摘要
牛粪高温厌氧发酵过程通常是严重非线性和时变的复杂动态系统,建立发酵过程沼气产量的精确模型具有很大难度,而且缺少测量重要过程参数的在线测量仪表。为了解决上述问题,本研究采用最小二乘支持向量机(LS-SVM),依据生产过程的温度、固体浓度、进出牛粪量和发酵体积等数据,对牛粪高温厌氧发酵沼气产量进行建模方法研究,给出了最小二乘支持向量机参数的调整策略和分析结果,建立了沼气产量的实时在线预估模型。结果表明用LS-SVM建立的在线预报模型误差小、方法简单、推广性能好,可以作为发酵过程的进一步控制和优化的参考依据。
The cattle manure Anaerobic -Thermophilic fermentation processes are usually characterized as seriously time var- ying and nonlinear dynamic systems. It is difficult to model the fermentation biogas output precisely. Furthermore important process parameters online measuring instruments is falling. In order to solve the above - mentioned problems, the paper uses least squares support vector machines. An approach via least squares support vector machines based on pilot experimental data is proposed for modeling the cattle manure Anaerobic - Thermophilic fermentation process, and the adjusted strategy for parameters of LS--SVM is presented. Based On the proposed modeling method,the predictive models of biogas output are obtained by using very limited on - 1 ine measurements. The results show that the models established are more accurate and efficient, and suffice for the requirements of control and optimization for biochemical processes.
出处
《黑龙江科学》
2013年第2期40-42,共3页
Heilongjiang Science
关键词
厌氧发酵
沼气产量
最小二乘支持向量机
建模
Anaerobic fermentation
biogas output
least squares support vector machine