Research On State Diagnosis On Fan Based On Factor Analysis And BP Neural Network

With the community continuously development and progress, people demand for energy is increasing. As an ideal clean energy, wind power gets more and more attention. With the wind power installed capacity continues to expand and the intelligent level of fan equipment raising.the mandate of Wind Power Management Department become higher and higher. The contradiction between advanced fan equipment and the slow level of maintenance management of wind farm gets outstanding.It has become the barrier of wind power business and wind power industry. Fan equipment maintenance and management has become one Important part of Wind Business management. Advanced fan equipment urgently needs advanced maintenance techniques and Advanced management method.First of all,the article pre-sentenced factors affecting the amount of fan power in SCADA under the Knowledge of aerodynamics,it got possible factors,and tested and verified date with Knowledge of wind resources.and excluded invalid data. Secondly.the article analyzed the possible factors under factor analysis to reducing the number of factors, then the ariticle used the reducing the number of factors as the input and active power as the output of BP neural network to building fan status diagnostics model.In the end. the article presented concrete steps of using the model,and designed the overall structure of diagnosis expert system.and conceptual designed of the main modules.The article empirical researched Fujian Z Wind Farm, the diagnostic conclusion can be used as the foundation of the fan repair work.