Boiler flame detection algorithm based on PSO-RBF network

As the main production tool in the industrial environment, large boilers play a vital role in the conversion and utilization of energy. Therefore, the furnace flame detection technology for boilers has always been a hot issue in the field of industrial automation and intelligence. In order to further improve the timeliness and accuracy of the flame detection network, a radial basis function ( RBF) flame detection network based on particle swarm optimization ( PSO) algorithm is proposed. First, the proposed algorithm initializes the speed and position parameters of the particles. Then, the parame-ters of the particles are mapped to the RBF flame detection network. Finally, the algorithm is iteratively updated to obtain the global optimal solution. The PSO-RBF flame detection algorithm adopts a flame sample collection method similar to back propagation ( BP) flame detection algorithm, and further im-proves the collection efficiency. The experimental results show that the PSO-RBF flame detection network has good accuracy and faster convergence speed in the given data samples. In the flame data samples, the detection accuracy of the PSO-RBF flame detection algorithm reaches 90. 5％.