为了解决综采放顶煤过程的煤矸识别问题，避免过放和欠放情况的发生，将多传感器信息融合技术应用于参数化模型的建立。计算分析了各阶IMF分量的总能量、EMD能量熵和峭度，进而发现它们与煤矸含量的关系，确定了合理的特征参数。建立了基于规则的知识库形式的产生式数据融合算法，完成了煤矸的分类识别。将方法应用到放顶煤煤矸界面识别的试验中，取得了良好的识别效果，验证了所提出方法的有效性。%To solve the issue of gangue identification in fully-mechanized top coal caving process, and avoid the over caving and under caving, the multi-sensor information confusion technology has been used in establishing the parameterization model. The total energy of each IMF component,the EMD energy entropy and kurtosis are calculated and analyzed, thus the relationship among them and content of gangue is found out, and reasonable characteristic parameters are determined. The generative data fusion algorithm is established based on knowledge base in the form of rules, to accomplish the classification and identification of gangue. Applied in the test of top coal caving gangue interface identification, excellent effect is obtained, the effectiveness of the method proposed is verified.
展开▼
