Hasil Pencarian  ::  Simpan CSV :: Kembali

"PREDIKSI PENYEBARAN LATERAL POROSITAS DARI ATRIBUT SEISMIK MENGGUNAKAN ARTIFICIAL NEURAL NETWORK Abdullah Nurhasan*, Abdul Haris* ABSTRAK Atribut seismik merupakan informasi yang berasal dari data seismik yang bisa dipelajari untuk memprediksi suatu target baik dengan kalkulasi langsung atau dengan analisis. Walaupun hubungan antara atribut seismik dengan karakteristik batuan dan reservoar tidak dapat dideskripsikan secara khusus, banyak makalah yang membuktikan bahwa atribut seismik merupakan faktor yang efektif untuk mengklasifikasikan karakter batuan. Makalah ini menampilkan hasil proses yang dilakukan oleh ANN yang dapat mempelajari pola hubungan antara atribut seismk dengan porositas. Tujuan utama penelitian ini adalah memprediksi penyebaran lateral porositas dari atribut seismik. Pada tahap awal, hubungan antara log porositas dengan satu tras seismik yang berhimpit ditentukan dengan ANN. Setelah jaringan terbentuk, metode tersebut diterapkan untuk seluruh tras seismik yang ada pada suatu volume seismik. Atribut seismik yang dijadikan input adalah amplitudo, amplitudo sesaat, frekuensi sesaat, fasa sesaat. Jaringan yang digunakan adalah Backporopagation dengan 3 layer masing-masing memiliki 55, 15, dan 1 neuron. Hasil proses jaringan ini memiliki tingkat korelasi 0,7 dengan rata-rata error kuadrat 15,6%. Kata Kunci : Atribut Seismik, Artificial Neural Network, Porositas"

"ABSTRAKTelah dilakukan pengolahan dan interpretasi data seismik 2D dengan menggunakan informasi line BGR06-141 untuk mengidentifikasi kondisi geologi subsurface pada zona epicentral gempa megathrust tahun 2004. Data seismik tersebut merupakan bagian dari daerah subduksi Lempeng Samudera Indo-Australia dan Benua Eurasia yang memiliki seismisitas yang tinggi dan rawan terhadap gempa. Pengolahan data seismik dilakukan dari proses preconditioning, radon filter, analisis kecepatan dan Post Stack Time Migration untuk meningkatkan signal to noise ratio(S/N ratio). Hasil pengolahan berupa penampang seismik yang menggambarkan kondisi geologi bawah permukaan daerah penelitian. Interpretasi yang berhasil dilakukan berupa sedimen sebagai cekungan busur muka (fore arc basin), tinggian (ridge), dan prisma akresi. Hasil penelitian juga menunjukkan patahan aktif strike slip fault, normal fault dan reverse fault pada penampang seismik yang diperkuat dengan penerapan atribut seismic variance. Diharapkan hasil bisa dimanfaatkan sebagai informasi awal mengenai kegempaan dimasa datang.

---

ABSTRACTA 2D seismic data processing and interpretation of line BGR06-141 have been done for identifying geological subsurface condition around epicentral zone of 2004 megathrust earthquake. The seismic data is acquired crossing the subduction zones between Indo-Australian Oceanic Plate and the Eurasian Continent Plate where has a high seismicity and vulnerable against earthquakes. Several data processing has been performed systematically on the seismic including pre-conditioning data, filtering using radon transform, velocity analysis and Post Stack Time Migration to improve enhance the signal to noise ratio (S/N ratio). Data processing results is shown in the form of vertical seismic section that describes the subsurface geological condition in the study. Careful interpretation of data processing result shows that the form of fore arc basin sediments, ridge and a prism accretion. This study also shows active strike slip fault, normal fault and reverse fault on the vertical seismic section which are confirmed by seismic attribute variance. We hope this result could be a reference for early information about future earthquake study."

