Chapter- 7
REVIEW OF RESTORATION OF THERMAL POWER STATIONS
7.1 Background
The black start procedure has already been prepared by RLDCs and is available with all utilities. However, during the recent grid disturbances it has been observed that substantially longer time has been taken by certain generating stations to come on bars. In view of this, discussions were held with the utilities to review the time taken in restoration of generation after the recent grid disturbances. NRLDC, NLDC and UPSLDC also participated. Major observations and recommendations are given below:
7.2 Observations
i. Some of the utilities expressed that to initiate start up process, certain delays were encountered on account of commercial issues in obtaining the start- up power supply from other outside agencies.
ii. Most of hydro stations were ready to provide the start-up power immediately after the grid disturbance. However due to complete collapse of the entire grid, the required quantum of load commensurate with the generation build-up rate were not available despite close coordination and intimation given to the concerned load dispatch centres and personal contact with the counterpart distribution utilities. A pre-defined arrangement for availability of loads under such emergency conditions would have hastened the process of restoring the power supply. This may require to be looked into by the concerned agencies.
iii. NRLDC suggested that the load dispatch centres should be authorized to advise action to the concerned utilities for extending power supply immediately to the black-starting units through exchange of special emergency code between the concerned load dispatch centres. This process would facilitate quicker restoration by cutting down time required in taking administrative clearance which is otherwise obtained under normal grid operation conditions.
iv. Existing Black start procedures should be frequently reviewed in line with the fast changing grid scenario and addition of generation capacity. The facilities available with existing and upcoming IPPs should also form part of these procedures for the purpose of extending start up supply to black starting units in the vicinity.
v. All utilities felt the need to strengthen and have a dedicated communication network between SLDCs and all power plants in the respective control areas, which does not adequately exist at present and the agencies depend mainly on mobile phone facility, which is not completely reliable for such purposes. Availability of reliable and efficient communication facilities at all active installations connected to the grid is essential to ensure faster restoration.
vi. Various load dispatch centres, substations and generating stations, which are to implement the restoration operations in the real-time, upon receiving instructions from the apex load dispatch centres are not adequately managed in terms of experienced manpower and also particularly during odd hours. Utilities therefore expressed that the qualified operating personnel having undergone orientation courses under certification programme should be posted there.
vii. While examining the restoration data received from various utilities, it was observed that certain delays had occurred in lighting up the units, after start up supply was made available. The observed time duration ranged from 2 to 23 hours for Singrauli STPS and 2 hrs to 7 hrs for units at Unchhar, Rihand, Dadri(coal), Tanda and 1 to 16 hours for various units at Anpara, Obra, Paricha and Panki stations. At GGSSTP the time ranged from 2 to 9 hours. The utilities intimated that in case of some units LP diaphragms had burst during the occurrence, for which additional time was taken to rectify/replace the diaphragms.
viii. It was observed that after lighting up of the units, some of the units had taken longer time than others to synchronize with the grid. The observed time duration ranged from 2 hours to 4 hours in case of various generating units at Singrauli, Unchahar, Badarpur & Rihand TPSs and 2 hours to 9 hours for units at Anpara, Obra, Paricha and Panki stations. In case of gas based stations the time duration ranged between 1 to 6 hrs at Auraiya, Dadri and Faridabad for GTs and 3 to 7 hours for STGs.
On 30.7.2012, in DTL system GTs 1,2,5 were restored during 0250 to 0430 hours ( generation of order of 80 MW) and later GT3 was synchronised at 0640 hours (30 MW) with STGs 1,2,3 resuming generation between 0810 to 0840 hours ( order 60 MW). On 31.7.2012, in DTL system GTs 1,2,3,5,6 were restored during 1310-1445 hours (order 141 MW) and STGs 1,2,3 resuming generation between 1615 to 1646 hours ( order 51 MW)
Badarpur TPS had earlier survived and had operated in Island mode with units nos. 1,3 and 5 till 0658 hours on 30.7.2012 when the island collapsed and power was later extended to the station at 0710 hours from 220 kV DTL system at Sarita Vihar and unit no.3 was first synchronized at 1025 hours.
