|
In the automatic unpacking control system, the control accuracy of the flipping speed of the flipping platform is of great significance to the dumping effect and the service life of the equipment. This paper conducts research based on this. Aiming at the fact that the standard extreme learning machine (ELM) is prone to fall into local optimum, this paper proposes a model for extreme learning machine (ASSA-ELM) optimization based on improved salp swarm optimization algorithm is proposed and applied to an example of flipping speed prediction of flipping platform. Based on the salp group optimization algorithm (SSA), a position update strategy combining the adaptive weight method and the proportional weight of the improved step size Euclidean distance is introduced. The weights and hidden layer biases are optimized, which greatly improves the generalization ability of the ELM model and the accuracy of the predicted value. The algorithm models before and after the improvement are compared and analyzed. The results show that the predicted value of the ASSA-ELM model has the highest fitting degree with the actual value collected in the industrial field, and has a high prediction accuracy, which verifies the feasibility and effectiveness of the ASSA-ELM model in the prediction of the turning speed of the turning platform.
|
