Grand Rapids Excitation and PSS Interconnection Evaluation Study

Transcription

1

2 Table of Contents TABLE OF CONTENTS EXECUTIVE SUMMARY RECOMMENDATIONS Recommended Grand Rapids exciter parameters Recommended Grand Rapids PSS parameters Install a PSS disable switch for each Grand Rapids generator unit Funding for the modifications BACKGROUND INFORMATION The Excitation System and PSSs at Grand Rapids The Power System Stabilizers (PSSs) at Kelsey The Power System Stabilizers (PSSs) at Jenpeg The Power System Stabilizers (PSSs) at Kettle U1 and U The Future Power System Stabilizers (PSSs) at Wuskwatim Automatic Generation Control (AGC) Ponton Static Var Compensators (SVC) Significance of Mode Shape, Relative Participation Factor and Damping Criteria Explanations of Simulation Procedures and Study Case in PSSE and SSAT CONFIGURATIONS STUDIED Existing Grand Rapids Configuration: U1 and U3 with rotating exciters, U2 and U4 with static exciters, No PSSs installed Future Grand Rapids Configuration: U1 and U3 with rotating exciters; U2 and U4 with existing static exciters and PSSs installed Future Grand Rapids Configuration: Static exciters on all generator units with PSSs installed Frequency Domain SSAT Analysis Excitation and PSS Considerations APPENDIX A: INTERCONNECTION REQUEST FORM APPENDIX B: CAPITAL PROJECT JUSTIFICATION FOR GENERATION SOUTH EXCITATION PROGRAM APPENDIX C: CORRESPONDENCE Page 1

3 1. Executive Summary Today s Grand Rapids generators are operated with static exciters on units 2 and 4 and rotating exciters on units 1 and 3. No power system stabilizers (PSSs) are installed on any of the units at this generating station. This interconnection evaluation study (IES) report was written in response to an Interconnection Request submitted by the Generation Maintenance Engineering Department. According to the Interconnection Request, initially PSSs will be added to units 2 and 4. Secondly, the existing amplidyne rotating exciters on units 1 and 3 will be replaced with a static, fast-acting excitation system with built in PSS functionality. The electromechanical gate-blade angle governor controls will also be changed to a digital governor head with a separate compensated gate blade angle control. The study concludes that the proposed excitation system upgrades, and the addition of PSSs at Grand Rapids would improve the damping of the low frequency oscillations occurring in the system. It provides the generator owner with parameters for the PSSs and exciters. Also recommended, is the installation of PSS disable switches in the station control room, as well as the system control center (SCC), for PSS operation in the rough zone. The funding for the modifications will come from the Generation South Excitation Program. Page 2

4 2. Recommendations 2.1 Recommended Grand Rapids exciter parameters In order to allow for optimal PSS operation, it is recommended that the modified exciters on all of the units satisfy the following voltage response time criterion: The time for the excitation voltage to reach 95% of the difference between ceiling voltage and rated load field voltage shall be less than 25 milliseconds after a sustained 5% step drop is applied to the generator terminal voltage. The exciter will be required to circulate positive field currents only; source [9] Since each of the generators at Grand Rapids is connected to the high voltage bus through an individual step up transformer, the exciters should also include line drop compensation (LDC) from zero to 60% of the transformer impedance. It is recommended that bus fed exciters be used on each of the units. In addition, the recommended parameters for the E max and E min values for the exciters are 5.33 pu and -4 pu respectively. These values are currently used on the static exciters on Grand Rapids units 2 and Recommended Grand Rapids PSS parameters Figure 1: IEEE PSS2A stabilizer model; source [9] Table 1: Parameter values for Grand Rapids PSSs Parameter Study Value Parameter Range Parameter Study Value Parameter Range Tw1 10 sec 0 Tw1 30 Ks Ks1 40 Tw2 10 sec 0 Tw2 30 T1.06 sec 0 T1 5 T6 0 0 T6 10 T2.025 sec 0 T2 5 Tw3 10 sec 0 Tw3 30 T3.14 sec 0 T3 5 Tw4 0 0 Tw4 30 T4.025 sec 0 T4 5 (bypassed in PSS/E) T7 10 sec 0 T7 30 M 5 0 M 10 Ks2 (Tw1/(2*H))=.91 0 Ks2 5 N 1 1 N 2 Ks3 1 0 Ks3 1 Vpss Max.05 pu 0 Vpss Max.2 T8.5 sec.1 T8 3 Vpss Min -.05 pu -.2 Vpss Max 0 T9.1 sec.02 T9 1 Page 3

