Aiming at the reasoning problems that the Mutual integration of the numerical information and qualitative do‐main knowledge ,this paper proposes a hybrid computing neural network (HCNN) based on cloud transformation was pro‐posed .Using the reverse normal cloud generator can achieve the conversion of the uncertain relationship between the quanti‐tative values and qualitative concept description ,and build the mixed information reasoning logic and HNN model based on cloud transformation . Then transforming the numerical information into qualitative concept in the sense of probability through the cloud transform ,and expressing the inference rules as neurons ,and using the learning nature of the neural net‐works to achieve adaptive processing of mixed quantitative and qualitative information .In algorithm design ,integrating net‐work properties parameters for a particle ,the hybrid particle swarm optimization (pso) algorithm for computing the neural network to the overall optimal solution .In the automatic identification of sedimentary microfacies in the study of petroleum geology ,the results verify the validity of the model and algorithm .%针对数值信息与定性领域知识相互融合的计算问题，提出了一种基于云变换的混合计算神经网络模型。利用正向正态云发生器可实现定性概念到量化数值描述之间不确定关系的转换机制，建立基于云变换的混合信息计算逻辑和神经网络模型。将定性概念谓词通过云变换在概率意义下转换为数值变量，把计算规则表示为神经元，利用神经网络的学习性质来实现对定量与定性混合信息的自适应计算和推理。在算法设计中，将网络性质参数整合为一个粒子，利用粒子群算法对混合计算神经网络进行整体优化求解。以石油地质研究中的沉积微相自动识别为例，实验结果验证了模型和算法的有效性。
展开▼