"Interpretasi seismik merupakan salah satu tahapan dalam kegiatan eksplorasi seismik untuk memetakan struktur kedalaman, serta menemukan lokasi dengan prospek migas. Penelitian berlokasikan di Kalimantan dengan objek Cekungan X. Dengan menggunakan basis data berupa seismik 2D, checkshot borehole, log sumur, dan data-data geologi seperti stratigrafi regional dan peta geologi. Konversi kedalaman menggunakan metode layering cake, dengan dua strategi layering. Dimulai dengan picking horizon seismik dan fault, kemudian dilanjutkan dengan memetakan struktur berdasarkan two-way time. Konversi kedalaman bisa dilakukan setelah menentukan nilai k, Vo dan Vinterval. Dengan membuat permodelan Vo dan Vinterval yang kemudian diaplikasikan pada formula DT, maka didapatkan peta struktur kedalaman. Dari kalibrasi terhadap data log sumur, didapatkan data residual. Ketika permodelan peta residual diaplikasikan ke peta kedalaman sebelumnya maka akan didapatkan hasil akhir peta struktur kedalaman. Terakhir, dilakukan suatu analisis untuk menentukan prospek migas di cekungan tersebut, dan didapatkan dua lokasi yang memiliki prospek migas.

---

Seismic Interpretation is a part of exploration project to get depth structure map, also to find an interesting structure with hidrocarbon prospect. Located in Kalimantan with the main object is Basin X. 2D seismic data, borehole checkshot, well logs, and some geological data such as geological map and regional stratigraphy are the main database. In this study, depth conversion is performed by using layering cake method with two layering strategies. Starting with seismic horizon and fault picking, then mapping the time structure map based on two-way time. The depth conversion can be done after the value of k, Vo and Vinterval are inverted from checkshot. After make velocity modelling for Vo and Vinterval, used it with DT formula, then the result would be considered the fisrst depth structure map. Callibrated it with log data would gain residual depth, the residual modelling will give correction to the depth structure map, that would be considered as the final depth structure map. HC prospect and structure of interest are analyzed using this depth structure map, and found two structures with HC prospect."

"Atribut seismik merupakan informasi yang diperoleh dari data seismik yang dapat digunakan untuk memprediksi suatu target petrofisika baik secara numerik maupun analitik. Walaupun hubungan antara atribut seismik dengan karakteristik suatu batuan dan reservoar tidak dapat didefinisikan secara spesifik, banyak sumber yang menunjukkan bahwa atribut seismik merupakan salah satu parameter untuk mengklasifikasikan karakter dari suatu batuan. Skripsi ini menunjukkan hasil proses yang dilakukan oleh ANN yang dapat membuat suatu hubungan antara atribut seismik dengan saturasi air (Sw). Tujuan utama penelitian ini adalah memprediksi penyebaran lateral saturasi air (Sw) yang diperoleh dari atribut seismik. Pada tahap awal, hubungan antara log saturasi air (Sw) dengan satu tras seismik yang berhimpit ditentukan dengan menggunakan metode ANN. Setelah jaringan terbentuk, metode tersebut diterapkan untuk seluruh tras seismik yang ada pada suatu volume seismik. Atribut seismik yang dijadikan masukan adalah amplitudo, impedansi akustik, frekuensi sesaat, dan kuat refleksi (amplitudo sesaat). Jaringan yang digunakan adalah Backpropagation dengan 5 lapisan yang masing-masing memiliki 40, 30, 20, 10 dan 1 neuron. Metode pelatihannya menggunakan metode resilent backpropagation. Hasil proses jaringan ini memiliki nilai korelasi 96 % dengan nilai validasi sebesar 60 % dan nilai rata-rata error kuadrat (rmse) 3.01 %.

---

A seismic attribute information, which is obtained from seismic data, can be used to predict petrophysical properties analytically as well as numerically. Although the relationship between seismic attributes with rock properties can not be specifically defined, many papers indicated that seismic attributes can be used to characterize the rock.