5 It is recommended that the IEEE standard PSS2A type stabilizers be used on all of the units with generator electric power and rotor speed inputs. Figure 1 above provides the functional block of the PSSs and table 1 summarizes the study parameters used in the simulations. 2.3 Install a PSS disable switch for each Grand Rapids generator unit Individual disable switches for the PSSs on each of the Grand Rapids generator units should be installed in both the station control room and the System Control Center (SCC). These switches would allow the operators to conveniently disable the PSSs when operating in the rough region, below 65% gate. 2.4 Funding for the modifications The recommended modifications have already been approved and will be funded by the Generation South Department, who has prepared a capital project justification (CPJ) for the Generation South Excitation Program, see appendix B. Page 4

6 3. Background Information 3.1 The Excitation System and PSSs at Grand Rapids In time and frequency domain simulations, each of the Grand Rapids units was modeled separately. The models used are described below. Figure 2: Rotating Exciter Model Used in the Simulations Table 2: Rotating Exciter Parameters Parameter Study Value Parameter Study Value Tr 0 Ka 29 Ta.06 Vrmax 1 Vrmin -1 Ke 0 Te.143 Kf.1 Tf 1 E SE(E1).16 SE(E2).4 E Recurring forced outages have been attributed to the outdated excitation systems on Grand Rapids units 1 and 3. These problems are complicated by the lack of parts, as described in appendix B. Both of the excitation units were modeled using the standard IEEET1 exciter model, with the parameters provided in table 2 above. Figure 3: Static Exciter Model Used in the Simulations Table 3: Static Exciter Parameters Parameter Study Value Ta,Tb.1 K 125 Te.02 Emin -4 Emax 5.33 rc/rfg 10 The static exciters on Grand Rapids units 2 and 4 are fairly modern and have been reliable. Page 5

7 The model used to represent the static exciters at the Grand Rapids GS was the SCRX PSS/E model shown in figure 3 above, with the parameters outlined in table 3. These parameters satisfy the exciter time response criterion listed in the Recommendations section 2.1, and are able to accommodate PSS2A type stabilizers. Presently, there are no PSS installed at the Grand Rapids GS. The stabilizers are a requirement of both the MB Hydro Transmission System Interconnection Requirements [2] and the Midwest Reliability Organization (MRO) PRC-502-MRO-01 approved standard on small signal stability which states that the generator owner shall install power system stabilizers on all or substantially modified generator units with a nameplate rating 100 MVA or larger 3.2 The Power System Stabilizers (PSSs) at Kelsey Generator units 6 and 7 at the Kelsey GS are able to supply power to both the Northern AC and the NCS, while units 1-5 are operated solely on the Northern AC system. An interoffice memorandum composed in 2004 by B.A. Archer, and referenced in [11] recommended PSS2A type PSSs and provided the simulated settings. These modifications are currently in the process of being implemented. In the stability package used, the Kelsey generators were lumped as a single 224 MW generator model with the exciter and PSS parameters outlined in [11]. 3.3 The Power System Stabilizers (PSSs) at Jenpeg Jenpeg units 1-6 were also modeled as a single generator in the stability package. The PSSs and exciters at this generating station are integrated together. The accuracy of the parameters used to model the exciter and the PSS at this generating station has not recently been verified with actual system tests. 3.4 The Power System Stabilizers (PSSs) at Kettle U1 and U2 Kettle units 1 and 2 are able to operate both on the NCS and the AC system. The PSSs on these units have dual input, dual setting PSSs, which allow the units to operate with separate settings. In simulations, each generator units was modeled individually with parameters specified in [8,13]. 3.5 The Future Power System Stabilizers (PSSs) at Wuskwatim Wuskwatim will be a 223 MW generating station with an anticipated in service date of The exciter and PSS parameters used in the simulations are those recommended in the 2003 Requirement for Wuskwatim Generators and Controls report, referenced in [1]. Each of the Wuskwatim generators was modeled individually in the study case. 3.6 Automatic Generation Control (AGC) The Grand Rapids generating station AGC is used to maintain a stable frequency in the system, as well as on the tie-lines. This control can increase or decrease the generation at the station in order to maintain a stable system frequency of 60 Hz. Since the AGC is fairly slow, it is not expected to impact PSS operation or any of the recommended excitation system modifications proposed in the interconnection request. The AGC is not modeled in the simulations because its impact would not be seen in the transient simulations anyways since these simulations are usually approximately 10 seconds long. Furthermore, Grand Rapids is the swing bus in the system and is effectively acting as the AGC by providing the required power and maintaining frequency control. 3.7 Ponton Static Var Compensators (SVC) A Static Var Compensator (SVC) was placed at the Ponton bus to provide voltage support and allow for higher power transfer levels on the P19W line. The parameters used to model the SVC and the damping controller are detailed in the System Planning memorandum, reference [12]. Page 6

