基于PCA-SVM的高含硫油气混输管道腐蚀预测

为预测高含硫油气混输管道腐蚀情况,分析了导致腐蚀的原因,归纳影响腐蚀的因素。采用主成分分析法(PCA)对各种因素进行优选,摒弃相关联但贡献率较低的因素。将贡献率较大的因素作为支持向量机(SVM)的输入变量,以腐蚀率作为目标输出函数,建立管道腐蚀预测模型,并进行管道腐蚀率预测。以某公司川中地区8条运营管线为例,验证SVM方法预测管道腐蚀率的效果。结果表明,PCA-SVM方法的预测数据平均相对误差较小、吻合度高,预测结果符合实际情况。

Study on characteristics of pipeline transportation and sulfur fixing of cleaned coal logs

As special cylindrical briquettes of coal for long distance pipeline transportation and directly cleaned combustion the cleaned coal logs should possess two characteristics of transportation in pipeline and cleaned combustion. In order to make cleaned coal logs a rational technology for manufacturing, cleaned coal logs was designed and compound sulfur fixing binders with high effects of binding and sulfur-fixing was selected and combined. In addition, by means of characteristic experiments of strength, wear, waterproof and sulfur-fixing five different cleaned coal logs made with different compound sulfur fixing binders in different compaction conditions was tested and measured. Experimental results indicated that the manufacturing technology of cleaned coal logs was reasonable and the combination of compound sulfur fixing binders was scientific. Cleaned coal logs made up with the fourth group of coal mixture had high strength, good waterproof property, efficient sulfur-fixing, good characteristic of transportation, and achieved the performance requirement for pipeline transportation and sulfur fixing.