Optimization of Prefill Mode to Increase Reliability of Gas Turbine 13E2 PT. PJB UP Muara Tawar

IOSR Journal of Mechanical and Civil Engineering (IOSR-JMCE). e-ISSN: 2278-1684,p-ISSN: 2320-334X, Volume 14, Issue 3 Ver. VII. (May - June 2017), PP 13-18. www.iosrjournals.org

Abstract: Muara Tawar block V is constructed by PT. PLN (Persero) Power Plant Master Project ofJava, Bali and Nusa Tenggara Networks based on contract No. 261. PJ / 041 / DIR / 2007 (Concerning Gas-Fired Power Plant Extension Project). The plant uses gas turbine type 13E2. The operational concept of gas turbine type 13E2 is regulating the amount of fuel that is divided from three fuel control valves and the operating pattern is in an area known as the pilot valve and premix valve which must be maintained the stability of combustion shown by pulsation parameters. The prefill concept is used to avoid flame off when gas turbine loading often passes over point switch or back point switch. Prefill itself will momentary activate a pilot valve to fill the Fuel Distribution System (FDS) line from MBP43. However, prefill gives the effect of high pulsation that triggers Gas Turbine experience PLS or derating. This effect can be overcome by improving the prefill concept and modifying the prefill system through Logic Advant, resulting in a more stable burning in the gas turbine burner.

Keywords: Gas-Fired Power Plant, FDS, pilot valve, pulsation, prefill.

I. Introduction
The utilization of gas turbines in power generation has increased in recent years and it will be increased up to a medium level of application [1]. Muara Tawar Block V gas turbine system was designed to operate at peak or under load continuously in both open and combined cycles as per network requirement. Muara Tawar Block V power generation can generate 242.5 MW electricity (gross output). It comes from the gas turbine generator alone about 161.5 MW and from the steam turbine is about 81 MW [2]. The reliability of a power plant to
provide electricity to the electricity network is the most important task to ensure energy availability [3]. Gas Turbine type 13E2 consist of three fuel control valves to maintain mass and energy balances in the burner system under the pilot and premix valve. To maintain stable combustion, changes in the operating condition between the pilot valve and the premix valve are required and it's known as switch over and switch back point. During these conditions, the pulsation parameter should be monitored. If the gas turbine operates in high load or above switch over point continuously, it tends pilot valve MBP43 cannot provide sufficient gas fuel to the combustion system. Moreover, during the load decrease, the operating condition will change to a switchback point. It condition leads to mass unbalance and it can be very dangerous because the pilot valve must close immediately. Moreover, in this condition, the pilot valve status is closed due to operation in high load. This condition triggers flame off resulting in the unit trip. The prefill concept is used to overcome this hazard and unscheduled shut down by the pilot valve in the MBP43 area pulse opening. However, prefill can also affect the emergence of high pulsation that resulted in gas turbines experiencing PLS or derating.

II. Theory
Prefill is used to reduce the disturbance that may arise due to the changing of electricity load from high to low load and vice versa. Prefill is used to fill up the volume between Fuel Distribution System (FDS) to the burner [4]. This prefill is performed by opening the pilot valve or MBP43 at a certain time to ensure the volume of the FDS until the burner fills up with approximately 20% of additional fuel. Mass unbalance occurs when the load is over the SwitchBack Point (SBP) area. Therefore, the prefill on the FDS pilot will not cause the flame off when SBP is present.

The prefill will be activated when the decrease in electricity load occurs. The prefill system will open the MBP43 control valve from the minimum mass flow rate and increase up to the maximum mass flow rate gradually to fill up the volume between FDS to the burner. The gas turbine at Muara Tawar Block V mass flow rate was designed at 0.285 kg/s in maximum load and 0.250 kg/s in minimum load [5]. The prefill process will stop if any of the criteria are reached i.e. Delta TAT (Temperature After Turbine difference) reaches 5 deg C or high pulsation reaches 35 mbar, or maximum prefill time has reaches 10 seconds.

III. Method
3.1 Analyze DEPP Prefill System of Gas Turbine
Logic prefill system is simulated under Advant Egatrol 8 software. Simulation on the logic includes the operation of pilot valve MBP43 and the operating pattern of the prefill system.