8 3.8 Significance of Mode Shape, Relative Participation Factor and Damping Criteria Mode shape is used in identifying groups of coherent and un-coherent generators. Inter-plant modes with two groups of mode shapes which are significantly different show that the rotors of the plants have an angle difference with respect to each other and that the two plants are exchanging power between each other. They are usually normalized to the largest participation factor. The relative participation factor provides information on how significant a contribution each of the generating stations make to the particular mode [4]. This can be used as an indication of the location for the placement of power system stabilizers [5]. Adding damping to generators where the participation factor is real and positive will increase the damping of that particular mode, as stated in [5]. The MB Hydro Transmission System Interconnection Requirements (2007) document states that A power system oscillation damping ratio which exceeds 5% is acceptable, between 3% and 5% is marginal and below 3% requires mitigation. The criterion applies to the entire MB Hydro system, including the Northern Collector. 3.9 Explanations of Simulation Procedures and Study Case in PSSE and SSAT The studies and simulations presented in his report were performed using the 2005 stability package, which is based on the 2004 series MAPP models. The 2014 winter peak loadflow was used. Same loadflows were used in both PSSE time domain and SSAT frequency domain simulations. In the PSSE simulation, the Eigenvector Fit by Least Squares method was used to obtain the frequency modes along with their respective eigenvalues and eigenvectors. Although more computationally intensive than the Prony method, it is intended for use in systems with non linear inputs [3]. According to the manual in [3], the Least Squares method also provides results with a higher signal to noise ratio and lower percent error. The skip factor is used to control the number of points on the curve that are to be used in the fitting process [3]. It is defined as the number of points to be skipped before approximating the next value in the analysis. For example, a skip factor of 8 means that every 8th point of the original plot will be used in the fitting process. Once the user graphically selects the time range of the plot, the software will display the number of points in that time interval and prompt the user for the skip value. The number of points used in the fitting process will then be the number of points in the selected interval, divided by the skip value. The rule of thumb for this post-processing least squares analysis, according to Power Technology Inc. (PTI) is that the maximum number of points used in the fitting process when using the Prony method should not exceed 100 [3]. Frequency domain simulations in SSAT provide accurate mode shape and participation factor data. The DC system is detailed explicitly and in more detail in SSAT, compared to the way it is modeled in PSSE. Page 7

9 4. Configurations Studied 4.1 Existing Grand Rapids Configuration: U1 and U3 with rotating exciters, U2 and U4 with static exciters, No PSSs installed As mentioned earlier, Grand Rapids units 1 and 3 are operating with rotating exciters while units 2 and 4 are operating with static exciters. No PSSs are installed on any of the 4 Grand Rapids units. The models and their respective parameters were discussed and detailed in section 3.1 of the report. In order to excite inter-area and inter-plant modes, Kettle units 1 and 2 were operated on the AC system with maximum generation, and the real power outputs at Kelsey, Grand Rapids, Wuskwatim and Jenpeg were also maximized. Time and frequency domain simulations were performed by applying different contingencies to the system. Figure 1: P19W flow following a 30 MW drop at INCO Figure 2: Grand Rapids U1-4 output following a 30 MW drop at INCO Figure 1 above shows the power flow on the line P19W, connecting Dunlop and Ponton. Wuskwatim, Grand Rapids, Kelsey, Kettle (U1 and U2) and Jenpeg stabilizers were enabled and Page 8

10 a 30 MW drop was simulated at the INCO load. The original loading at INCO was set at 135 MW. The frequency of the oscillation is approximately.67 Hz. Figure 2 shows the power oscillations on each of the Grand Rapids generators, following the same contingency. Figure 3: P19W flow, following a 3 phase, 3 cycle Ralls Island bus fault Figure 4: Grand Rapids U1-4 output following a 3 phase, 3 cycle Ralls Island bus fault Figure 3 above shows the real power flow on line P19W. Wuskwatim, Grand Rapids, Kelsey, Kettle (U1 and U2) and Jenpeg stabilizers were enabled and a Ralls Island 3 phase, 3 cycle bus fault was simulated. As shown in figure 4, the same contingency excited a 1.2 Hz oscillation on each of the Grand Rapids units as well. From the figures above, without performing extensive calculations, it can be concluded that the oscillations are well below the 5 % damping criteria outlined in the MB Hydro Transmission System Interconnection Requirements. Page 9

