论文部分内容阅读
为了对电镀废水中的间硝基苯磺酸钠催化湿式过氧化氢氧化(CWPO)降解条件进行模拟及优化,建立了间硝基苯磺酸钠CWPO降解过程BP神经网络模型。经验证,模型预测值与试验值的平均相对偏差为0 81%,相关系数r和Nash-Suttcliffe模拟效率系数NSC分别为0.992 5和0.983 9。相对灵敏度分析表明,影响间硝基苯磺酸钠去除率(以TOC表示)的顺序从大到小为:反应温度、间硝基苯磺酸钠质量浓度、pH值、H_2O_2用量、反应时间、初始氧分压、催化剂用量。结合遗传算法以TOC去除率最高作为优化目标,分别对降解条件进行优化。经对比,带成本约束的优化降解结果(99.36%)比试验中的TOC去除率平均值(85.51%)提高了 10%以上,同时,优化后的降解成本(2.03元)相比无成本约束条件下的降解成本(2.38元)降低了近15%(0.35元)。
In order to simulate and optimize the degradation conditions of sodium nitrobenzene sulfonate in electroplating waste water (CWPO), a BP neural network model was established for CWPO degradation of sodium nitrobenzene sulfonate. The average relative deviation between model predictions and experimental values is 0 81%, and the correlation coefficient r and the Nash-Suttcliffe simulation efficiency coefficient NSC are 0.992 5 and 0.983 9 respectively. The relative sensitivity analysis showed that the order of affecting the removal rate of m-nitrobenzene sulfonate (expressed as TOC) is as follows: the reaction temperature, the concentration of sodium nitrobenzene sulfonate, the pH, the amount of H 2 O 2, the reaction time, Initial oxygen partial pressure, catalyst dosage. Combined with genetic algorithm, the highest TOC removal rate was taken as the optimization target, and the degradation conditions were optimized respectively. In contrast, the optimized degradation results with cost constraint (99.36%) increased more than 10% from the average TOC removal rate (85.51%) in the experiment, while the optimized degradation cost (2.03) Degradation costs under (2.38 yuan) decreased by nearly 15% (0.35 yuan).