Fig. 2. describes prefill system, where prefill will be enabled or start (PltPrfSqPrel_START) with 2

modes i.e. periodically time (PltPrfSeqCyc_START) or decrease in load. Every 4 hours, the gas turbine will experience periodic prefill at any load that causes interruption to Protective Load Shedding (PLS). Interruption to PLS by high pulsation is caused by an over mass flow rate from MBP43 or pilot valve.

Fig. 3. shows the high pulsation which resulted in PLS 148 mbar. High pulsation occur because MBP43 

(number 4) is activated and mass flow rate or gas fuel flow rate of 0.250 kg/s up to 0.285 kg/s during 10 seconds. This logic set can be observed in Fig. 4. Periodic prefill is set up for every 4 hours which often results in high pulsation. Total mass flow rate can be obtained in two ways i.e. the mass flow rate on the scale is too large or the setting time is too long for each prefill process.

3.2. Fix the Prefill Function on the Operational Concept from the Advant logic side
Based on the investigation of the logic prefill compared to the DEPP data trending it is necessary to make some parameters modification that will affect the prefill process as well as Prefill operation pattern. Improvement of the prefill logic is done in several stages. The first stage is to modify the maximum time for each prefill (TIME_MAX_PRF_C1). The maximum time for prefill is gradually lowered and applied directly when the gas turbine operates by monitoring DEPP trending data continuously.

Based on Table 1. It can be seen that a decrease in setting time will reduce the pulsation that occurs in the burner. However, the maximum limit of changes is shown by the absence of diff TAT. Diff TAT indicates that the prefill has perfectly filled the FDS. The maximum prefill time decrease gradually and reaches the optimum value of 7 seconds. However, a change in time setting is not the best choice because the gas turbine is still experiencing PLS due to prefill for a while. The next step is decreasing the pilot valve mass flow rate value in logic prefill shown by Fig. 6. The mass flow rate value decreases gradually while pulsation pattern and changes in TAT are continuously observed.
Changing of pilot valve mass flow rate parameters compared to the pulsation values can be observed in Table 2.

Any decrease in mass flow rate will result in a decrease of high pulsation value. However, the maximum limit of changes is indicated by the absence of diff TAT. Change in mass flow rate value obtained minimum high pulsation at the point of 0.2 kg/s for both minimum and maximum mass flow rate.

IV. Results And Discussions
Prefill system on turbine gas will help avoid flame off when there is a change in the area of switch over or switchback point. However, the active prefill will cause high pulsation so PLS or Protective Load Shedding will be active. PLS will cause turbine gas load to drop until a normal condition is reached and PLS is reset.

In Fig. 7. There is an active PLS signal from signal number 7 that is 51MBX41EA000_XU03 PLS 2oo3 or PLS 2 out of 3 active signals. PLS is active because of signal high pulsation number 9, 51MBM30AX010Breach 148 mbar. High pulsation reaching 148 mbar would cause PLS. PLS always happens when prefill process takes place. The total event for 2012 is 52 times starting from February 10, 2012 and the last event is November 7, 2012. High pulsation results in PLS because Prefill occurs in an area of about 60% to 65% Relative Power or approximately 112 MW. The operating load in 112 MW is the demand from the network at certain times. However, if prefill operates at a relatively high load (above 120 MW), then the prefill process will not cause high pulsation because the proportion of premix is so large that the effect of the prefill is relatively small. Using this information, logic improvements are needed by modifying the enabled function
PLT_PREFIL_CYC_EN, so when turbine gas operates at a load the cyclic prefill will not be active. This function can be activated when a gas turbine operates with a relatively high load. When under load, the prefill will be active when the gas turbine experiences a minimum load loss of 0.35 MW/s for 10 seconds. Fig. 10. show prefill process after modifying function has been done. Prefill occurs when the Gas Turbine experiences a decrease in load from 143 MW to 113 MW within 2 seconds or a decrease of 15 MW per second. This prefill occurs optimally by being marked by a small power swing and no increase in pulsation value.

