DER Volt/var Control Function. Available: Var-priority Mode of DER Volt/var Control Function. | Energy Central, 2018
48. Nokhum Markushevich, Comparison of the effectiveness of the var priority mode for reducing and increasing voltage. Available: Comparison of the efficacy of the var priority mode for reducing and increasing voltage, EnergyCentral
49. Regulating voltage: recommendations for smart inverter, GridLab Report, 2019. Available: GridLab_Regulating-Voltage-report.pdf
50. Nokhum Markushevich, Operations of Smart Inverters in Active Distribution Networks. Available: Operations of Smart Inverters in Active Distribution Networks, EnergyCentral
51. Nokhum Markushevich, Dispatchable Reactive Load in Active Distribution Networks. Available: https://www.scribd.com/document/376759501/Dispatchable-Reactive-Load-in-Active-Distribution-Networks