This work shows the application of ANN algorithm to generate the relationship between seismic attributes and water saturation (Sw). The main objective of this study is to predict the lateral distribution of water saturation (Sw), which is derived from seismic data. The first step, the relationship between water saturation (Sw) and a trace seismic, which coincide with well log data, is determined using the ANN. After the network is defined, the method can then be applied to all existing seismic traces in a seismic volume. The input of seismic attributes is amplitude, acoustic impedance, instantaneous frequency, and reflection strength (amplitude envelope).

We use bacpropagation network with 5 layers each having 40, 30, 20, 10 and 1 neuron. The training method is resilent backpropagation. This network produces good agreement between predicted water saturation (Sw) and targeted water saturation (Sw), which is indicated by correlation coefficient of 96 %, validation coefficient of 60 % and root mean square error (rmse) of 3.01 %."

"Analisa multi atribut adalah salah satu metode statistik menggunakan lebih dari satu atribut untuk memprediksi properti fisik dari batuan. Tujuan analisa ini adalah adalah mencari hubungan antara log dengan data seismik. Hubungan ini digunakan untuk memprediksi Volome dari properti log pada semua volume seismik Pada penelitian ini analisa multiatribut diaplikasikan pada lapangan X daerah cekungan sumatera selatan dengan menggunakan 5 data sumur. Target dari penelitian ini adalah memprediksi penyebaran porositas di lapangan X. Sumursumur yang dipilih adalah sumur yang tersebar merata dan mewakili area yang akan diprediksi penyebaran porositasnya. Jumlah atribut yang digunakan di tentukan oleh proses step wise regression. Metode multiatribut yang linier transformasinya terdiri dari deret bobot yang diperoleh dari minimalisasi least square. Pada metoda non linier, neural network di gunakan dalam proses training dengan menggunakan atribut yang sudah ditentukan sebelumnya.Tipe neural network yang digunakan adalah PNN ( Probabilistic Neural Network ),tipe ini dipilih karena mempunyai hasil korelasi yang paling baik dibandingkan dengan tipe neural network yang lain. Untuk mengetahui tingkat kepercayaan dari transformasi multiatribut dilakukan proses crossvalidasi. Hasilnya multiatribut menunjukan korelasi sebesar 0.65 dan neural network 0.69.

---

Multi-attribute analysis is a statistic method using more than one attribute to predict physical properties of rocks. The aim of this analysis is to find a relationship between log and seismic data. The relationship is used for predicting volume of log property at all seismic volumes. In this study the multi-attribute analysis is applied to area X, which is a cavity region in South Sumatera, using five well data. The aim of the study is to predict porosity distribution at area X. The wells that were chosen were those that were spread evenly and represented areas where the distribution of porosity will be predicted. The quantity of attributes used is determined by a step wise regression process. A linear multiattribute method comprises of a series that is achieved by a minimised least square. In a non-linear method, neural network is used in the training process with predetermined attributes. The neural network type used was PNN (Probabilistic Neural Network ), this type was chosen because of the best correlation result. To verify the validity of the multi-attribute transformation, a crossvalidation was conducted. The result shows a 0.65 correlation and a 0.69 neural network."

"Reservoir karbonat diperkirakan mengandung hampir 60% dari total cadangan hidrokarbon dunia dan diperkirakan memiliki 50% dari total produksi hidrokarbon. Hidrokarbon umumnya terdapat pada batuan berpori. Porositas batuan karbonat umumnya memiliki heterogenitas yang tinggi, kompleksitas, dan random. Salah satu metode yang efektif untuk mengatasi heterogenitas adalah metode neural network. Sehingga penelitian ini bertujuan untuk menetukan distribusi porositas dengan neural network pada batuan karbonat dengan menggunakan 2 data sumur dan data seismik 2D post stack time migration (PSTM) pada lapangan T. Seismik atribut yang digunakan sebagai input proses probabilistic neural network berupa data seismik dan hasil inversi serta log yang akan diprediksi penyebarannya. Digunakan step wise regression dan validation error untuk menentukan atribut terbaik yang akan digunakan. Hasil prediksi nilai porositas menggunkan probabilistic neural network dengan input atribut terbaik yang telah terpilih menghasilkan korelasi yang lebih baik 0.81 dengan error 0.03 dibanding dengan metode multiatribut yang menggunakan persamaan linier yaitu 0.66 dengan error 0.04 dan hasil model log prediksi mendekati log aktual. Hasil distribusi porositas dapat dianilisis bahwa nilai porositas pada sumur C1 memiliki nilai porositas efektif yang rendah dibandingkan dengan sumur C4.