V. Conclusion
Fixing the prefill function from gas turbines can reduce and eliminate PLS due to High Pulsation. Prefill function improvement is done by disabling cyclic prefill function under load. With the improvement of prefill function, there is no more derating because of prefilling. Despite the improvement of the prefill system to the operational concept of gas turbine, monitoring of combustion parameters must continue.

Endrik Purbo Yunastyo1, Danan Tri Yulianto1, Asnawadi Hidayat1,

Kevin Sanjoyo Gunawan2, Totok R. Biyanto2*

1PT. PJB UP Muara Tawar, Muara Tawar, Indonesia

2

Engineering Physics Department, Institut Teknologi Sepuluh Nopember (ITS), Surabaya, Indonesia.

Implementing Risk-Based Maintenance Strategies for Distributed Control System as Power Plant Asset Management

doi:10.1088/1757-899X/1096/1/012108 

Abstract. The electricity generated from the power plant is subject to several requirements for active power, voltage, and frequency according to the grid system, so the machine must be controlled to achieve requirement by a power plant control system known as Distributed Control System (DCS). DCS system in block 1 Muara Tawar power plant using Procontrol P13 for gas turbine and Procontrol P-14 to control steam turbine. These systems had operated since 1997 and had been nearly operating for 23 years until now, and several failures tend to increase from time to time. The failures of the DCS and the lack of control cards will result in the loss of production. DCS system assets must handle properly to maintain the overall reliability of the power plant system. A method and strategy to maintain DCS must be carried out and ensure reliability and risk always under controlled conditions. Implementing Risk-based maintenance by carrying out quantitative calculations through the reliability approach and the level of the consequence of failure to calculate equipment risk is one of the methods of the DCS system. The result of the Risk-based maintenance method shows the highest risk on the risk map of the DCS system was in cubicles 14CBA02, 14CBA03 with a high-risk level and gas turbine and HRSG cubicle in medium-high level. The interval preventive maintenance time calculated by reliability within a year showed that cubicles 14CBA02 and 14CBA03 suggested being maintained every 29 days to reduce possibilities of failure.

        The number of DCS failures both cards control and monitoring system (human-machine interface) based on record during 2016 to mid-2020 shows the Procontrol DCS system contribute failures and result in operations and production decreasing. Based on data, DCS Procontrol P13 and P14 manufacturers in 2015, show that the card control and the whole system are in the classic, limited, and absolute phases. The classic phase to absolute has a span of 10 years. This situation brought a major issue on the availability of spare cards, especially how to keep and maintain the reliability of the overall DCS system. 

        According to card assessment, the majority of cards are in the classic and limited phases. This happened in all GT blocks 1-2 and ST 14 that used P13 and P14 Procontrol systems. DCS card failures data according to the type of card control can be shown in figure 3 the Pareto diagram according to failure of each type of card. Maintenance of DCS systems is an asset management challenge, especially on how "card control" management takes part in the situation where the availability of spare cards is starting to be limited and it will greatly affect the operation, production and at the same time the safety factor of the plant. A maintenance pattern is needed to accommodate and maintain the reliability of the DCS system. The ability to maximize the capabilities and reliability of DCS is an important asset to improve the power plant performance and profitability. 

        This research seeks to formulate a method of maintenance based on DCS’s Risk-based Maintenance and is expected to answer some maintenance problems in the DCS system then classify the possible causes of the failure of the DCS system. This research also shows how much the consequences arising from interference are, how likely the occurrence of the interference is, how big is the risk of DCS failure, and how to plan a DCS system maintenance strategy.

        Data analysis is carried out following the steps in the preparation of equipment maintenance strategies related to the DCS in its cubicle system that uses Risk-based maintenance. The following steps are: 

Classification of DCS generating systems based on cubicle according to Pareto chart that reflected the number of failures. 

