An intelligence-expert system for bench blasting design and its application at the Lanjian iron mine

摘要

A new technique in which artificial intelligence and an expert system are applied to bench blasting design is put forth in this paper. The technique called "Intelligence-expert system for bench blasting design" (IESBBD) includes the optimization of blasting parameters, the auto-adjustment of blast hole layouts, the calculation of design parameters, the prediction and analysis of blasting effects, automatic plotting, and blast management. The composition and structure of the IESBBD system are discussed. The content and structure of the database and the knowledge base as well as their automatic renewal are described. Based upon blast hole layout principles, the forward and backward reasoning mechanisms for determining the spatial position of the blast holes are presented. The reasoning flowchart behind the auto-adjusting of blast hole layouts is also given. The feed forward network models for both the optimization of blasting parameters and the prediction of blasting effect are established by the neural network theory and their training process is determined. Rock mass blasting is considered as a grey system and the principal factors affecting the quality of rock blasting are analyzed by the grey correlation method. The main design content of the IESBBD system and its realization is introduced. The IESBBD package has been developed by means of Visual Basic and Access database. The IESBBD system has been tested and applied to production blasting at the Lanjian iron mine. The results of the six blasting experiments illustrate how the various kinds of design charts provided by the system be directly applied to a blasting operation to yield the desired production requirements. They provide a convenient basis for revising the design and improve operability.