---

Reservoir carbonate mostly contains 60% of total hydrocarbon preserves in the world, and it is predicted about 50% which is produced hydrocarbon. Commonly, hydrocarbon is found in the rock pores. The porosity of carbonate, generally, has high heterogeneity, complexity, and random. One of effective methods to solve the problem is neural network. The aim of this study is to determine the distribution of porosity using neural network for carbonate in T field. Seismic attribute is used as input in neural network process which is seismic data, inversion result, and well log. Step wise regression and validation error are used to determine the best attributes that will be used to.

The prediction result of porosity using probabilistic neural network with the best attribute has better correlation than using multi attributes for linier method. The correlation and error value using neural network are 0.08% and 0.03%, while the value of correlation and error using multi attribute for linier method are 0.06% and 0.04%, respectively. The predicted log model is approaching the actual log. The result of porosity distribution shows that the porosity value of well C1 has lower effective porosity than well C4."

"Penelitian ini menggunakan data seismik 3 dimensi dan 5 data sumur dari lapangan w. Target penelitian yaitu batuan karbonat pada formasi Tuban di cekungan Jawa Timur Utara. Penelitian bertujuan menentukan distribusi porositas karbonat, dengan menggunakan neural network berdasarkan inversi dan atribut seismik. Inversi seismik model based dan linier programming sparse spike, menghasilkan impedansi akustik pada lapisan di bawah horizon Top Carbonate hingga horizon Base Carbonate, mengalami peningkatan signifikan pada rentang 38076 - 46857 ((ft/s)*(g/cc)). Atribut seismik sweetness, rms amplitude, dan reflection intensity, digunakan sebagai atribut eksternal, untuk tahap multiatribut linier regresi dan neural network. Multiatribut linier regresi dan neural network dilakukan untuk memprediksi porositas bedasarkan atribut-atribut internal maupun eksternal. Hasil analisis multiatribut yang diaplikasikan pada data raw seismik dan 5 volum atribut eksternal, yaitu log porositas prediksi, memiliki nilai korelasi sebesar 0.712 terhadap log porositas. Dan, nilai validasinya sebesar 0.573. Sedangkan, Probabilistic Neural Network menghasilkan porositas prediksi dengan nilai korelasi sebesar 0.661 dan nilai validasinya sebesar 0.485. Berdasarkan multiatribut linier regresi maupun probabilistic neural network, porositas rata-rata pada lapisan reservoar karbonat sebesar 10-15% di bagian utara. Sedangkan, di bagian selatan, porositas rata-rata hanya di bawah 6%.

---

This study uses three-dimensional seismic data and 5 well data from w field. The research target is carbonate rocks of the Tuban formation in North East Java basin. The study aims to determine the distribution of porosity carbonate, by using neural network algorithm, based on acoustic impedance inversion and seismic attributes. Models based inversion and linear programming sparse spike inversion result in acoustic impedance, in the layers below the horizon Top Carbonate to horizon Base Carbonate, experienced a significant increase impedance in the range 38076-46857 ((ft/s)*(g/cc)). Some seismic attribute; sweetness, rms amplitude, and reflection intensity, are used as external attributes for multi attribute linear regression and neural network. Multi attribute linear regression and neural network is done to predict porosity based on attributes of both internal and external.

The results of the analysis that is applied to the data multi attribute raw seismic and 5 volumes of external attributes, is called log porosity prediction, have a correlation value of 0.712 to log porosity original. And the value of its validation is 0.573. Meanwhile, Probabilistic Neural Network is producing log porosity prediction with correlation value of 0.661 and the value of its validation by 0485. Multi attribute based linear regression and probabilistic neural network, average porosity of the reservoir layer of carbonate of 10-15% in the north. Meanwhile, in the southern part, average porosity of just under 6%."

