Application of neural tools in geological data analyses

  • Slides: 28
Download presentation
Application of neural tools in geological data analyses Dr. Tomislav Malvić, Grad. in Geol.

Application of neural tools in geological data analyses Dr. Tomislav Malvić, Grad. in Geol. INA-Industry of Oil Plc. , E&P of Oil and Gas, Reservoir Engineering and Field Engineering Dept. (advisor) Faculty of Mining, Geology and Petroleum Engineering, Institute of Geology and Geological Engineering (visiting lecturer) Visiting lecture for IAMG student chapter in Szeged, Hungary 14 th Nov 2008

INTRODUCTION IN NEURAL ARCHITECTURE Generally, neural networks can be described as: Biological (human) and

INTRODUCTION IN NEURAL ARCHITECTURE Generally, neural networks can be described as: Biological (human) and Artificial or simulated (computer algorithms based network). Fig. 1: Biological (human) neurons Fig. 2: Artificial neurons (schematic)

Fig. 3: The artificial neuron model (extended) q. The input layers collects and distributes

Fig. 3: The artificial neuron model (extended) q. The input layers collects and distributes data into the network. q. The hidden layer(s) process such data. q Equation (1) represents a set of operations performed on the neuron. q Equation (2) detects activation of the neuron. The Activation function – the value of output (U) is compared with condition necessary for hypothesis acceptance (t). The function is started only if this value is appropriate.

Fig. 4: Schematic organization of neural network through the layers The basic Equation 1

Fig. 4: Schematic organization of neural network through the layers The basic Equation 1 impies: previously coefficients, determined weighting Condition of hypothesis acceptance, Number of layers, Number of neurons in layer. Coefficient estimation PROPAGATION process (or backerror procedure). Fig. 5: Adoption of weighting coefficient and error decreasing is BACK

Simple (basic) neuron architecture recognize inputs behaviour through finding linearity (it is perceptron concept).

Simple (basic) neuron architecture recognize inputs behaviour through finding linearity (it is perceptron concept). Back-propagation network by backing error and adopting coefficient overcome this limitation using hidden layers. Backpropagation network is also called Multilayer Perceptron Network. Such error is determined for each neuron, and applied for adopting weighting coefficient and activation value. It is learning (training) and validating of the network. The weighting coefficient are calculated through Equation 3 and 4.

Backpropagation (disadvantages) – the most used paradigm, but often characterised with long lasting training.

Backpropagation (disadvantages) – the most used paradigm, but often characterised with long lasting training. Simple (basic) neuron architecture recognize inputs behaviour through finding linearity (it is perceptron concept). It resulted from the gradient descent method used in backprop. This problem is often expressed in geophysical neural application. The very large dataset, and sending each channel (attribute, input) back can significantly decreased learning rate (slow processing) and paralyze the network. Resilient Propagation Algorithm (r. Prop) – one of the often improvements of backprop. The main difference is using only of partial derivations in process of weighting coefficient adjustment. It is about 45 times faster than the standard backprop algorithm. Radial Basis Function Algorithm (RBF) – is an artificial network that uses radial basis fnction as activation function. Very often it is applied in function approximation, time series prediction etc. A radial basis function is a real-valued function whose value depends only on the distance from the origin or alternatively on the distance from some other point c, called a center.

Fig. 6: The Multi Layer Perceptron (MLP) backprop network Fig. 7: The Radial Basis

Fig. 6: The Multi Layer Perceptron (MLP) backprop network Fig. 7: The Radial Basis Funcion (RBF) network

ARCHITECTURE OVERVIEW: The networks architecture includes: 1. 2. 3. 4. Distribution of neuron in

ARCHITECTURE OVERVIEW: The networks architecture includes: 1. 2. 3. 4. Distribution of neuron in different layers; Defining of connection types among neurons; Defining of the way how neurons receiving inputs and calculate outputs; Setting of the rules how to adjust weighting coefficient. The application of neural network includes: 1. Learning of training of network; 2. Testing of network; 3. Applying of the network for prediction.

ANALYSED AREAS (CROATIAN PANNON) • The Okoli field (prediction of facies) in 2006; •

ANALYSED AREAS (CROATIAN PANNON) • The Okoli field (prediction of facies) in 2006; • The Beničanci field (porosity) in 2007 and • The Kloštar field (lithology and saturation) in 2007/08. Fig. 8: Areas analyzed by neural networks in Croatia

OKOLI FIELD The neural analysis was performed using c. VISION – Neuro Genetic Solution

OKOLI FIELD The neural analysis was performed using c. VISION – Neuro Genetic Solution commercial software. Available at: http: //www. bestneural. net/

The Okoli field, located in the Sava depression, is selected as the example for

The Okoli field, located in the Sava depression, is selected as the example for clastic facies prediction using neural network. The significant oil and gas reserves are proved in Lower Pontian sandstones. The analysis is based on r. Prop algorithm. The network is trained using log data (curves GR, R 16", R 64", PORE/T/W, SAND & SHALE) from two wells (code names B-1 & B-2). The neural network was trained based on selected part of input data and registered lithology from c 2 reservoir (as analytical target) of Lower Pontian age. Positions of facies (sand/marl sequences) were predicted. The results indicate on over-trained network in the case of sandstone sequences prediction (Figures 10, 11), because the marl sequences in the top and the base are mostly replaced by sandstone. The further neural facies modelling in the Sava depression need to be expanded with additional logs that characterised lithology and saturation (SP, CN, DEN). Then, r. PORP algorithm could be reached with more than 90% probability of true prediction (in presented analysis this value reached 82. 1%).