11 Figure 5: P19W flow, following a 3 phase, 3 cycle P58C line fault Figure 6: Grand Rapids U1-4 output, following a 3 phase, 3 cycle P58C line fault A 3 phase, 3 cycle line fault was simulated on the P58C line. Figure 5 and 6 shows the power oscillations on the P19W line and the Grand Rapids generators, respectively. The oscillations on the Grand Rapids generators indicate that the units are oscillating in synchronism. Again, the plots in figures 5 and 6 both indicate that the damping on these oscillations is poor and that it requires significant improvement. Page 10

12 Figure 7: P19W flow, following a 3 phase, 3 cycle G8P line fault Figure 8: Grand Rapids U1-4 output, following a 3 phase, 3 cycle G8P line fault G8P is a major transmission line which connects the Grand Rapids generating station to the William River and Ponton bus. Figures 7 and 8 above demonstrate that a 3 phase, 3 cycle G8P line trip can cause significant oscillation to occur on both the generator units and the P19W line. Figure 8 shows the frequency of the power oscillations at approximately 1.2 Hz. Page 11

13 4.2 Future Grand Rapids Configuration: U1 and U3 with rotating exciters; U2 and U4 with existing static exciters and PSSs installed According to the CPJ in appendix A, the first modification at Grand Rapids will be adding PSSs on units 2 and 4, which is scheduled for implementation in 2012/2013. The simulations were performed again with this configuration and shown in the figures below. Figure 9: Effect of adding Grand Rapids U2 and U4 PSSs, following a 30 MW drop at INCO Figure 10: Effect of adding Grand Rapids U2 and U4, following a 30 MW drop at INCO Figure 9 above shows the real power flow on line P19W, connecting Dunlop and Ponton. Wuskwatim, Grand Rapids, Kelsey, Kettle (U1 and U2) and Jenpeg stabilizers were enabled and a 30 MW drop was simulated at the INCO load. The original loading at INCO was set at 135 MW. The solid line demonstrates the contingency with PSSs installed on Grand Rapids units 2 and 4, while the dotted line demonstrates the situation with no PSSs installed at Grand Rapids. The damping improvement is apparent. Similarily, figure 10 shows an improvement in the damping of oscillations on Grand Rapids unit 2, following the same contingency. Page 12

14 Figure 11: Effect of adding Grand Rapids U2 and U4 PSSs, following a Ralls Island bus fault Figure 12: Effect of adding Grand Rapids U2 and U4 PSSs, following a Ralls Island bus fault Figure 11, shows the real power flow on line P19W. Wuskwatim, Grand Rapids, Kelsey, Kettle (U1 and U2) and Jenpeg stabilizers were enabled and a Ralls Island bus fault was simulated. The solid line demonstrates the contingency with PSSs installed on Grand Rapids units 2 and 4, while the dotted line demonstrates the situation with no PSSs installed at Grand Rapids. In figure 12, the dotted line represents the power output from U2 with PSSs installed at Grand Rapids U2 and U4, while the solid line represents the power output on U2 with no PSSs installed on any of the Grand Rapids units. Page 13

15 4.3 Future Grand Rapids Configuration: Static exciters on all generator units with PSSs installed According to the CPJ in appendix B, once all of the modifications are completed, each of the Grand Rapids generator units will be equipped with fast acting, static exciter and PSS. In the simulations, all four Grand Rapids PSSs were modeled using identical static exciters and stabilizers presented in section 3.1 of the report. Figure 13: Grand Rapids upgrades complete: power flow on unit 1; 30 MW load drop at INCO Figure 14: Grand Rapids upgrades complete: P19W power flow; 30 MW load drop at INCO Figure 13 above shows the power output of Grand Rapids unit 1. Wuskwatim, Grand Rapids, Kelsey, Kettle (U1 and U2) and Jenpeg stabilizers were enabled and a 30 MW drop was simulated at the INCO load. The original loading at INCO was at 135 MW. The damping on the oscillations is improved with the addition of PSSs on units 2 and 4, and is further improved by placing stabilizers on all 4 Grand Rapids units. Figure 14 shows the power flow on line P19W Page 14

16 following the same contingency. It clearly demonstrates the ability of the PSSs at Grand Rapids to improve the damping of the oscillations. Figure 15: Grand Rapids upgrades complete: P19W power flow; Ralls Island bus fault Figure 16: Grand Rapids upgrades complete: power flow on unit 4; Ralls Island bus fault Figure 15 shows the results of a simulated Ralls Island bus fault. It shows the power flow on P19W as a result of the contingency. In this case, the damping of the oscillations on the lines with the PSSs at Grand Rapids GS appears to be identical in both situations: with U2 and U4 and with U1-U4 PSSs installed. Figure 16 shows the power output of Grand Rapids unit 4, following the same disturbance. It illustrates the improvement in the damping among the cases with no PSSs installed, U2 and U4 PSSs installed and all 4 Grand Rapids units equipped with PSSs. Again, the inter-machine mode has a frequency of approximately 1.2 Hz. Page 15

