A thermal power plant of Sinopec has 9 boilers, which generally have problems of high exhaust gas temperature and high flying ash carbon content. In order to improve the adaptability of coals, the stability of coal po...A thermal power plant of Sinopec has 9 boilers, which generally have problems of high exhaust gas temperature and high flying ash carbon content. In order to improve the adaptability of coals, the stability of coal powder ignition, the burn-off rate of pulverized coals and the boiler efficiency, a series of renovation projects about importing hot air into mill exhauster are proposed. For the sake of verifying the renovation effects, an efficiency performance test is conducted on the renovated #5 boiler. The test result shows that the boiler heat efficiency has improved by 0.4% and it operates more safely and reliably after the renovation. At last, this paper recommends an optimized operation mode.展开更多
考虑潜在高价值旅客特有的数据高度不平衡、旅客特征和价值类别弱相关等问题,提出一种基于三重混合采样和集成学习的潜在高价值旅客发现模型。采用RFM(Recency Frequency Monetary)方法标注旅客类别;使用三重混合采样对不平衡旅客数据...考虑潜在高价值旅客特有的数据高度不平衡、旅客特征和价值类别弱相关等问题,提出一种基于三重混合采样和集成学习的潜在高价值旅客发现模型。采用RFM(Recency Frequency Monetary)方法标注旅客类别;使用三重混合采样对不平衡旅客数据集进行重采样;使用融合特征选择算法遴选旅客特征;使用梯度提升决策树作为分类器,构建旅客价值预测模型,识别潜在高价值旅客。在PNR数据集上的实验结果表明,与基准算法相比,该模型能取得更好的AUC值和F1值,可以较好地识别潜在高价值旅客。展开更多
文摘A thermal power plant of Sinopec has 9 boilers, which generally have problems of high exhaust gas temperature and high flying ash carbon content. In order to improve the adaptability of coals, the stability of coal powder ignition, the burn-off rate of pulverized coals and the boiler efficiency, a series of renovation projects about importing hot air into mill exhauster are proposed. For the sake of verifying the renovation effects, an efficiency performance test is conducted on the renovated #5 boiler. The test result shows that the boiler heat efficiency has improved by 0.4% and it operates more safely and reliably after the renovation. At last, this paper recommends an optimized operation mode.
文摘考虑潜在高价值旅客特有的数据高度不平衡、旅客特征和价值类别弱相关等问题,提出一种基于三重混合采样和集成学习的潜在高价值旅客发现模型。采用RFM(Recency Frequency Monetary)方法标注旅客类别;使用三重混合采样对不平衡旅客数据集进行重采样;使用融合特征选择算法遴选旅客特征;使用梯度提升决策树作为分类器,构建旅客价值预测模型,识别潜在高价值旅客。在PNR数据集上的实验结果表明,与基准算法相比,该模型能取得更好的AUC值和F1值,可以较好地识别潜在高价值旅客。