Figure 9: Structural map of c 2 reservoir top with selected well's positions

Figure 9: Structural map of c 2 reservoir top with selected well's positions

Figure 10: Relations of errors in periods of training (T), learning (L) and validation

Figure 10: Relations of errors in periods of training (T), learning (L) and validation (V) and position of Face and Best configurations (the symbols F, B in legend) for B-1 well Figure 11: Relations of errors in periods of training (T), learning (L) and validation (V) and position of Face and Best configurations (the symbols F, B in legend) for B-2 well

CONCLUSIONS (Okoli field) 1. This is the first neural analysis in hydrocarbon reservoir analysis

CONCLUSIONS (Okoli field) 1. This is the first neural analysis in hydrocarbon reservoir analysis in Croatia 2. Excellent correlation was obtained between predicted and true position of sandstone lithology (reservoir of Lower Pontian age in the Sava depression); 2. On contrary, positions of predicted and true marlstones positions (in top and bottom) mostly do not correspond; 3. The best prediction (so called Face machine) is reached in relatively early training period. In B-1 well such prediction is observed in 2186 th iteration, and in B-2 well in 7626 th iteration; 4. It means that in similar facies analyses in the Sava depression, it is not necessary to use large iteration set (here is used about 30000); 5. The input dataset would need to be extended on other log curves that characterize lithology, porosity and saturation, like SP (spontaneous potential), CN (compensated neutron), DEN (density) and some other; 6. The wished true prediction could reached 90% (Face machine could be configured with 90% probability).

BENIČANCI FIELD The neural analysis was performed using NEURO 3 – Neural Network Software.

BENIČANCI FIELD The neural analysis was performed using NEURO 3 – Neural Network Software. It is freeware E&P Tools published by the National Energy Technology Laboratory (NETL), owned and operated by the U. S. Department of Energy (DOE) national laboratory system. (http: //www. netl. doe. gov/technologies/oil-gas/Software/e&ptools. html)

GENERAL LITHOLOGY AND NETWORK TYPE: The reservoir is represented by carbonate breccia (and conglomerates)

GENERAL LITHOLOGY AND NETWORK TYPE: The reservoir is represented by carbonate breccia (and conglomerates) of Badenian age. Locally the thickness of entire reservoir sequence is locally more than 200 m. The three seismic attributes were interpreted – amplitude, phase and frequencies making 3 D seismic cube, averaged and correlated by well porosities at the 14 well locations. The 14 seismic and porosity point data made the network training. The network was of the backpropagation type. It was fitted through 10000 iterations, searching for the maximal value of correlation between attribute(s) and porosities and the minimal convergence.

The best training was reached using all three attributes together, what indicated on: q

The best training was reached using all three attributes together, what indicated on: q tendency that neural networks like numerous inputs; q physical connection among seismic attributes. Results are presented for: q Kriging (Figure 12 a); q Cokriging (Figure 12 b) and q Neural network (Figure 12 c). Neural map is based at cell estimation, rarely reaching of hard-data porosity minimum and maximum (the scale is 5 -10%, and the geostatistics interpolated in 3 -11%). It means that neural estimation is more “conservative” than geostatics (Figure 12 c). The cokriging approach includes one attribute. The neural approach favours using of three attributes. The possible attribute physical connection alerts us on carefully and geologically meaningful selection of the network inputs.

Figure 12 a: Kriging porosity map (colour scale 4 -10%) Figure 12 b: Cokriging

Figure 12 a: Kriging porosity map (colour scale 4 -10%) Figure 12 b: Cokriging porosity map (colour scale 3 -11%) Figure 12 c: Neural network porosity map (colour scale 5 -10%)

CONCLUSIONS (Beničanci field) 1. The neural network was selected as the tool for handling

CONCLUSIONS (Beničanci field) 1. The neural network was selected as the tool for handling uncertainties of porosity distribution in breccia-conglomerate carbonate reservoir of the Badenian age; 2. The lateral changes in averaged reservoir's porosities are influenced by the Middle Miocene depositional environments; 3. The best porosity training results are obtained when all three seismic attributes (amplitude, frequency, phase) were used; 4. The reached correlation of neural results for each attribute is R 2=0. 987 and convergence criteria Se 2=0. 329; 5. These values can slightly (a few percent) differs in every new training, what is consequence of stochastic (random sampling) is some processes of the network fitting; 6. The result indicates that neural network very favour the numerous inputs, but also can be easily applied in the Beničanci field for porosity prediction.