17 Figure 17: Grand Rapids upgrades complete: P19W power flow; V38R line fault Figure 18: Grand Rapids upgrades complete: unit 2 power flow; V38R line fault A 3 phase, 3 cycle V38R line (connecting Raven Lake to Vermilion) fault was also simulated in order to test the ability of the PSSs at Grand Rapids to damp out oscillations. Figure 17 above illustrates the power flow on the P19W line, following the simulated contingency. Placing PSSs on Grand Rapids units improves the damping on the line. Figure 18 shows the power flow on unit 2 of Grand Rapids, during the same contingency. Without the PSSs, the damping ratio is below 5 % and requires improvement. As was expected, maximal damping is achieved when all units are equipped with static exciters and stabilizers. Page 16

18 Figure 19: Grand Rapids upgrades complete: P19W power flow; G8P line fault Figure 20: Grand Rapids upgrades complete: unit 3 power flow; G8P line fault A 3 phase fault of 3 cycle duration was simulated on the G8P transmission line. Figures 19 and 20 compare the effects of this contingency with all of the PSSs disabled, and all of the PSSs enabled. Both figures clearly verify the PSSs ability to provide sufficient damping to the resulting oscillations. Page 17

19 4.4 Frequency Domain SSAT Analysis Existing Excitation System Configuration: U1 and U2 with rotating exciters, U2 and U4 with static exciters, no PSSs installed As described in the background section, the frequency domain simulations were performed in SSAT in order to determine the participation factor and mode shape. The loadflows were imported from PSSE, along with the dynamic data. Unfortunately, the Ralls Island fault could not be simulated in frequency domain because the contingency is too severe for PSAT to solve. In addition, SSAT does not allow fault clearing and therefore some of the PSSE time domain simulations could not be replicated in SSAT. Figure 21: Grand Rapids inter-machine oscillation a) mode shape b) participation factor Figure 21 above shows the mode shape and participation factor of the marginally damped1.575 Hz inter-machine mode that is seen at Grand Rapids. A P58C line fault was simulated in order to excite this oscillation mode. Wuskwatim, Grand Rapids, Kelsey, Kettle (U1 and U2) and Jenpeg stabilizers were enabled. The mode shape shows that the units modeled with rotating exciters are oscillating against the units modeled with static exciters. The relative participation factor proves that all of the units have a high participation factor in the mode and that each of them would be a good location for PSS placement. Page 18

20 Figure 22: Grand Rapids inter-machine oscillation a) mode shape b) participation factor The figure above shows an inter-machine mode occurring between Grand Rapids units 1 and 2, as a result of a continuous p52e line fault. Both of these units are equipped with the obsolete rotating excitation systems. A damping ratio of 3.57 % is marginal according to the criteria in the MH Transmission System Interconnection Requirements. Page 19

21 Figure 23: Grand Rapids area oscillation a) mode shape b) participation factor The Grand Rapids generators are also participants in.886 Hz inter-area mode shown above. A G8P line fault was simulated and demonstrated.886 Hz mode with a negative damping which means that in theory, the oscillations increase in amplitude. The mode is unstable because the real eigenvalue is positive and the mode is located in the right side of the root locus plot. In this situation, Grand Rapids units are swinging against units which are located outside of Manitoba Hydro s system. The participation factor shows that all of the units at Grand Rapids are suitable locations for PSS placement. These results are consistent with the time-domain simulations shown in figure 8 on page 11. The fact that the damping in frequency domain is lower than that in the time domain is attributed to the lack of fault clearing ability in SSAT. Page 20

22 4.4.2 Proposed Excitation System and PSS Configuration: Static exciters and PSSs on all generator units The configuration represents the implementation of the proposed upgrades, as described in the IES Request (appendix A). Static exciters are added on all of the units along with power system stabilizers. The PSS parameters used were provided in Table 1 on page 3. Again, the same dynamic data and loadflows were used in the simulations. Figure 24: Grand Rapids inter-machine oscillation a) mode shape b) participation factor With static exciters and PSS on all of the Grand Rapids, the inter-machine mode frequency drops from Hz to Hz as the P58C line fault is applied. In comparison with figure 21 on page 18, with the new exciters and PSSs, the damping of the mode increases from 4.94 % to % and the control mode begins to manifest itself. The mode shape shows that the damping controller on the HVdc system is interacting with the stabilizers. Since the control oscillation is very well damped, it is not likely to be a concern from the simulation point of view. Page 21