Faridabad GPS which had tripped at 0233 hours on 30.7.2012 was synchronized with grid at 0552 hours. It however tripped at 0658 hours with the collapse of Badarpur island. Faridabad GPS was later synchronized at 0844 hours with grid.
ix. While extending the power to Singrauli TPS, through HVDC Vindhyachal by pass route there had been tripping at Vidhyachal resulting delay in making start up power available to the station. The possibility of extending power to Singrauli from Pipri Hydro on 30th July in closer coordination between NRLDC, SLDC, UPJVUNL, UPRVUNL and NTPC would have resulted time-saving in affecting quicker start-up power to NTPC generating units. UPRVUNL suggested that Pipri-Hydel should be synchronized with Western grid through Vindhyachal-Shaktinagar-Anpara 400 kV line and 132 kV Anpara-Pipri line. This would ensure stability of voltage of Pipri machines and more machines of Pipri-Hydel could be started up.
x. In case of major hydro station, BBMB intimated that on 30th July 2012 the system was fully connected to grid by 0902 hours and on 31st July 2012 at 1553 hours. GGSSTP Ropar received start up power from Bhakra source at 0841 hours on 30th July 2012 and unit #2 boiler lighted up first at 1020 hours. The unit was synchronized at 1218 hours via Bhakra (220KV)-Ganguwal (220/132KV)-132KV Ropar-GGSSTP route and on 31st July 2012 unit # 2 boiler lighted up at 1640 hrs and synchronised at 1740 hours via Nalagarh(PG)-220KV Mohali-GGSSTP route.
xi. In ER, Kahalgaon and Farakka STPS received start up power from WR through Sipat-Ranchi and could not receive power earlier from hydro due to tripping of Teesta HPS on 3rd harmonic, over-voltages and under frequency and mismatch in generation with remote loads over long EHV lines. Despite multi-attempts of black starts at Tala, Chukha and Teesta,the startup power could not be extended to Farakka / Kahalgaon from the hydro sources.
xii. It was also brought out that start-up power could be extended to number of stations simultaneously so that stations could use them for preparatory activities like CW pumps, compressors etc. and actual start-up could be attempted after specific clearance from the source providing start-up power. This could considerably expedite the start-up as preparatory activities not needing much power could be taken up by number of stations simultaneously thus considerably reducing the start-up times.
7.3 Analysis of Restoration Process of Thermal Power Stations
Detailed analysis of start-up process for the grid disturbance of 30th July 2012 has been made so as to examine the restoration process and areas of possible improvements. The salient observations are given in subsequent paras:-
7.3.1 Availability of start-up power
Salient abstracts of the receipt of start-up power in the region are as under:-
Table 7.1: Availability of start-up power in Northern Region 30th July 2012
It may therefore be seen that more than 50 % of the affected stations in the region received start-up power after 3hrs. Only 2 stations in the region could receive start-up power within 1 hour. Also 8 stations received start up power after 4 hours. Maximum time taken for any station to receive start-up power was 6 hours for Ropar TPS.
The reasons for delay in receiving start-up power by most of the thermal power stations may require to be looked into from grid system point of view. A normative or bench mark time frame for extending start-up power to each of the TPS may be evolved by the RPCs in consultation with the constituents and RLDC so as to ensure that significant delays are not encountered in extending start-up power. A fact that emerged during discussions was that since the present grid failure occurred after almost a decade, the preparedness and response was perhaps not upto the level expected. More frequent need of having periodic mock exercises to ensure preparedness of all stakeholders involved as actual grid disturbances needs emphasis.
7.3.2 Restoration of Thermal Power Stations
Coal fired thermal power stations involve considerable amount of preparatory actions before actual start-up like operationalizing major auxiliary systems like circulating water (CW) system, compressed air system. Also start-up power is required to be provided to each unit and station auxiliary which involves charging up of number of transformers within the station sequentially and in turn is time consuming. In the above context, suggestions made at Para B above that start-up power should be extended as soon as possible so that stations could initiate preparatory activities and actual start-up process could be attempted immediately upon receiving start-up power.
It was also brought out by the stations that sudden tripping of the unit at high load lead to bursting of LP Turbine diaphragms in many of the units requiring replacement before start-up could be taken up and involved about 4 hrs for replacement of diaphragms for each unit.
7.3.3 Start-Up And Restoration Times
With a view to analyze the restoration process of thermal stations, data regarding time of availability of start-up power, time of taking up Boiler light up (BLU), time till synchronization and time of achieving full load were sought from the stations. The status of receipt of data from the stations is furnished below in Table-7.2.
Table 7.2: Receipt of restoration data from stations
As may be seen, while data has been received from most of the stations, the data from several other stations was not received and thus their restoration pattern could not be analyzed. The analysis for stations for which data for boiler light up was received have been made in respect of time of synchronization after BLU.
