基于SATLBO与氧化率自修正的CaSO_(3)浓度预测

Prediction of Ca SO_(3) Concentration Based on SATLBO and Self-correction Oxidation Rate

作为表征湿法脱硫吸收塔内浆液氧化反应优劣与按需启停氧化风机的重要参考指标,CaSO_(3)浓度缺乏有效的在线预测手段,为此提出一种基于SATLBO(supervised adaptive teaching-learning-based optimization algorithm)与氧化率自修正的CaSO_(3)浓度预测方法。方法依据反应机理设计实时工况的自适应欠氧计算环节、反应过程的容积迟延环节以及吸收塔内CaSO_(3)的滚动积分环节,并结合寻优算法优化模型参数,最终建立了预测模型;针对所选教学优化算法不足,增加教师答疑及学生互学空间距离判定环节,应用标准函数测试,验证了改进算法优越性。

CaSO_(3)concentration is an important reference for characterizing slurry oxidation reaction in wet desulfurization absorption tower and intermittent running of oxidation fan on demand. There is no effective online prediction method for CaSO_(3)concentration. Therefore, a prediction method for CaSO_(3)concentration based on SATLBO and self-correction oxidation rate is proposed. According to the reaction mechanism, the adaptive oxygen deficiency calculation link in real-time condition, the volume delay link in the reaction process and the rolling integration link of CaSO_(3)in the absorption tower were designed. The optimization algorithm was applied to optimize the model parameters, and finally the prediction model was established. In view of the shortcomings of the selected optimization algorithm, teachers’ question answering and students’ mutual learning space distance judgment were added in the teaching-learning-based optimization algorithm(TLBO),and the standard function test verified the superiority of the improved algorithm.