23 Figure 25: Grand Rapids inter-machine oscillation a) mode shape b) participation factor The simulated line P52E contingency shown above confirms the ability of the PSSs to provide sufficient damping of the inter-machine mode. The damping of this mode without the proposed changes is 3.57 % (figure 22, page 19) and increases to 58.1 % with the implementation of the proposed changes. The mode shape again demonstrates the presence of a control mode occurring between the HVdc damping controls and the stabilizers at Grand Rapids. A change in the mode shape and participation factor is attributed to the addition of the PSSs and change in exciter models on the units. Page 22

24 Figure 26: Grand Rapids inter-area oscillation a) mode shape b) participation factor In comparison with figure 23 on page 20, figure 26 shows that the proposed changes increase the damping of the inter-area mode from -.66 % to %, as a G8P line fault is applied. The frequency of the mode decreased as the mode imaginary eigenvalue is decreased. The real eigenvalue also decreases, as the mode is pushed from the right to the left side of the root locus plot. Again, the mode shape and participation factor are different due to the changes in the models used for the exciters and PSSs at the Grand Rapids GS. Page 23

25 4.5 Excitation and PSS Considerations Static vs. rotating excitation systems Grand Rapids Rotating Exciter Phase Response 180 Phase [degrees] Frequency [Hz] Grand Rapid Static Exciter Phase Response 180 Phase [degrees] Frequency [Hz] Figure 27: Grand Rapids a) Rotating and b) Static Exciter Phase Response Figure 27 shows the phase response between the exciter input and the electrical torque output for the a) rotating and b) static exciters at Grand Rapids. As described in [7], this simulation was performed with a 100 times larger generator inertia, in order to ensure that the rotor speed and angle remain constant. According to figure 27 a), a speed based PSS would need to provide approximately 155 degree lead and an electrical power based PSS approximately a 65 degree lead (power based stabilizers have an inherent 90 degree phase lead), in order to effective damp the 1.5 Hz inter-machine mode, if rotating exciters are used. Figure 27 a) also shows a significant phase difference in the 0-2 Hz frequency range, which makes the PSS design with these exciters difficult and prone to noise, [7]. Figure b) shows that a speed based PSS would need approximately a 50 degree lead, while a power based would need a 40 degree lag, in order to provide damping for the 1.5 Hz inter-area mode. Since the phase difference in the static exciters is significantly smaller in the frequency range of 0-2 Hz than that in the rotating exciters, the static exciters would allow the PSS to provide sufficient damping for inter-area and inter-plant oscillations which are present at Grand Rapids. Page 24

26 4.5.2 Exciter Voltage Response Time In order to allow for proper power system stabilizer operation, it is necessary to ensure that the existing, as well as future exciters at Grand Rapids have a sufficiently fast response time. Sufficient voltage response time was defined in [9] as follows: The time for the excitation voltage to reach 95% of the difference between ceiling voltage and rated load field voltage shall be less than 25 milliseconds after a sustained 5% step change is applied to the generator terminal voltage. The exciter will be required to circulate positive field currents only. The parameters used in the simulations to model the existing and future exciters were presented in table 1 in section Line Drop Compensation (LDC) Each of the units at the Grand Rapids GS has an individual step-up transformer that connects it to the high-voltage bus. If the exciters are connected to the terminal voltage of the generator, each of the generators would be attempting to regulate the bus voltage, which would result in hunting problems. These problems can be mitigated by using line drop compensation (LDC) which takes in a measured current and passes it through a known compensating impedance. The voltage across this impedance is then taken, subtracted from the terminal voltage and the result used as a set point for the exciter. At Manitoba Hydro, 60 % of the transformer impedance is typically used for impedance compensation, as discussed in [9]. Since closer control to the high voltage bus contributes to system stability improvement, it is recommended that each static exciter have the ability to control a point from zero to a minimum of 60 % of its generator transformer. Page 25

