基于k-Means算法的生活垃圾焚烧设施烟气二恶英防控

为深入探究生活垃圾焚烧设施二恶英的防控方法，采用k-Means算法对华南某生活垃圾焚烧发电厂焚烧设施的运行状态和不同炉内状态下二恶英排放特征及控制机理进行研究。结果表明，该厂3台焚烧炉具有显著不同的运行状态 (类别1和2) (p<0.05) ，在不同状态下，其常规污染物及二恶英排放特征差异显著 (p<0.05) ，类别1炉内状态较差，其排放的CO质量浓度是类别2状态下的2倍，二恶英的质量浓度是类别2状态下的1.5倍。类别1和类别2状态下二恶英的主导生成机理分别为“前驱物合成反应”和“从头合成反应”。为降低二恶英的排放，该厂首先应保证烟气流量大于10.5×10⁴ m³·h⁻¹ (均值11.8×10⁴ m³·h⁻¹) ，其余参数最优范围为：烟气含氧量大于7.0% (均值8.6%) 、CO质量浓度低于9.0 mg·m⁻³ (均值3.5 mg·m^-3) 、烟气湿度小于23.7% (均值21.3%) 、烟气温度为140~154 ℃ (均值145.7 ℃) 。当通过运行控制手段无法保持上述参数控制范围时，应分析原因并考虑停炉检修。本研究结果可为优化焚烧过程、降低二恶英排放、生活垃圾焚烧发电厂长期稳定运行的判别和精细化管理提供参考。

Prevention and control of dioxins emission in domestic waste incineration facilities based on k-Means algorithm

In order to further explore the prevention and control of dioxins emission from domestic waste incineration facilities, pollution characteristics and control mechanisms of dioxins emitted from a domestic waste incineration power plant in South China under different operation status of incineration facilities and different furnace conditions were studied by k-Means algorithm. The results showed that three incinerators in the plant had significantly 2 different operation status (Status 1 and 2) (p<0.05), with distinct emissions of air pollutants including conventional pollutants and dioxins. The mass concentrations of CO and ∑PCDD/DFs were 2 times and 1.5 times higher, respectively, in Status 1 than Status 2. The dominant formation mechanisms of dioxins in status 1 and status 2 were the precursor synthesis reaction and the de novo synthesis reaction, respectively. To reduce the dioxins emission, the plant should firstly ensure that the smoke volume should be above 10.5×104 m3·h−1, and then guarantee that the other parameters of domestic waste incineration are in the following optimal range: flue gas oxygen content above 7%, carbon monoxide concentration below 9.0 mg·m−3, flue gas temperature between 140 ℃ and 154 ℃, and flue gas humidity below 23.7%. The incinerator should be shut down for maintenance when the limits of the optimal parameters are exceeded. The results could provide a scientific support for optimizing the incineration process and reducing dioxin emission, and also provide a theoretical basis for the identification and fine management of long-term stable operation of domestic waste incineration power plants.