"ABSTRAKStudi yang dilakukan dalam penelitian ini untuk menganalisis multiatribut dari data seismik dan data log dalam mengidentifikasi litologi bawah permukaan, terutama dalam mengidentifikasi penyebaran sourcerockpada suatu daerah target tertentu. Metode multiatribut dalam penelitian ini dilakukan untuk memprediksi log vshaleyang diturunkan dari log resisitivity. Untuk mendapatkan gambaran distribusi sifat fisik batuan di bawah permukaan, maka hubungan antara data log dengan data seismik dapat membantu menghasilkan volume sebaran sifat fisik batuan yang tadinya hanya didapat dari tempat yang memiliki data log hingga menjadi sebaran pada seluruh area yang terdapat data seismik. Kemampuan neural network untuk mengenali pola dan menemukan hubungan non linear antara data seismik, data multiatribut dan sifat fisik batuan dari data log dimanfaatkan untuk mendapatkan sebaran sifat fisik batuan pada volume seismik dengan mengacu pada data log. Aplikasi geostatistik digunakan untuk mendapatkan model distribusi spasial di mana hasil dari prediksi volume shale dipakai sebagai data sekunder sedangkan data dari lubang sumur digunakan sebagai data primer, sehingga diharapkan mendapatkan kemenerusan secara lateral.

---

ABSTRACTThe Study conducted in this research to analyze the multi attributes of seismic data and log data in identifying subsurface lithology, especially in identifying sourcerock deployment in a given target area. Multi attributes method in this research is done to predict log vshale derived from log resistivity. To obtain a description of the distribution of the physical properties of rocks below the surface, the relationship between log data and seismic data can help to generate volume distribution of physical properties of rocks that were only obtained from places that have log data to be distributed in all areas of seismic data. The ability of neural networks to recognize patterns and find nonlinear relationships between multi attribute data seismic data and the physical properties of rocks from log data are utilized to obtain the distribution of physical properties of rocks at seismic volumes with reference to log data. The geostatistical application is used to obtain spatial distribution model where the result of the shale volume prediction is used as secondary data while the data from the wellbore is used as the primary data, so it is expected to obtain lateral generation."

"[ABSTRAK Lapangan-X merupakan lapangan gas di Cekungan Kutai yang dikembangkan sejak tahun 1986. Reservoar lapangan-X merupakan endapan delta Miosen akhir yang berlapis, dimana dikarakterisasikan oleh formasi yang didominasi oleh lempung. Perselingan antara batupasir dan batuserpih menghasilkan heterogenitas porositas yang cukup tinggi. Salah satu metode yang efektif dalam mengatasi tingkat heterogenitas yang tinggi adalah dengan metode Artificial Neural Network (ANN). ANN menggunakan algoritma Probabilistic Neural Network (PNN) mampu mendiskriminasikan daerah yang memiliki sebaran porositas yang tinggi dan rendah dengan baik pada zona Fresh Water Sand (FWS) lapangan-X dibanding dengan metode Multiatribut linier yang cenderung merupakan nilai sebaran porositas rata-rata. Nilai korelasi hasil prediksi terhadap target menggunakan metode PNN mencapai 0.8610 dengan rata-rata kesalahan (average error) sebesar 0.0283, sementara nilai korelasi hasil metode Multiatribut linier hanya sebesar 0.7098 dengan rata-rata kesalahan (average error) sebesar 0.0398. Hasil PNN pada sayatan waktu +10 ms dari horizon FS33 berhasil mengkarakterisasikan sebaran porositas batupasir yang bersih dari lempung di bagian selatan daerah penelitian, dimana fasies pengendapan batupasir tersebut diinterpretasikan berasal dari dataran delta. Sementara sayatan waktu -10 ms dari horizon FS42, menunjukan sebaran porositas batugamping dengan fasies pengendapannya diinterpretasikan berasal dari lingkungan neritik (shelf). Dari penelitian ini, dapat disimpulkan bahwa metode PNN berhasil menggambarkan sebaran porositas batuan di zona Fresh Water Sand (FWS) lapangan-X dengan baik sehingga hasil prediksi penyebaran yang dilakukan mampu mendekati data- data sumuran.

