基于退火算法优化的BP神经网络的电力通信网业务可靠性评价

Reliability Evaluation of Power Communication Network Business Based on BP Neural Network Optimized by Simulated Annealing Algorithm

对于规模越来越大的电力通信网来说,快速预测和把握可靠性对其发展具有至关重要的作用。对可靠性的定量化度量的研究,多为对实际操作中人工进行业务评价,这里根据电力通信网故障影响因素,从设备层和光缆层构建了故障时长预测指标体系,对其中的每个指标进行量化。然后基于此指标体系,利用BP神经网络模型和基于退火算法优化的BP神经网络进行预测构建了定量化的可靠性模型,给出了度量可靠性的方法,为可靠性量化评价提供了一种解决方法。

For an increasingly large-scale power communication network, the rapid prediction and control of its reliability issues are crucial to its development. The existing research on the quantitative measurement of its reliability is mostly the artificial evaluation of the operations in actual operations. Based on the factors affecting the failure of the power communication network, a fault forecasting indicator system is constructed from the equipment layer and the cable layer, and each of these indicators is quantified. Then based on this index system, BP neural network model and BP neural network based on annealing algorithm optimization are used to forecast and build a quantitative reliability model. The reliability measurement method is given and a solution to the quantitative reliability evaluation is provided.