Application of Evolutionary Neural Networks for Well-logging Recognition in Petroleum Reservoir.
ABSTRACT A critical task of well-logging interpretation is to differentiate oil-gas-water layers. Other approaches based on data exploration and low recognition rate are difficult to generalize oil-gas-water layers identification because of the high moisture content in the later period of development. In this research we utilize evolutionary neural networks to build the interpreting model of oil-gas-water layers and extracting well-logging parameters. By using an evolutionary neural network method to recognize reservoir stratum, it can efficiently distinguish oil-gas-water layers.
- SourceAvailable from: psu.edu
Conference Paper: Efficient evolution of neural network topologies[Show abstract] [Hide abstract]
ABSTRACT: Neuroevolution, i.e. evolving artificial neural networks with genetic algorithms, has been highly effective in reinforcement learning tasks, particularly those with hidden state information. An important question in neuroevolution is how to gain an advantage from evolving neural network topologies along with weights. We present a method, NeuroEvolution of Augmenting Topologies (NEAT) that outperforms the best fixed-topology methods on a challenging benchmark reinforcement learning task. We claim that the increased efficiency is due to (1) employing a principled method of crossover of different topologies, (2) protecting structural innovation using speciation, and (3) incrementally growing from minimal structure. We test this claim through a series of ablation studies that demonstrate that each component is necessary to the system as a whole and to each other. What results is significantly faster learning. NEAT is also an important contribution to GAs because it shows how it is possible for evolution to both optimize and complexify solutions simultaneously, making it possible to evolve increasingly complex solutions over time, thereby strengthening the analogy with biological evolutionEvolutionary Computation, 2002. CEC '02. Proceedings of the 2002 Congress on; 02/2002
Conference Paper: Qualitative interpretation of process trends by using neural networks[Show abstract] [Hide abstract]
ABSTRACT: Qualitative interpretation is a process to convert numerical output of sensors into symbolic representation. This process is one of the most critical path to connect intelligent systems with real world. In this paper, qualitative interpretation is realized as pattern-based classification of time-series signal by using ART2 neural networks. As an example, automatic classification of flow patterns in a pneumatic conveyor is successfully demonstratedKnowledge-Based Intelligent Electronic Systems, 1998. Proceedings KES '98. 1998 Second International Conference on; 05/1998
- [Show abstract] [Hide abstract]
ABSTRACT: The issue of the modeling for the distribution management based on demand forecasting. ANN model is applied to the field of demand forecasting. The modeling of market demand forecasting is built using BP algorithm. The model of forecasting is established by Simulation model of MATLAB. Based on them, a simple numerical example is given to test and forecast with three-layer BP network model.Distributed Computing and Applications to Business Engineering and Science (DCABES), 2010 Ninth International Symposium on; 09/2010