Study of Optimizing Logistic Distribution Routing Based on Improved Ant Colony Algorithm
Zhang Yon · Control Engineering of China · 2015
To solve the problem that production process indicators expressed by concentrate grade and recovery in flotation is difficult to establish a precise mathematical model, while ordinary control methods require expensive instrumentations, a combined method of chaos ant and neural networks is proposed. This method selects the beneficiation flotation process as the research object to achieve the optimal set of process indicators of the flotation production process. This model uses principal component analysis to reduce the dimensionality of input data, and adjust premise parameters and target values with the hybrid algorithm combined from the chaos ant colony algorithm and the least squares method. The algorithm replaces the quadratic programming and solves the optimization problem with high speed and accuracy. Even under the situation of random disturbances or measurement noise, this method can still achieve good training results, which improves the convergence speed of the identification for network parameters. Simulation results show that the proposed model can well predict the economic indicators of the flotation process and meet the optimization computing requirements of flotation reagents addition.