Var-priority Mode of DER Volt/var Control Function.

Var-priority   Mode of DER Volt/var Control Function.

Nokhum Markushevich

The “Volt / var control (VVC) with var priority” is suggested as one of the functions of the Active Distribution Networks [1]. The primary objective of this function is to support voltage quality and/or var requirements by providing more reactive power from the distributed energy resources (DER) at the expense of real power injections by the DERs. This article discuss the case, when the var priority mode is considered for mitigating under-voltage.

The available reactive power from an inverter depends on its rated power factor, actual kW, and voltage at the terminals of the inverter. For each fixed value of the DER real power, the available reactive power of the DER is presented in Table 1 for the voltage equal 1pu, and in Table 2 for the voltage equal 0.93pu. The rated real power is assumed 100%, and the reactive power is presented in the same scale.  

Table 1. Var capabilities of inverter (without kW reduction). Rated kW = 100%, DER Volt=1

In many cases the VVC function in var priority mode is applied when the voltage is below the acceptable level at the DER terminals. In these cases, the DER’s var capability is significantly smaller.

Table 2 presents the case when the voltage at the DER terminals is 0.93 pu. As seen in this table, the available kvars are less than in Table 1.

Table 2. Var capabilities of inverter (without kW reduction). Rated kW = 100%, DER Volt=.93

 The same reduction of kW, but from different initial kW of the DER, does not provide the same additionally available kvars, as seen in Table 3.

Table 3 presents the additional kvars that become available after the kW are reduced by 10%, For instance, when the PF=0.90 and the voltage is 0.93pu, reduction of the kWs by 10% from the initial 100% provide 24.7 % of additional kvars. Reduction of the kWs by 10% from the initial 90% of kWs provides 14.6 % of additional kvars.  As seen in the table, the lower is the initial DER kW, the smaller are the additionally available kvars. 

Table 3. Additional kvars available due to reduction of DER kW by 10% from different initial DER kWs, %. DER Volt =0.93pu

Notice that the incremental vars are greater under the greater DER’s power factor. This is because under the greater power factor the initial var capability of DERs is smaller.

If the additional reactive power is needed to raise the voltages in some voltage-critical points, the impacts of both the reduction of the DER real power and the increase of the DER reactive power should be compared.  The following conditions should be met:

Δkvar(DERk) x Reactance(k-l)+ ∆kW(DERk) x Resistance(k-l) > 0              (1)

Volti,min  ≤Volti  ≤  Volti,max                (2)

Ampj ≤ Ampj,max                    (3)

(Operating kW Reserve) ≥ (Operating kW Reserve)min ,      (4)

Where

k – is the k-th DER involved in the kw-kvar exchange

k-l  - is the electrical path either between the l-th critical point and source of supply, or between the k-th DER and source of supply, depending on the mutual allocation of the critical point and DER

i – is the i-th node in the subject network

j – is the j-th element in the subject network

min – is the minimum limit

max – is the maximum limit 

As follows from (1)

∆kvark / ∆kWk > resistance(k-l) / reactance(k-l)                     (5)

The figures below illustrate the effects of the VVC in var priority mode for a single node connected to distribution circuits with different R/X ratios (Figure 1). Figure 2 illustrate the difference in the effect of this VVC mode for DERs with different rated power factors, and Figure 3 shows the difference of the effects for different initial DER injections of the kWs.

Figure 1. Change of load voltage due to reduction of DER’s kW from 100% and increase in DER’s var capabilities for two initial load voltages (93% and 98%) for different R/X ratios (DER’s PF=0.95)

Figure 2. Change of load voltage due to reduction of DER’s kW from 100% and increase in DER’s var capabilities for two DER’s PF (0.95 and 0.90), R/X = 1/3

Figure 3. Change of load voltage due to reduction of DER’s kW from 100% and from 50% and corresponding increase in DER’s var capabilities (PF=0.95, R/X = 1/3)

As seen from the above tables and illustrations, the required effect of this mode of VVC may not be achievable under a number of operational and infrastructural conditions.

The dominant R/X ratios of the circuits upstream from the voltage critical point and the initial DER’s KW define whether the reduction of the voltage drop due to greater injections of DER’s vars is sufficiently greater than the increase of the voltage drop due to the reduction of DER’s Watts. E.g., it means that the var priority mode of DERs connected to the secondaries of the distribution transformers may be efficient for the voltages in the relevant secondary circuit, but may have detrimental effect on other load node connected through the primary distribution due to the voltage-reducing effect of the reduction of DER’s kW.