27 Figure 28: IEEE PSS2A stabilizer model; source [9] Table 3: Study values for Grand Rapids stabilizers Parameter Study Value Parameter Range Parameter Study Value Parameter Range Tw1 10 sec 0 Tw1 30 Ks Ks1 40 Tw2 10 sec 0 Tw2 30 T1.06 sec 0 T1 5 T6 0 0 T6 10 T2.025 sec 0 T2 5 Tw3 10 sec 0 Tw3 30 T3.14 sec 0 T3 5 Tw4 0 0 Tw4 30 T4.025 sec 0 T4 5 (bypassed in PSS/E) T7 10 sec 0 T7 30 M 5 0 M 10 Ks2 (Tw1/(2*H))=.91 0 Ks2 5 N 1 1 N 2 Ks3 1 0 Ks3 1 Vpss Max.05 pu 0 Vpss Max.2 T8.5 sec.1 T8 3 Vpss Min -.05 pu -.2 Vpss Max 0 T9.1 sec.02 T Stabilizer Considerations and Parameters This section will discuss the proposed PSS parameters and provide the study values used in the simulations. Final PSS settings will come from System Performance department. There are currently no power system stabilizers installed at the Grand Rapids generating station. The study results demonstrate a need for an improvement in the damping of the.8 Hz inter-area mode and the 1.5 Hz inter-machine mode. In the studies, the frequency of the modes changed depending on the severity of the disturbance applied. Therefore, the PSS design needs to provide phase lead for a wide frequency bandwidth and not destabilize modes outside of the design frequency range. It is recommended that IEEE standard PSS2A stabilizers be used on all Grand Rapids generator units. The block diagram of this stabilizer is shown in figure 28 above. Table 3 provides the designed study parameters for the Grand Rapids PSSs. These parameters were used in all of the time and frequency domain simulations performed as a part of this study. Page 26

28 Figure 29: Root locus plots of the a) 1.5 Hz mode b).8 Hz mode The root locus plots in figure 29 a) and b) are used to demonstrate the effect of the stabilizer gain (Ks1 parameter) on the damping of the oscillations at Grand Rapids. The plotted eigenvalues provide information on the stability of the oscillations. An eigenvalue with a positive real component represents an unstable oscillation that is increasing in magnitude. A negative real eigenvalue represents a decaying magnitude of the oscillation. Imaginary eigenvalue gives the frequency of the oscillations. The results were gathered after modeling each of the generators at Grand Rapids using static exciters and PSS2A type stabilizers and simulating a G8P line (connecting the Grand Rapids GS and the Ponton SVC) outage with different stabilizer gain values. With the PSSs at Grand Rapids disabled (gain=0), the 1.5 Hz mode is marginally damped and the.8 Hz mode is negatively damped. With the Ks1 gain value of the stabilizers increased to the recommended study value of 20, the damping of the modes is increased far beyond the 5 % minimum acceptable level. The damping of the 1.5 Hz inter-plant mode is approximately % and the.8 Hz inter-area mode is %. Page 27

29 6.0 References [1] B.A. Archer, Wuskwatim Generating Station Requirements for Generators and Controls, System Planning Report, System Planning File SPD 02/7, [2] Manitoba Hydro, Transmission System Interconnection Requirements, Rev. 1, [3] Power Technologies Inc., PSSPLT Program Manual, User Manual Ver. 26, [4] Powertech Labs Inc, SSAT User Manual, User Manual Ver. 8, [5] G. Rogers, Power System Oscillations, Ref No /Rog, [6] A.M. Mian and B.R. Shellrude, Commissioning and Operating Experience with PSS at Kettle and Long Spruce, Report H&GPD File No. 420H1-1 & 427H1-1, [7] B.A. Archer, Kettle Units on AC - Stabilizer Considerations, Inter-office Memorandum to J.B. Davies, System Planning File No. 5-8, [8] D.E. Ans, Kettle G.S. Units 1 and 2 Power System Stabilizers - Completion and Remaining Work, Inter-office Memorandum to multiple recipients, File No , [9] B.A. Archer, Kelsey Exciter Replacement - Static Exciter and Stabilizer Considerations, Interoffice Memorandum to J.B. Davies, [10] B.A. Archer, L.E. Midford and J.B. Davies, Dual Configuration, Dual Setting, Digital Power System Stabilizer Simulation and Tuning Experience at Manitoba Hydro, Proceedings of IEEE PES Winter Meeting 2002, New York, February [11] B.A. Archer, Kelsey Excitation System and PSS Technical Specification, Inter-office Memorandum to J.B. Davies, System Planning File No 7-3, [12] D. Diakiw, Areva Ponton SVC and Damping Controller PSS/E Model Validation after Final Commissioning, Inter-office Memorandum to D. Jacobson, [13] D.E. Ans, Kettle G.S. - Replacement of Unit 1,2 Stabilizers, Inter-office Memorandum to G.A. Maher, Page 28