---

ABSTRACT X-field is a gas field in Kutai Basin and it has been developed since 1986. Reservoir of X-field is a multi layered upper Miocene deltaic deposits and characterized by a shaly formation. A highly intercalation between sand & shale unit in X-field has been contributed to the heterogeneity of porosity in the area. One of the effective methods to spatially quantify such heterogeneity of porosity is by using Artificial Neural Networks (ANN). ANN with Probability Neural Network (PNN) algorithm has been successfully retained more dynamic range, high and low frequency porosity content, compare to the Multiattributes linear which is tend to show a smoothed, or more averaged prediction. The correlation value from PNN methods can be up to 0.8610 with average error is 0.0283, while correlation value from Multiattribute linear only up to 0.7098 with average error is 0.0398. The time slice of PNN result at +10ms from horizon FS33 has been clearly figured out an accumulation of high porosity in the southern area of the interval target which is indicated as a clean sand lithology based on sensitivity analysis. And such accumulation has formed a distributaries channel trend which is interpreted as delta plain deposits. Meanwhile, the time slice of PNN result at - 10 ms from horizon FS42 has indicated a carbonate lithology which is interpreted as shelf deposits. From this study, it?s concluded that PNN algorithm as a nonlinear function has been successfully showed a better porosity distribution in the Fresh Water Sand (FWS) zone of X-field.;X-field is a gas field in Kutai Basin and it has been developed since 1986. Reservoir of X-field is a multi layered upper Miocene deltaic deposits and characterized by a shaly formation. A highly intercalation between sand & shale unit in X-field has been contributed to the heterogeneity of porosity in the area. One of the effective methods to spatially quantify such heterogeneity of porosity is by using Artificial Neural Networks (ANN). ANN with Probability Neural Network (PNN) algorithm has been successfully retained more dynamic range, high and low frequency porosity content, compare to the Multiattributes linear which is tend to show a smoothed, or more averaged prediction. The correlation value from PNN methods can be up to 0.8610 with average error is 0.0283, while correlation value from Multiattribute linear only up to 0.7098 with average error is 0.0398. The time slice of PNN result at +10ms from horizon FS33 has been clearly figured out an accumulation of high porosity in the southern area of the interval target which is indicated as a clean sand lithology based on sensitivity analysis. And such accumulation has formed a distributaries channel trend which is interpreted as delta plain deposits. Meanwhile, the time slice of PNN result at - 10 ms from horizon FS42 has indicated a carbonate lithology which is interpreted as shelf deposits. From this study, it?s concluded that PNN algorithm as a nonlinear function has been successfully showed a better porosity distribution in the Fresh Water Sand (FWS) zone of X-field.;X-field is a gas field in Kutai Basin and it has been developed since 1986. Reservoir of X-field is a multi layered upper Miocene deltaic deposits and characterized by a shaly formation. A highly intercalation between sand & shale unit in X-field has been contributed to the heterogeneity of porosity in the area. One of the effective methods to spatially quantify such heterogeneity of porosity is by using Artificial Neural Networks (ANN). ANN with Probability Neural Network (PNN) algorithm has been successfully retained more dynamic range, high and low frequency porosity content, compare to the Multiattributes linear which is tend to show a smoothed, or more averaged prediction. The correlation value from PNN methods can be up to 0.8610 with average error is 0.0283, while correlation value from Multiattribute linear only up to 0.7098 with average error is 0.0398. The time slice of PNN result at +10ms from horizon FS33 has been clearly figured out an accumulation of high porosity in the southern area of the interval target which is indicated as a clean sand lithology based on sensitivity analysis. And such accumulation has formed a distributaries channel trend which is interpreted as delta plain deposits. Meanwhile, the time slice of PNN result at - 10 ms from horizon FS42 has indicated a carbonate lithology which is interpreted as shelf deposits. From this study, it’s concluded that PNN algorithm as a nonlinear function has been successfully showed a better porosity distribution in the Fresh Water Sand (FWS) zone of X-field., X-field is a gas field in Kutai Basin and it has been developed since 1986. Reservoir of X-field is a multi layered upper Miocene deltaic deposits and characterized by a shaly formation. A highly intercalation between sand & shale unit in X-field has been contributed to the heterogeneity of porosity in the area. One of the effective methods to spatially quantify such heterogeneity of porosity is by using Artificial Neural Networks (ANN). ANN with Probability Neural Network (PNN) algorithm has been successfully retained more dynamic range, high and low frequency porosity content, compare to the Multiattributes linear which is tend to show a smoothed, or more averaged prediction. The correlation value from PNN methods can be up to 0.8610 with average error is 0.0283, while correlation value from Multiattribute linear only up to 0.7098 with average error is 0.0398. The time slice of PNN result at +10ms from horizon FS33 has been clearly figured out an accumulation of high porosity in the southern area of the interval target which is indicated as a clean sand lithology based on sensitivity analysis. And such accumulation has formed a distributaries channel trend which is interpreted as delta plain deposits. Meanwhile, the time slice of PNN result at - 10 ms from horizon FS42 has indicated a carbonate lithology which is interpreted as shelf deposits. From this study, it’s concluded that PNN algorithm as a nonlinear function has been successfully showed a better porosity distribution in the Fresh Water Sand (FWS) zone of X-field.]"

