线性化厌氧一号模型在干式厌氧系统中的应用

APPLICATION OF THE LINEARIZED ADM1 FOR A DRY ANAEROBIC DIGESTION SYSTEM

厌氧消化已被广泛应用于能源回收和固体废物处理。由于厌氧工艺的复杂性,在工业领域很难实现厌氧消化的优化运行。因此,厌氧消化工艺的有效控制方法已成为热门研究领域。厌氧一号模型(ADM1)是最常用的厌氧消化过程控制模型,但是该模型在准确性和灵敏度上有所欠缺。通过生化过程速率计算方法线性化修改对ADM1进行优化,ADM1和线性化ADM1(LADM)均使用来自干式厌氧中试系统的输入数据,并进行测试评估。结果表明:LADM的性能与ADM1非常相似,模型预测结果的平均相对误差(MRE)为0.013。在Matlab软件平台上分别使用ODE45和ODE15s 2种方法进行计算,LADM的计算成本相比ADM1分别降低18%和15%。将LADM应用于中试规模干式厌氧消化系统的模型预测控制(MPC)。结果表明:LADM对沼气产量的预测性能非常准确,5个MPC阶段的模型预测输出结果和实际数值对比的MRE分别为3.8%、4.2%、5.2%、5.4%和6.4%。

Anaerobic digestion(AD)has been widely applied for energy recycling and solid waste treatment.The optimal operation of AD is barely achieved in the industrial field because of the complexity of the AD process.Therefore,the effective control method for the AD process has become a popular research field.Anaerobic digestion model No.1(ADM1)is the most commonly applied model for process control of AD.However,it requires modification to improve accuracy and sensitivity.In this study,ADM1 was modified by the linearization of the calculation method in biochemical process rates.First,both ADM1 and the linearized ADM1(LADM)were evaluated with input data from a pilot-scale dryAD system.The results showed that the LADM presented a very similar performance to that of ADM1,as the mean relative error(MRE)of model prediction result was 0.013.However,the computing cost of the LADM was 18%and 15%lower than that of ADM1,as calculated by using ODE45 and ODE15s,respectively.Secondly,LADM was applied for the model predictive control(MPC)of the lab-scale AD system.LADM presented a very accurate prediction performance of the biogas production,while the MRE of model prediction output of 5 MPC phases was 3.8%,4.2%,5.2%,5.4%,and 6.4%,respectively.