The var priority mode of the DERs with a lower rated power factor has a smaller effect (Figure 2) because their initial injection of vars is greater, and the increments of the additional vars are smaller. (Remember, the var priority mode is applied only after the initial DER’s var injection is the maximum available). The same applies to the smaller than 100% initial DER’s kW (Figure 3).

 There is another function of the Smart Distribution Grid – the Volt-Watt function [1]. The main objective of this function is reducing the overvoltage by reducing the kW of some DERs. Because the DERs have the capability of absorbing kvars, this function is similar to the Volt/var with var priority mode. In this case, reduction of DER’s kW releases additional available absorbing kvars. In these cases, the reduction of the DER’s kWs and the increase of the absorbing kvars may reduce the voltages.

As follows from the above, the effect of the VVC in var priority mode depends on many operational and facility factors.  To assess these effects in near-real time, an optimization procedure based on comparative power flow simulations should be used.

Consider an example of the application of the “Volt/var with var priority” function in a circuit with multiple DERs.  Figure 4 presents the example diagram. The transmission equivalent is represented by an 115kV network. The distribution is a 12kV network. The sample initial operating conditions are such that the secondary voltages in nodes 1210 and 1209 are below the lower voltage limits. The bus voltage is kept constant by the LTC controller, and the needed level of the bus voltage is determined by requirements from other feeders. Hence, the Volt/var control with var priority function is considered. Different cases of DER kW reduction have been analyzed. The paths between the voltage critical points and the main source of supply do not include the DERs in nodes 1202, 1203, and 1204. That is why these DERs were not considered in the analysis. A number of other combinations of reductions of DER kW for additional kvars were considered (see Table 4)

Figure 4. Diagram of the example circuits

Table 4. Different cases of reduction of the DER’s kWs (%) for additional kvars

As seen in Table 4, Case 1 represents the initial conditions, when the voltage in node 1210 is 92% (see Table 5). Cases 2 through 6 represent different degrees of reduction of DER kW in the secondaries of the critical node. As seen here, the voltage does not reach the standard level (95%). The best effect of kW reduction for additional kvars just in node 1210 gets the voltage in this node at 93.9% (case 5 in Table 5). Combinations 7 through 11 also do not provide the desired results. Combination 12, which involves all DERs contributing fully or partially kWs and kvars to the paths between the critical points and the source of supply, brings the voltages in the critical points into the standard range. 

As it was mentioned before, the substation voltage controller keeps “constant” voltage, which is true with the accuracy defined by the bandwidth of the voltage controller. As long the bus voltage is within the range of the bandwidth, the bus voltage changes with the changes of the net kWs and kvars due to the changes of the voltage drop in the transformer according to the impedance of the substation transformer.  However, if the changes of the kWs and kvars result in bus voltage beyond the bandwidth, the controller will change the voltage in the opposite direction by, at list, one LTC step. It means that the bandwidth of the substation voltage controller introduces an uncertainty in the effects of the var priority mode of VVC.

If the bus voltage is controlled with Line Drop Compensation (LDC), the LDC changes the bus voltage with the change of the net kW and kvars according to its LDC-R and LDC-X factors (still with the bandwidth uncertainty). In this case, reduction of DER’s kW increases the bus voltage, but increase of DER kvar injections reduces the bus voltage. The results depend on the ratio of the LDC-R/LDC-X factors. In our example with the ratio of the LDC-R/LDC-X factors equal 0.5, (the same as the R/X ratio of the primary distribution), the standard voltage in node 2010 can be reach with a different from case 12  allocation of the DER kWs.  