"Aplikasi metode Multiatribut pada data poststack seismik dan hasil inversinya telah dilakukan untuk mengestimasi kecepatan interval melalui pendekatan Neural Network. Estimasi kecepatan interval yang dihasilkan tersebut digunakan untuk memprediksi tekanan formasi di Lapangan Texaco 3D, Louisiana. Tujuan dari studi ini adalah untuk mengaplikasikan pendekatan geostatistik dan analisis Multiatribut dengan keterbatasan data yang dimiliki untuk memprediksi tekanan formasi. Hasil estimasi kecepatan interval menggunakan Multiatribut (10 atribut) menunjukkan korelasi yang sangat baik yaitu rata-rata korelasi prediksi log hasil atribut dan log validasi mencapai 79%, dengan tingkat kesalahan yang kecil berkisar rata-rata 175 - 292 m/s dari kecepatan validasi. Pendekatan Neural Network menghasilkan atribut polaritas semu (apparent polarity) sebagai atribut terbaik dalam estimasi kecepatan dengan error berkisar 108 m/s (berdasarkan hasil PNN) hingga 166 m/s (berdasarkan hasil MLFN). Anomali kecepatan rendah terdeteksi pada kedalaman 2800 - 2900 m dan sekitar kedalaman 3000 m, dengan gradient tekanan rata-rata mencapai 18 ? 22 ppg.

---

Application of Multiattribute to poststack seismic data and the the seismic inversion result has been carried out to estimate the interval velocity, by using Neural Network approach. The result of estimated interval velocity is used to predict formation pressure in Texaco 3D Field, Louisiana. The purpose of this study is to apply the geostatistical approach and Multiattribute analysis to predict the formation pressure.

The results of estimated interval velocity using Multiattribute (10 attributes) show excellent correlation of the average correlation between predicted log and the real log reached 79%, with an error training and validation of a fairly small range from an average of 175-292 m/s validation of the velocity. The Neural Network approachment generating apparent polarity attribute as the best attribute of velocity estimation with errors ranging from 108 m/s (based on PNN) up to 166 m/s (based on the results of MLFN). Low velocity anomaly was detected at a depth of 2800 - 2900 m and approximately 3000 m depth, with the pressure gradient averaged 18-22 ppg."