State evaluation of power cable based on RBF information fusion algorithm using multi-parameters

With the development of electric power industry, power cable will play a more and more important role not only widely used in distribution network, but also in the field of AC/DC and high voltage transmission system. Power cable is generally underground or underwater, the problem that fault concealment and long time troubleshooting make serious impact on the power supply reliability of power system. Evaluate real-time operation state of power cable by using multi-parameters from variety of monitoring and detection information, and strengthen early defect equipment monitoring on the basis of evaluation result, it can reduce equipment failure rate and prevent accident expanding and it has very important significance for improving power supply reliability, From the perspective of information fusion, the method of evaluation algorithm based on RBF (Radial basis function) neural network is proposed. Multi-parameter including online detection data, environmental factors, testing and equipment information is used to conduct a comprehensive evaluation on cable status. The method using data fusion in decision-making level by RBF neural network for results of different algorithms including catastrophe theory and ANP fuzzy theory can make up applicability problem of different evaluation algorithm, and achieve complementary functions between the algorithms. To further adapt to the actual conditions of parameters, the integrity of input parameters was associated to credibility of evaluation results, it will make evaluation results more complete. The affectivity of the algorithm was verified by practical data.