The invention relates to the technical field of railway
catenary monitoring, in particular to a multi-source sensing fusion method and
system for state evaluation of a
catenary compensation device, and the method comprises the steps: obtaining multi-source
sensing data such as vibration, temperature and current, generating a
phase space trajectory through employing a self-adaptive
phase space reconstruction technology, and carrying out the state evaluation of the
catenary compensation device; multiple topological features are extracted to form topological feature vectors, and the vectors are combined with temperature and current data to construct a
fault probability model. According to the method, the health degree
score of the
contact network compensation device can be calculated,
trend prediction can be carried out, early warning information can be generated, high-precision early warning is realized, the clamping stagnation risk of the
pulley block can be predicted three months in advance, passive response is converted into active prevention, the maintenance efficiency and safety of the
contact network are remarkably improved, and the maintenance cost is reduced. By means of the innovative technology, the safety of railway operation is improved, the maintenance cost is reduced, and revolutionary transformation is brought to the railway industry.