KLOŠTAR FIELD Neural analysis was done by package Stat. Soft STATISTICA 7

KLOŠTAR FIELD Neural analysis was done by package Stat. Soft STATISTICA 7

The field is located in the Sava depression. The largest oil reserves are in

The field is located in the Sava depression. The largest oil reserves are in Upper Miocene sandstones in: q I. series (Lower Pontian age), q. II. series (Upper Pannonian age). Neural networks were trained in two wells (Klo-A and Klo-B). Inputs were conventional log data (curves SP, R 16 and R 64). The neural networks were used to predict: q. Lithology and q. Saturation with hydrocarbons.

DATA ANALYSIS The networks designing included: q Number of hidden layers and neurons in

DATA ANALYSIS The networks designing included: q Number of hidden layers and neurons in each layer; q Selection of the best training algorithm; q Number of epochs (iterations); q Learning rate (or here called momentum coefficient).

LITHOLOGY PREDICTION Input data: • Spontaneous potential (SP) log • Resistivity logs R 16

LITHOLOGY PREDICTION Input data: • Spontaneous potential (SP) log • Resistivity logs R 16 and R 64 • Paper description of available cores Lithology was defined as a categorical variable - sand (1) or marl (0). Neural network type and properties Well Training errora Selection errora RBF 3– 31– 1 Klo-A 0. 152942 0. 172753 MLP 3– 4– 6– 3– 1 Klo-A 0. 31438 0. 133478 RBF 3– 1 Klo-B 0. 156621 0. 149185 MLP 3– 6– 4– 2– 1 Klo-B 0. 255012 0. 214935 a. Error value ranges from 0 to 1, where 0 represents 100% success of prediction, i. e. , no error.

LITHOLOGY PREDICTION (example in well “Klo-B”). The better results are obtain by RBF network.

LITHOLOGY PREDICTION (example in well “Klo-B”). The better results are obtain by RBF network. Figure 13: RBF network training (II. sandstone series UP, I. sandstone series DOWN)

SATURATION PREDICTION Input data: • Spontaneous potential (SP) log • Resistivity logs R 16

SATURATION PREDICTION Input data: • Spontaneous potential (SP) log • Resistivity logs R 16 and R 64 • Paper description of available cores and saturation from DST Hydrocarbon saturation was defined as a categorical value – saturated (1) and unsaturated (0). Neural network type and properties Training error Selection error MLP 5– 6– 8– 1 0. 056897 0. 091173

SATURATION PREDICTION (examples from Klo-A and Klo-B). The better results are obtain in both

SATURATION PREDICTION (examples from Klo-A and Klo-B). The better results are obtain in both wells by MLP network. . Figure 13: MLP network training (both series are shown) (Klo-A UP, Klo-B DOWN)

CONCLUSIONS (Kloštar field) 1. Neural networks were trained with the tasks of: q Analyzed

CONCLUSIONS (Kloštar field) 1. Neural networks were trained with the tasks of: q Analyzed sandstone series of Upper Pannonian and Lower Pontian age; q Predicting lithology; q Predicting hydrocarbon saturation. 2. RBF network was used for prediction of lithology; 3. MLP network was used for prediction of hydrocarbon saturation; 4. Results were very good, with small error; 5. Neural network could be applied in sandstone reservoir characterisation; 5. In the Sava depression, RBF and MLP networks are good tool for acquiring useful results from well logs and extending properties along the reservoir (lateral).

RECOMMENDED REFERENCES ANDERSON, J. A. and ROSENFELD, E. (1989): Neurocomputing: Foundations of Research. Cambridge,

RECOMMENDED REFERENCES ANDERSON, J. A. and ROSENFELD, E. (1989): Neurocomputing: Foundations of Research. Cambridge, MA: MIT Press. CHAMBERS, R. L. & YARUS, J. M. (2002): Quantitative Use of Seismic Attributes for Reservoir Characterization. RECORDER, Canadian SEG, Vol. 27, pp. 14 -25, June. CVETKOVIĆ, M. (2007): Petroleum geology use of neural networks on the example of reservoir in Kloštar field. University of Zagreb, Faculty of Mining, Geology and Petroleum Engineering, Graduate thesis, mentor Prof. Dr. J. Velić, 15. June 2007, 49 p. MALVIĆ, T. (2006): Clastic facies prediction using neural networks (Case study from Okoli field). Nafta, 57, 10, 415 -431. MALVIĆ, T. and PRSKALO, S. (2007): Some benefits of the neural approach in porosity prediction (Case study from Beničanci field). Nafta, 58, 9, 455 -467. Mc. CORMACK, M. D. (1991): Neural Computing im Geophysics. The Leading Edge, 10/1, Society of Exploration Geophysicists. RIEDMILLER, M. and BRAUN, H. (1993): A direct adaptive method for faster backpropagation learning: The RProp algorithm. Proc. of the IEEE Intl. Conf. on Neural Networks, San Francisco, p. 586 -591. ROSENBLATT, F. (1958): The perceptron: A probabilistic model for information storage and organization in the brain. Psychological Review, 65, 386 -408. ZAHEDI, F. (1993): Inteligent systems for business, expert systems with neural networks. Wodsworth publishing Inc.