• Determine the scenario of failure by analyzing the possible causes of disruption of the DCS system.     
• Determine the evaluation of consequences in a semi-qualitative way by determining the risk criteria for ranking loss of performance of units, financial, ecological, and health and safety so that the total consequences result from a system failure. The four factors are then combined to produce a                 total assessment of the consequences of equipment failure on the system and formulated with Consequence = [0.25A2 + 0.25B2 + 0.25C2 + 0.25D2 ] ^0.5 
• Conduct probabilistic analysis with precise distribution in accordance with the characteristics of the failure interval data so that the reliability value can be found as a basis for calculating the failure probability of the DCS system. The reliability value can be calculated using the equation.
• A risk assessment by combining the consequences of risk and analyzing the probability of failure that produces an acceptable or rejected value. At this stage, a matrix of risk is drawn to describe the position of DCS system risk.
• Based on the risk level and duration of operation of the DCS system, the most appropriate type of maintenance interval can be determined to ensure the reliability of DCS system. Evaluation of maintenance by making comparisons the maintenance process before risk-based maintenance is carried out and after this method is implemented. The reference of reliability value must not be less than 80% or 0.80 [8]. The results of the comparison were then plotted into a graph to find out the relationship between DCS reliability values and operational time. 

       The risk mapping of DCS system based on each cubicle in Muara Tawar Block 1 power plant from likelihood risk identification and consequences based on the failure report then visualized in a risk matrix (figure 12) as part of DCS system risk mitigation process in case of failure prevention that has an uncontrollable impact. The DCS system that is classified as high risk is 14CBA02 and 14CBA03, which are the system that handles the control of the steam turbine system. Meanwhile, the medium-risk were 11CRC30, 12CRC30 cubicle that handles the control of gas turbine and 12CBA11 cubicle handles control of HRSG 12. Risk control is carried out by making recommendations for maintenance and modification based on the level of risk and the likelihood of damage to the DCS system in a cubicle.

    

    Maintenance planning is based on the level of risk and the time interval for DCS system failure so that the most appropriate type and maintenance intervals can be seen and the DCS can be maintained. Based on the calculation of the decline in reliability values obtained maintenance time intervals that have been recapitulated in Table

    Determination of the maintenance interval of the DCS system on the cubicle with a critical level is crucial if it were done in a period based on the failure rate from time to time. Spare part readiness can help maintain DCS assets as an important part of a power plant.

Implementing Risk-Based Maintenance Strategies for Distributed Control System as 

Power Plant Asset Management

D T Yulianto, R M Isman, S N Ihsan and H G Susanto 

Pembangkitan Jawa Bali UP Muara Tawar, PLTGU Muara Tawar 1 street, West Java, Indonesia Corresponding email: *danantriyulianto@ptpjb.com

Analisa Low Pressure Feed Water Pump (Lp Pump) Dengan Metoda Failure Mode Effect Analysis Dan Fault Tree Analysis

 1.       Pendahuluan

1.1.  Overview Muara Tawar Blok 1

PLTGU Muara Tawar merupakan pembangkit yang dikelola oleh PT Pembangkitan Jawa Bali yang terdiri dari 5 blok, blok 1 3 Gas turbine dan combine dengan 1 steam turbine, blok 2, 3, dan 4 beroperasi masih secara open cycle, dan blok 5 dengan 1 gas turbine dan 1 steam turbine secara combined. Blok 1 yang memiliki 3 gas turbine dicombined dengan memanfaatkan sisa gas panas yang memiliki termperatur berkisar 520  degC – 550 degC (mengikuti pembebanan) untuk membangkitkan uap untuk dimanfaatkan sebagai penggerak steam turbine yang terdiri dari 2 stage yaitu sisi HP turbin dan LP turbin.

Salah satu peralatan yang berperan dalam system combined cycle ini adalah Low Pressure Pump (LP Pump) yang berfungsi untuk mentransfer kondensasi uap setelah steam turbine (dari kondensor ke hotwell) yang menuju ke LP drum. LP feedwater pump terdiri dari 3 pompa dan ketika beroperasi combined cycle dengan 3 gas turbine maka akan dioperasikan 2 pompa, dengan salah tidak dioperasikan atau standby. satu Kegagalan fungsi LP pump akan mengakibatkan berhentinya operasi system Heat Recovery Steam Generator (HRSG).