7.3.4 Initiation of Boiler Light Up
As brought out above, initiation of start-up of a coal fired station takes considerable time after receipt of start-up power due to preparatory activities involved. As the time for BLU have not been received from number of stations, the actual time taken for preparatory activities as also maintenance like replacement of diaphragms etc. could not be ascertained.
Details of BLU undertaken are furnished in Table-7.3. From the data on BLU available it is seen that there are considerable variations between the time taken for BLU of the first unit after receipt of start-up power.
Table 7.3: Time elapsed (Hrs) before BLU was undertaken after receipt of Start up power
Note:Timelines indicate total elapsed time before successive units were taken for start up and DO NOT refer to unit numbers
From the table it may be seen that few Stations like Obra and Panki undertook first BLU after 1 hour of receipt of start-up power. Ropar and Singrauli attempted first BLU after 01:40 hrs after receiving start-up power. Several Stations could undertake first BLU only after 2.5 to 3 hrs of receiving start-up power.
Also large variations are seen in undertaking further unit start-ups after taking BLU of first unit. The data of time elapsed before undertaking subsequent BLUs have also been analysed and presented in the Table-7.4. From the table it may be seen that while for some of the units the BLU was taken up within very short interval of 10-20 minutes of BLU of previous unit, in most of the cases the BLU for subsequent unit was taken up 2 to 3 hours after the BLU of preceding unit and in many cases exceptionally large time of 8 to 10 hours have been taken. The utilities were asked for the reasons for delay in start up of the units; however no reasons for delay have been furnished. Further discussions need to be undertaken in this regard.
Table 7.4: Time interval between successive BLUs
Note-Time for 1st unit indicates time taken after start-up power. Timelines for other units indicate time taken after BLU of previous unit.
7.3.5 Unit Synchronization after Boiler Light Up
The details of time taken for synchronization after BLU are furnished in Table- 7.5. Even from the limited number of Stations where data of both BLU time and synchronization time is available, it is seen that the time taken for synchronization after BLU varies considerably.
Table 7.5: Time taken for synchronization after BLU
As may be seen that the time for synchronization after BLU varies from a low of 1.2 hrs to as high as 3 to 4 hrs and even exceptionally high at 8 to 10 hrs for some of the units. Many of the NTPC units took 3 to 4 hrs for synchronization after BLU.
It may be mentioned that time taken from BLU to synchronization is expected to be fairly comparable for the units of similar design with similar start up regimes and thus such large differences in timelines are not understood. Details of problems/constraints encountered during start-up process and delays occurred may require to be looked into further in respect of constraints bottlenecks faced in this context.
7.3.6 Time from start-up power to Unit Synchronization
Since timeline for undertaking BLU were not made available by most Stations, an analysis of total time taken upto synchronization from receipt of start-up power has been made to understand the trend that emerged. These timelines however would be indicative of combined impact of constraints/delays occurred in preparatory activities (before BLU) and during the start-up process. The details of time taken for synchronization after start-up power are furnished in Table-7.6
Table 7.6: Total Time S.U Power to Synchronization
Note: Timelines indicate order of synchronization of units and not unit numbers
The variability seen here is similar to the variability seen in timelines from start-up power to BLU and BLU to synchronization – rather the variability seen here is much more prominent. Amongst the 210 MW units, Kota was the first station to achieve synchronization within 2.55 hrs from start-up power followed by Ropar achieving synchronization of one unit in 3.19 hrs. Singrauli achieved synchronization of first unit in 3.44 hrs. Rest of the Stations achieved first synchronization beyond 4 hrs and some Stations like Paricha and Anapara could achieve their first synchronization in 6 to 8 hrs. Badarpur could achieve synchronization of its 210 MW unit in 8 hrs.
Amongst the smaller size units, Tanda and Rajghat TPS achieved their first synchronization in 2.10 hrs after start-up power whereas other Stations like Badarpur and Panki took 3 to 4 hrs to achieve first synchronization. Obra achieved synchronization of 50 MW unit in 6.30 hrs though it was the second synchronization for the Station. The lignite fired Circulating Fluidized Bed Combustion (CFBC) units had taken exceedingly long time to achieve their first synchronization.
