In existing multi-population multi-objective cultural algorithms, information are exchanged among sub-populations by individuals. However, migrated individuals can not reflect the evolution information enough, which limits the evolution performance.In order to enhance the migration efficiency, a novel multi-population multi-objective cultural algorithm adopting knowledge migration is proposed. Implicit knowledge extracted from the evolution process of each sub-population directly reflects the information about dominant search space. By migrating the knowledge among sub-populations at the constant interval, the algorithm realizes more effective interaction with less communication cost. Taken benchmark functions as the examples, simulation results indicate that the algorithm can effectively obtain the Pareto-optimal sets of multi-objective optimization problems. The distribution performance is also improved.

Applying Wall tapping and roof sounding for coal blasting mining, excavation face to determine whether the status of roadway roof separation occurs. Firstly through beating the different parts of rock layers and record the received voice signal, which do the DTW feature extraction of the voice signal. Secondly based on the model database of constructed roof separated layer, determining the state of the roof combining with BP neural network recognition algorithm, after the extraction of DTW feature and the sequence data generated. Finally simulation and experiments show that identification method based on DTW, determining the state of the roof characteristic. Whether the roof separation of roadway occurs, to avoid the misjudgment problems of lack of experience in practical works.

The ground penetrating radar and radar wave propagation in the subsurface environment is very complex. All kinds of noise and clutter interference is very serious, and detection echo data is a variety of with clutter. Therefore, the key techniques of data processing is to suppress clutter processing of ground penetrating radar record data. Surfacelet transform can efficiently capture and represent local surface singularities with different sizes. In order to improve the reliability of 3D ground penetrating radar detection results and accuracy, this paper presents a three-dimensional ground penetrating radar signal denoising method based on Surfacelet transform. Using Surfacelet transform and 3D context model for ground penetrating radar (GPR) analog signal to denoising, the noise in the case of low signal noise ratio (SNR) still can obtain a better result, and the simulations prove the effectiveness of the method.