Table 5. Effects of kW reduction for additional kvars

When  the kWs and kvars of DERs are changed, the power flows in both the distribution and transmission networks are also changed. This means that the losses, nodal voltages, loading of elements, operating reserves, and other dependent parameters are changed. We can say that there are “cost” and “benefits” of such changes. If the VVC in var priority mode is included in the system planning process and is considered as an alternative to other means of voltage support in distribution, the cost/benefit relationship between the alternatives should be ananlyzed. It may be a challenge to adequately determine these relationships for each specific distribution network

Changes of relevant operational parameters for the sample circuit are presented in Table 6.  As seen in the table, the natural real load in the non-critical points is increased by 1.8%, and the reactive load is increased by 5.5% due to the increase in the weighted average secondary voltages by 2.0%. The customer power factor is reduced, which may cause additional cost to some customers. The total DER’s kW reduction was 13%, while the DER’s kvars are increased almost three times. The loading of the transmission line feeding the subject substation is increased by 10%. The kWs flowing into the distribution system from the transmission system in case 12 are increased by 24%, while the net kvars are reduced by 78%, which resulted in a greater net power factor.  The losses in distribution and in transmission systems are also increased. 

Table 6.  Additional factors affected by the reduction of DER’s kW for more kvars

*- percent of Case 12

The sample distribution circuit (for cases 1-12) was an overhead distribution system with the R/X ratio of the involved primary circuits around 0.5. The R/X ratio of the distribution transformer and the secondary circuits in the voltage critical node is 0.3. If the circuit were with a higher ratio (e.g., underground), the effect would be much smaller, if any, and the cost would be much higher.  On the other hand, if the circuit were of a smaller R/X ratio, then the effect would be greater, and the cost would be smaller.  Such an example is presented in case 13. In this example, the R/X ratio is around 0.3.

As seen in case 13, the kWs flowing into the distribution system from the transmission system are reduced in comparison with Case 12.  The loading of the transmission line feeding the subject substation and the losses in distribution and transmission systems are also reduced.

As follows from the above analyses, the many entities of different ownerships paid some “cost” to improve the voltages in two distribution nodes. This cost should be compared with the benefits that this way of voltage improvement provides in comparison with other approaches.

Consider another use of the Volt/var with var priority function. In this case, all DER owners autonomously run the Conservation Voltage Reduction (CVR) mode of Volt/var control.  In order to reduce the voltage, they reduce the injection of vars in the circuit. Because of this, the voltages in the entire circuit are reduced, and in one node, it is reduced below the standard voltage limits. The DER in this voltage-critical point cannot produce enough, if any, reactive power to support the needed voltage at its bus. In order to provide the customers in this node with quality voltage, the distribution system operator (DSO) or the DMS needs to increase the voltage at the bus of the main feeding substation, increasing the voltages in the entire distribution network fed from this bus. The CVR-running DERs further reduce the var injections fighting the increase of voltage by the DSO. This may last until the full capability of the DER inverters is utilized. Such an increase of the bus voltage may be unacceptable due to a number of operational constraints. In such a case, reduction of the kW injection by the DER in the voltage-critical point to release its reactive power may help in supporting the voltage in this point and, at the same time, to avoid unacceptable increase of the substation voltage.

Consider an example. The sample circuit is the same as in the previous example with the R/X ratio about 0.5. The results of the analysis are presented in Table 7 (cases 14 and 15).

Table 7.  Results of analysis of the CVR case, % of the parameters of the reference Case 12. 

As seen in the table, the combined actions by the DSO and the DER owner in node 1210 (case 15) result in reduction of total customer loads and in reduction of total losses.  These reductions benefit all customers. However, the customer in node 1210 needs to take additional 50% of kW from the grid to compensate for the reduction of its DER injection. This additional cost to the customer 1210 can be considered as payment for the improved voltage quality.

Case 16 in Table 7 presents a case when the Volt/var function is coordinated by the DSO/DMS. In this case, the objective of the function is still CVR, but the DSO requests providing maximum available kvars from the DERs.  The additional supply of vars from the DERs allows the DSO to reduce the substation bus voltage. The difference between this case and the case of autonomous Volt/var controls is presented in the last column of Table 7. As seen in the column, the total losses are reduced, however, the total customer load is slightly increased. The reduction of the net kWs means that the loss reduction in distribution exceeds the increase of the customer real load.  

As follows from above, the effectiveness of the var priority mode is the highest, when the real power injections of the DERs are at 100%.  When the DER’s kW injections are significantly less than the maximum, the available reactive power capability of the DERs is much greater. However, if this increase in the reactive power injection does not compensate for voltage reduction due to the reduction of DER's kW, the additional reduction of the DER's kW may be ineffective.

Figure 5 presents a sample of a "duck" load curve, which was applied to the sample circuit presented above.   