Further, analysis of time taken for subsequent synchronizations have also been analysed and presented in the Table-7.7. Here again similar large variability is seen. Ropar and Kota TPS achieved most rapid successive synchronizations with second and fourth synchronization at Ropar in 18 and 15 minutes and second and third synchronization at Kota in 48 minutes and 05 minutes, however the subsequent synchronizations took longer. Amongst the smaller size units, Tanda TPS achieved rapid second synchronization in 35 minutes.
Table 7.7: Time taken for successive Synchronizations
Note-Time for 1st unit indicates time taken after start-up power. Timelines for other units indicate time taken after synchronization of previous unit
The large variations in synchronization times and successive synchronization times with many of the stations achieving timelines far better to others is indicative of the potential improvements possible in most of the stations with better/faster preparation.
7.3.7 Gas Based Stations
Gas fired stations are looked upon as rapid source of power after such grid disturbances as preparatory activities required are far less and start ups of Gas turbines (GTs) are fast. Details of time taken for synchronization of GT Stations a re given in Table-7.8.
Table 7.8: Gas fired stations - Time taken from start-up power to Synchronization
Note: Cells coloured yellow indicate Steam Turbine unit
It may thus be seen that even the Gas turbine units have taken unduly long time for start-up. While the first Gas turbine units could come up at Anta and Auraiya in about 50 minutes, the Gas turbine units at most other stations and even the subsequent Gas turbine units at these stations took far longer time of 2 to 4 hours. Such large start-up times for Gas turbine units need to be looked into with a view to make Gas turbine units a dependable source of rapid restoration power.
7.4Conclusions and Suggestions
Based on the discussions brought out in the foregoing paragraphs, the suggestions for faster restoration of thermal power stations are as under:-
1. A well coordinated and documented process for supply of start-up power may be put in place under the overall coordination of NLDC clearly bringing out the following:
a. Existing Black start procedures should be frequently reviewed in line with the fast changing grid scenario and addition of generation capacity. The facilities available with existing and upcoming IPPs
should also form part of these procedures for the purpose of extending start up supply to black starting units in the vicinity.
b. Explicit instructions to all stake holders to supply start-up power without any commercial considerations that could be settled later.
c. Micro level load management for lines and loads may be envisaged for extending start-up power through pre-defined arrangement for availability of loads under emergency conditions so as to avoid frequent tripping while extending start-up supply.
d. Authorizing Load dispatch centres to advise action to the concerned utilities for extending power supply immediately to the black-starting units through exchange of special emergency code between the concerned load dispatch centres.
e. A normative or bench mark time frame for extending start-up power to each of the TPS. This may be evolved by the RPC in consultation with the constituents and RLDC so as to ensure that significant delays are not faced in extending start-up power.
2. There is a need to strengthen communication and have a dedicated communication network between SLDCs and all power plants in the respective control areas. The mobile phone facility presently used, is not considered reliable for such communication. Reliable and efficient communication facilities at all active installations connected to the grid is essential to ensure faster restoration.
3. Key installations/sub-stations should be managed by experienced manpower particularly during odd hours – like Qualified operating personnel having undergone orientation courses under certification programme.
4. A system to extend start-up power to number of stations simultaneously for preparatory activities could be considered. The actual start-up could be attempted after specific clearance from the source providing start-up power. This could expedite start-ups as preparatory activities not needing much power could be taken up by number of stations simultaneously.
5. Large variations are observed in time taken for initiation of unit start up (Boiler light up) by the stations after availability of start-up power and also for start ups/light up of subsequent units. Also, while subsequent start-ups were very fast (10-20 minutes) in some of the units, in other cases they took considerably longer – several hours.
Reasons for the delays in attempting first start-up and subsequent start-ups may be examined by the utilities in consultation with CEA. A standard procedure for preparatory activities and sequence of start up may be put in place by the stations to restore units as early as possible.
6. Large variations have also been observed in time taken by the stations for synchronization after Boiler light up which are normally not expected. Detailed analysis of times taken especially from BLU to synchronization may be made by the utilities in consultation with CEA so as to identify constraints/bottlenecks faced during unit start-up for remedial action.
7. The large variations in synchronization times and successive synchronization times achieved with many of the stations achieving timelines far better to others is indicative of the potential improvements possible in most of the stations with better/faster preparation. Further, in most cases the start-up times appear to be considerably higher than the manufacturers/OEM recommendations. Optimal start-up/ restoration procedures by the stations in consultation with OEMs.
8. Long start-up times taken by most Gas turbine units need to be investigated to develop Gas turbine units as reliable source of fast restoration power.
