The whole state of Sarawak plunges into a state of total darkness yesterday. On first instance, we could see activity on Social Media asking if there is a major power failure or not. This follows by scattered info in when the power will come back and at the 3rd hour, some cursing and ranting has go one on Social Media. By midnight, all the fingerpointings starts.
A little bit on why there was that major blackout. Firstly, all our electricity supply are channeled into this big network called a state grid. All the major power generations are done at Bakun, Batang Ai, Sejingkat, TAR, Bintulu, Miri and Mukah, each supplies different amount of loads to the grid,
The supply is then channeled into distribution zone/segments which can be linked and unlinked into the grid. These segment is what that is connecting our premises to the supply.
During the fateful moment, the power frequency drop occurred in one of the Bakun’s supply causing an imbalance to grid. The nearest power generator from Bakun – Kemena Bintulu, was the first to trip to prevent any damages to the generators. This causes a large chunk of supply lost and created a higher demand than supply scenario. Technically, this is called a cascading failure.
An analysis on the sequence of event, Sarawak Energy first reported a trip in the transmission line to the generator in Bintulu. Here is what causes it. The drop of frequency from the power generation from Bakun causes the grid to have poor quality supply. The grid system will automatically rectify the balance from the zone by drawing more load from the area, The reason why Bakun’s line was not detected down in the first place was because the load detection was measured by the Ampere while the grid monitors based in the load in the same bandwidth of 50 Hz. Grid system have lost synchronicity with Bakun and the corrective measure causes all the load to be leveraged on the next to nearest generator – Kemena-Bintulu. Due to the load been higher than what the generator can supply, Bintulu-Kemena line was tripped to prevent damages to the generator.
With Bakun and Bintulu out, there is a shortage of over 1000Mw from the grid which causes load shift to other Power Generation line to trip in the same manner to how Bintulu has tripped. Hence a Cascading failure.
A lot of people blatantly blasting at Sarawak Energy and the Government on this failure.
Now, Let’s take a look at recovery.
In order to fully recover, Each and every one of the generators need to be shutdown and reconnect into the grid one by one. Starting with Batnag Ai, Sejingkat, TAR, Bintulu, Miri and Bakun. Once the grid has supply to it, supply segment is connected into the distribution grid one by one making it a pain staking process. This is done in sequence whereby area nearest to the grid node is restarted first. This is followed by Intermediate and Secondary intermediate and the rest of the sequences.
A grid the size of Sarawak will take between 2-4 hours minimum to restart. This is considering that there is no confounding factor to the scenario. Yesterday was not the best scenario and The problem with Bakun that was serving more than half of the grid load (56.8%) at the time. Even if we load all the rest of the station to the theoretical 100% load/supply ratio will also be insufficient to rectify the load factor. It is only at after 10pm when Bakun has restarted and reconnected into the grid that allows almost complete restorations to the state grid. Prior to Bakun coming online, the grid was split  into two-island with one at the north and another south. The casualty of this is the disconnection of Central grid which supplies to major part of Sibu. This is done to balance the load factors from the North and the South.
The reason why there is a frequency drop is because our grid being a single lined grid. The state grid has no redundancy and load balancing from the point of supply. It seems that this is the lesson for the supplier, in this case, Sarawak Energy and the State Government (owner of Sarawak Energy) to invest into it. Have we got the secondary feed on the grid the drop frequency would not be that significant and will not cause the cascading failure to the state grid, theoretically speaking.
So, before you start blasting them on the inefficiency, please understand how much work they have to do before aiming your arsenal.
*** This is a personal opinion and analysis. Reader discretion is advised.
15 responses to “Sarawak Blackout – A cascading failure.”
Christina Law liked this on Facebook.
Peter Tan liked this on Facebook.
Nick Alfred Jaffer liked this on Facebook.
Jenny Sim liked this on Facebook.
Thank you pierce for a good analysis, only one comment, the Bakun plant have black start facility, apparently it still took some time to utilize this. All remote HEP, should and normally will have BS facility, because of the lengthy transmission systems they are connected to and hence, back feeding power to these remote location normally is challenging in a system failure mode. Thank you. Torstein
Great article, Pierce. I am puzzled by a few things. Why were there no risk analysis or risk management done to ensure synchronicity is not an issue? Why would loads be detected with two different systems separate from each other? Why was there no isolation protection system in place to prevent a cascading effect? Isolation Protection allows fault areas to be isolated promptly without pulling everything down with it. I see it also as a failure of the Power Company to properly manage the risk and have systems in place to mitigate and isolate potential problems downstream before cascading effects happens. Other countries are able to do it on their systems and its nothing new. Your thoughts?
Cool. Now i know a bit more about the facility. Like I’d said, my info is outdated.
Pete, there was studied done, but sometimes it takes time get through the head of some people.
Dear Pete, the synchronization of two systems is a difficult process, because voltage and effect as well as I believe frequency, must be very close in the two separate system to be able to connect them otherwise you create another tripping. So to synchronize south and north demand careful management of supply and demand as well as many other factors to align the two systems before you connect. As you know our system operate at 275 000 V so you can simply not connect without a very controlled operation. Thank you for your interest. My background is not electrical, so I am saying this as a layman, but I believe the main understanding is correct. Hope it helps. Torstein
Pete, the drop is not n the voltage, but rather on the frequency. If my memory serves me right, the load was approximate to 40% load on Bakun generator, and a drop of 2.5Hz been registered. On bandwidth calculation, For load factor 100%, if the frequency dropped by 1 Hz, you will get a 50% drop in bandwidth(wattage) produced. Based on this rule, at a load factor of 40% and 2.5Hz drop will also bring the bandwidth down by the same ratio. There was no trip because the voltage did not drop. Trip detection is usually built checking voltage change. Under the unique circumstance, voltage drop didn’t occur. That should be my theory on the cause of the problem that is liken pulling a handbrake while the car is moving on motorized motion. The jolt could stalled the engine completely.
An isolation protection system should be able to detect both the difference in frequency and voltage not just the voltage…and in the process provide prompt isolation of fault area. These systems have been around for years and are nothing new. Not sure why they rely only on voltage drop for trip detection. Kind of odd to say the least.
Dear Peter, I may misunderstand your point, but just to be sure our message has been clear, the total tripping of our system came from drastic drop in the supply of power whilst the customers still were taking load, this lead to drastic drop in frequency in the whole system and, to protect the remaining generators from serious damage they are auto shut down when frequency drops below 47.5. Then also the north south system split/tripped, but this had no effect on the black out because the total system was already almost with out supply due to the earlier mentioned auto shut down of all capacity. Torstein
Torsten, thank you for your explanation. What puzzles me is why the 7 generation areas shut down rather than isolate themselves from the main grid due to the imbalance. This way the Bakun generation has a chance to correct the frequency shift and synchronization can kick back in after.
Again I must admit I am not an expert, but are quite sure of the following: we can not isolate the generators from the grid without shutting it down. The unit must deliver the power when it runs and the grid is the only outlet for the power, hence we are isolating it to protect the units, but that means shutting them down also. Thanks Torstein
Wow, this is absolutely informative. Thanks for such great article which enables me to have greater understanding about Bakun.
However, may I ask if… Bakun Dam has proven to be the best source for electricity for Sarawak ?