Figure 5. Sample Load shapes with solar generation

In this case, the natural load at the time of DER’s peak kW is about 0.9 of the evening peak load, and it is about 10% higher at the time, when the DER’s kW contribution is about 50% of its peak kWs.

Table 8 presents the initial conditions for the 50% DER’s kW contribution in relation to the initial condition for the 100% DER’s contribution (case 1). The lowest load voltage in this case is higher than in Case 1 due to a greater var capability of the DERs, but still below the voltage limit. The second column of the table presents the increase of the critical voltage to the low standard limit by raising the voltage at the feeding substation bus by 3%. In this case, all load voltages in the distribution nodes were increased, which led to an increase in customer loads by about 3%. The effort to apply the var priority mode was unsuccessful: the reduction of DER’s kW did not produce enough kvar to raise voltage. Instead, a 600-kvar capacitor was suggested near to the voltage critical points (the third column in the table below). It brought the voltage to the standard level and raised the average voltage just by 0.7% with a corresponding smaller increase in the customer load.

By the way, under the conditions of 100% contribution of the DER’s kWs, the addition of the 600-kvar capacitor near the voltage critical point would solve the voltage quality problem by raising the bus voltage by just 0.7%, or by reducing the DER’s kW by just 4% in node 1210.

Table 8. Results of analysis of var priority mode under 50% kW injections from DERs, % of the initial case 1

Under smaller R/X ratios, and/or higher load power factors, the increase of the DER's var capability might be sufficient to compensate for the voltage reduction due to the DER's kW reduction.

Conclusions

1. The efficiency of the Volt / var with var priority function in distribution with high penetration of DER depends on a number of factors, such as:

• the dominant R/X ratio of the circuits,

• the mutual allocation of the voltage-critical points and the DERs,

• the sizes of the DERs,

• the rated power factors of the DERs,

• the initial loading of the DERs,

• the load power factors,

• the operational objectives of the VVC function

• the availability of centralized vs  autonomous control of distribution operations

• the mode of voltage control in substations feeding the distribution system

• the operational constraints in related circuits

• the current connectivity of the distribution circuits

• the objectives of the transmission operations regarding volt/var control

• the availability of the DERs to participate in the Volt / var with var priority function

• other.

1. To take into account all the above factors and find the optimal solution for the var priority modes of multiple DERs, comparative power flow analyses should be performed, in other words, an optimization procedure  should be applied. It also implies that adequate communications between the major DERs, microgrids, and the DMS should be available.

2. The cost of mitigating voltage violations in distribution by using the VVC with var priority function with the dominant R/X ratios above 0.5 may be very high, if possible at all.  Under such conditions, an upgrade of the circuit facilities may be a more efficient solution and should be considered in the planning stage.

3. If the voltage violation happens when the DERs are contributing about 50% of their maximum kW capability, the var priority mode is ineffective, even when the R/X ratio is low.

4. Applying different means of the Volt/var control function by the DSO/DMS in a coordinated manner instead of autonomous controls by the DER owners may provide greater benefits to all customers.

References and further reading.

1. John Berdner, Advanced Inverters and Grid Support. Available: http://www.clean-coalition.org/site/wp-content/uploads/2014/05/Grid-support-With-Advanced-Inverters.pdf

2. Nokhum Markushevich, ‘What will the Microgrids and EPS Talk about?’ Part 1 and 2. Available: http://www.energycentral.com/gridtandd/gridoperations/articles/2858  and http://www.energycentral.com/gridtandd/gridoperations/articles/2864

3. Coordination of Volt/var control in Connected Mode under Normal Operating Conditions.   Available: http://smartgrid.epri.com/Repository/Repository.aspx/

4. Update aggregated at PCC real and reactive load-to-voltage dependencies. Available: http://smartgrid.epri.com/Repository/Repository.aspx/

5. Updates of capability curves of the microgrid’s DERs. Available: http://smartgrid.epri.com/Repository/Repository.aspx/

6. Understanding Coordinated Voltage and Var Control in Distribution Systems: Is Power Factor = 1 Always a Good Thing? Available: http://www.energycentral.com/gridtandd/gridoperations/articles/1553/Understanding-Coordinated-Voltage-and-Var-Control-in-Distribution-Systems-Is-Power-Factor-1-Always-a-Good-Thing-