 1.2.  Studi Kaus

LP Feed Water Pump baik nomer 1,2 dan 3 terdapat indikasi adanya vibrasi dari pengecekan secara berkala yang mengharuskan dilakukan perbaikan dari sisi pompa, akan tetapi vibrasi tidak menunjukkan indikasi penurunan. Terdapat 5 komponen utama yang akan dilakukan analisa yang terdiri dari;        

Laporan gangguan LP feedwater pump yang tercatat dari sistem Ellips adalah sebagai berikut;

1.    2.    FMEA

FMEA merupakan salah satu metode pengukuran resiko peralatan dengan mengidentifikasi mode-mode penyebab kegagalan yang ditimbulkan oleh setiap komponen terhadap suatu system dalam hal ini komponen pompa LP feedwagter pump. Penilaian terhadap komponen-komponen ini akan menghasilkan suatu penilaian yang dikenal sebagai risk priority number (RPN). Standar penilaian menggunakan pendekatan severity, occurrence dan detection, dimana severity adalah dampak yang timbul apabila suatu kesalahan (failure) terjadi, occurrence adalah kemungkinan atau probabilitas atau frekuensi terjadinya kesalahan dan detection adalah kemungkinan untuk mendeteksi suatu kesalahan akan terjadi atau sebelum dampak kesalahan tersebut terjadi. Nilai tiap pendekatan terdapat di table 1. 

Hasil analisa FMEA dihasillkan table sebagai berikut

Nilai RPN tertinggi dari sisi base plate dengan nilai 75 yang dihasilkan dari nilai severity sebesar 5 yang menenunjukkan efek moderate dari kegagalan, occurance total kejadian lebih dari 10 dalam 3 tahun terakhir, akan tetapi untuk kejadian kegagalan yang diakibatkan base plate terjadi sebanyak 3 kejadian (dari 3 LP feedwater pump). Sedangkan detection 5 karena kegagalan hanya bisa dilakukan deteksi awal dari suara.

3.  FTA

Fault Tree Analysis adalah analisis diagram terstruktur yang mengidentifikasi elemen-elemen yang dapat menyebabkan kegagalan sistem. Teknik ini didasarkan pada logika deduktif dan dapat disesuaikan dengan identifikasi risiko untuk menganalisis bagaimana dampak risiko yang muncul. Berdasar studi kasus LP feedwater pump dapat diturunkan diagram fault tree analysis dengan analisa berdasarkan 5 komponen utama, yaitu;

4. Kesimpulan

Failure Mode Effect Analysis dan Fault Tree Analysis adalah metode pengukuran resiko peralatan dengan mengidentifikasi mode-mode penyebab kegagalan yang ditimbulkan oleh setiap komponen terhadap suatu system. FMEA mengukur resiko peralatan berdasarkan penilaian terhadap severity, occurrence dan detection. Penilaian tertinggi (RPN) dari baseplate dengan nilai 100. Sedangkan Fault tree analysys adalah metode mengukur resiko kegagalam peralatn dari komponen-komponen yang dihitung nilai probabilitas kegagalan komponen penyusunnya terhadap komponen utama.

 

LOADING PROCEDURE ON ADVANT CONTROLLER

"Postingan lama"

AC 160 Advant Controller 

A. Single Controller – ON LINE 

1. Must be on ON-LINE condition 

2. Only to change the parameter, maximum 3 changes 

3. Make sure the changing didn’t effect on operation or protection side 

4. Step 

a. Make change on parameter/variable 

b. Save  save and load

B. Single Controller – OFF LINE 

1. Create/generate new target code 

2. Create new signal 

3. OFF-LINE condition 

4. Steps; 

a. Make change on logic 

b. Save then Generate target Code 

c. Follow the step on this Target menu (mean while keep open the status report status menu to watch the process); c.1. Block Program 

c.2. Load Application 

c.3. Save in PROM 

C. Dual Controller – Slave must be Updated 

1. Check the connection between master and slave controller 

2. To activate the slave  kick out the P1 (master) by force switching 

a. Click Reset on P1  b1 will come 

b. Acknowledge on b1  P1 will come 

c. And the configuration became  

d. Load the controller as usual 

e. Reset the P1  bf will come 

f. Acknowledge the b1  P1 will come 

g. And the configuration will be 

h. To update the slave go back to EDS and activate the update slave menu *not necessary since it                 can be done by hardware updating 

i. To update the slave by hardware updating, pres ACK on bf for about 3 seconds until the P.4                     comes.  

j. If appear then remove the ACK and bf will be b1 for the next 4 minutes.

