量子自组织网络模型及在油田水淹层识别中的应用

Quantum Self-organizing Network Model and Its Application in Oilfield Flooding Layer Identification

为解决油田测井解释中的水淹层识别问题,研究了一种量子自组织网络模型以及应用在油田水淹层的识别效果。首先样本数据由量子比特进行描述,将描述后的量子比特映射到Bloch球面上,竞争层权值则表示为Bloch球面上的任意分布的量子比特,然后根据球面坐标求得样本与权值之间的最短距离,距离最短的节点为获胜节点。针对油田水淹层水淹程度的识别问题,研究了一种量子自组织网络算法。最后以实际数据为例,进行了水淹层识别处理并与普通过程神经网络对比。结果表明,本文提出的量子自组织网络算法在水淹层识别的问题上具有较高精度。

In order to solve the problem of water-flooded zone identification in oilfield logging interpretation,this paper establishes a model using quantum self-organizing network and applies it in oilfield water-flooded zone identification.First,the sample data are described by qubits,and the described qubits are mapped to the Bloch sphere,and the weights of the competition layer are expressed as qubits randomly distributed on the Bloch sphere.Then,the distance between the sample and the weight is obtained according to the spherical coordinates.The node with the shortest distance is the winning node.Aiming at the problem of identifying the degree of flooding of oilfield water-flooded layers,a quantum self-organizing network algorithm is studied.Finally,taking the actual data as an example,the water-flooded layer identification processing was carried out and compared with the normal process neural network.The results show that the quantum self-organizing network algorithm proposed in this paper has high accuracy in the identification of water-flooded layers.