造纸废水处理中温室气体减排的溶解氧智能优化控制

Optimal Control of Dissolved Oxygen for Greenhouse Gas Reduction from Wastewater Treatment Plant in Paper Mill

关键词:  温室气体  造纸废水  优化控制  溶解氧

Key Words:greenhouse gas  wastewater from papermaking mill  optimum control  dissolved oxygen

摘要:造纸工业普遍采用活性污泥法处理废水，然而对该过程中温室气体排放却关注很少，缺少温室气体在线监测和减排措施。针对这一问题，提出了一种基于自适应回归神经网络PI控制（Adapted Kernel Regression Back Propagation Neural Network-Proportional Integral Control，AKRBP-PI）的溶解氧分层优化控制策略，旨在保证出水质量的同时，减少温室气体排放。该优化控制基于溶解氧对温室气体排放的作用机制，采用分层思想，根据出水污染物含量与溶解氧的函数约束关系，采用遗传算法求解溶解氧优化设定值，实现神经网络跟踪控制。仿真结果表明，对比开环控制，AKRBP-PI方案的出水污染物含量均未超标，温室气体总排放量减少了8.6%。其中，减少的温室气体主要来自曝气机耗电量的下降。

Abstract:Activated sludge process is widely used for wastewater treatment in pulp and paper industry. A mass of greenhouse gas (GHG) generates from the processing, which merely gets a handful of attention and lacks of on-line detection and mitigation. To solve this problem, a layered optimization dissolved oxygen (DO) control strategy based on adaptive kernel regression propagation neural network proportional integral (AKRBP-PI) control was proposed. The object of the study was to reduce the GHG emission in the premise of insuring effluent quality. The optimal control was based on the action mechanism of dissolved oxygen on greenhouse gas emissions and the idea of stratification was adopted. According to the functional constraint relation between pollutant content of effluent and dissolved oxygen, the set value of dissolved oxygen optimization was solved by genetic algorithm, and the neural network tracking control was realized. The simulation results showed that compared with the open-loop control, the effluent pollutant content of AKRBP-PI scheme met the regulation requirement, and the greenhouse gas emission reduced by 8.6%. Wherein, the reduction of greenhouse gases mainly came from the reduction of power consumption of the aerator.