 

Menurunkan Kegagalan Start Gas Turbine Dengan Metode Lean Six Sigma di PLTGU Muara Tawar Blok 1

PLTGU Muara Tawar sebagai unit peaker (yang beroperasi pada beban puncak), dioperasikan dengan menggunakan pola start-stop, sehingga dituntut mempunyai kesiapan dan keandalan yang tinggi untuk menjamin ketersediaan tenaga listrik. Bidang pemeliharaan bekerja untuk menunjang operasional unit pembangkit guna memenuhi kontrak kinerja yang telah dibuat antara Unit Pembangkitan dan pelanggan (dalam hal ini PT. PLN Persero). Pemeliharaan dalam pembangkitan berfungsi untuk menjamin peralatan dapat bekerja dengan baik, sehingga unit bisa beroperasi dengan maksimal dan menghasilkan keuntungan bagi perusahaan. 

Pemeliharaan meliputi segala aktifitas yang terlibat dalam peralatan dengan aturan kerja yang berlaku. Aktifitas pemeliharaan pada unit pembangkit bertujuan untuk mengembalikan performance mesin, memperbaiki efisiensi, meningkatkan faktor ketersediaan (Availability Improvement) dan meningkatkan keandalan (Reliability Improvement). 

    

        Data kegagalan start tahun 2018-2019 dalam tabel  7 menunjukkan jenis kegagalan start yang paling dominan adalah gagal start DN/DT atau pelambatan percepatan turbin (low acceleration) ketika proses start up.  DN/DT atau low acceleration merupakan bagian proteksi turbin untuk mencegah kerusakan material hot part (blade turbine) akibat temperatur pembakaran di awal start lebih dari 450°C.

        Setelah dianalisa dengan menggunakan rumus Pareto, gagal start berpengaruh 50% lebih dari jumlah gangguan yang terjadi di tahun 2018-2019 di Blok 1. Sehingga untuk menjaga nilai kesiapan dan keandalan unit sesuai dengan kontrak kinerja, perlu dilakukan optimalisasi tindakan preventif untuk mendeteksi lebih dini gangguan serta mencegah kegagalan pada sistem atau peralatan-peralatan yang mengakibatkan kegagalan start. 

        Untuk mengatasi permasalahan gagal start, maka harus dipilih metode yang sesuai dengan operasi unit pembangkit. Six sigma adalah suatu metodologi sistematis yang berfokus pada faktor kunci yang mengendalikan performansi suatu proses, mengaturnya pada tingkat yang paling baik dan menjaganya agar tetap berada pada level tersebut. Lean adalah suatu metodologi sistematik untuk mengurangi kompleksitas dan melancarkan proses dengan mengidentifikasi dan mengeliminasi sumber dari pemborosan (waste) dalam proses, karena pemborosan bisa mengakibatkan macetnya aliran. Pendekatan Lean bertujuan untuk menghilangkan pemborosan, memperlancar aliran material, produk dan informasi serta peningkatan secara terus-menerus. Sedangkan pendekatan six sigma untuk mengurangi variasi proses, pengendalian proses dan peningkatan terus menerus. Integrasi antara Lean dan Six sigma akan meningkatkan kinerja melalui peningkatan kecepatan dan akurasi (zero defect).

Tahapan Six sigma     Lean six sigma secara proses terdiri dari beberapa fase yang dikenal dengan istilah DMAIC (Define, Measure, Analyze, Improve, Control).

1. Define adalah fase menentukan masalah, menetapkan persyaratan-persyaratan pelanggan dan mengetahui CTQ (Critical to Quality).
2. Measure adalah fase mengukur tingkat kecacatan produksi maupun proses.
3. Analyze adalah fase menganalisis faktor-faktor penyebab masalah/cacat.
4. Improve adalah fase meningkatkan proses dan menghilangkan faktor-faktor penyebab cacat.   
5. Control adalah fase mengontrol kinerja proses dan menjamin cacat tidak muncul.

detail analisa dan hasil rekomendasi ada di buku ini, sila mampir perpustakaan unit muara tawar.