30 Appendix A: Interconnection Request Form Page 29

31 Manitoba Hydro ATTACHMENT 1 INTERCONNECTION REQUEST (All requested information and payment must be provided to constitute a valid Interconnection Request) 1. The undersigned Generator submits this Interconnection Request to install and operate generation interconnected with the Manitoba Hydro System pursuant to the Manitoba Hydro Open Access Interconnection Tariff. 2. This Interconnection Request is for (check one): A proposed new generating facility. An increase in the generating capacity of an existing generating facility. X A Substantial Modification to an existing generating facility. A generating facility proposed for inclusion in a Resource Solicitation Process. 3. The type of Interconnection Service requested is (check one): Energy Resource Interconnection Service X Network Resource Interconnection Service 4. Is Generator requesting expedited procedures for new generating facilities of less than one (1) MW or generating capacity additions of less than one (1) MW to existing generating facilities? _ X_ Yes No 5. Generator provides the following: a. Location of the proposed new generating facility site by section township and range, or by geographical coordinates, or, in the case of an existing generating facility site, the name and specific location of the facility: b. Maximum megawatt electrical output of the proposed new generating facility or the amount of megawatt increase in the generating capacity of an existing generating facility and/or a description of the Substantial Modification: Description of the Substantial Modification: The work would consist of: First, adding power system stabilizers to the existing Units 2 & 4 static exciters. Second, replacing the Grand Rapids GS existing Units 1 and 3 amplidyne rotating excitation equipment for high initial response full static excitation. (similar system the installation at Great Falls Unit 3, Kelsey Units 5 and 2; i.e.: ABB full static excitation system, with Power System Stabilizers (PSS2A/PSS2B) capability). Third, replacing the electro-mechanical gate-

32 Manitoba Hydro blade angle governor controls to a digital governor head with head compensated separated gate-blade angle control. c. Planned-in-service date by month and year of the new generating facility or increase in capacity of the existing generating facility or Substantial Modification: ISD of the first unit would be 2009 and as outages permit thereafter. d. Date of Application: October 09, 2008 e. Name, address, telephone number and address of Generator s contact person: Earl Borschawa Electrical Section Generation Maintenance Engineering Dept 1565 Willson Pl-Level 200, Winnipeg, MB, R3T 4H1 eborschawa@hydro.mb.ca f. A deposit (in the form of a certified cheque payable to Manitoba Hydro in the amount of $10, (Cdn.). As per existing accounting practice 6. This Interconnection Request and deposit shall be submitted to the representative indicated below unless submitted in response to a Resource Solicitation Process. In the latter case, the Interconnection Request shall be forwarded to the solicitor initiating the Resource Solicitation Process for submission to Manitoba Hydro. Manager, Transmission Services Department, Tariff, Administration Manitoba Hydro Box Taylor Avenue Winnipeg, Manitoba Hydro R3C 2P4 7. Manitoba Hydro shall maintain on OASIS a list of Interconnection Requests in accordance with the provisions of the Manitoba Hydro Open Access Interconnection Tariff. 8. Manitoba Hydro shall apply the Generator s deposit towards the cost of an Interconnection Evaluation Study. Manitoba Hydro shall either: (1) refund to Generator or solicitor of a Resource Solicitation Process any portion of the deposit that exceeds the actual cost of the Interconnection Evaluation Study within 30 days of completion of the Interconnection Evaluation Study or withdrawal or termination of the Interconnection

33 Manitoba Hydro Request; or (2) at the direction of the Generator or solicitor of a Resource Solicitation Process, apply any excess funds towards the cost of an Interconnection Facilities Study. 9. The terms and conditions of the Manitoba Hydro Open Access Interconnection Tariff as in effect on the date of this Interconnection Request are incorporated herein and made a part hereof. 10. I, the undersigned an authorized representative of the Generator, submit this Interconnection Request to Manitoba Hydro, with the understanding that Manitoba Hydro will subsequently provide an Interconnection Evaluation Study Agreement in accordance with the provisions of the Manitoba Hydro Open Access Interconnection Tariff. For Interconnection Requests submitted in response to a Resource Solicitation Process, Generator authorizes the solicitor of the Resource Solicitation Process (as identified below) to act as agent for the Generator in processing the Interconnection Request, until such time as the Interconnection Facilities Study Report is provided to the solicitor. The Interconnection Evaluation Study Report should be mailed to the following address: Nick Read, Manager Generation Maintenance Engineering Dept 1565 Willson Pl-Level 200, Winnipeg, MB, R3T 4H1 11. This Interconnection Request is submitted by: Earl Borschawa Electrical Section Generation Maintenance Engineering Dept 1565 Willson Pl-Level 200, Winnipeg, MB, R3T 4H1 eborschawa@hydro.mb.ca

34 Appendix B: Capital Project Justification for Generation South Excitation Program Page 30

35

36

37

38

39

40

41

42

43

44

45

46 Appendix C: Correspondence Page 31

47

48

49