基于支持向量机和粒子群算法的生物质气化过程建模与优化

The biomass gasification process modeling and optimization based on SVM and PSO

生物质气化过程的最终目标就是尽可能得到更多的高品质可燃气体。目前国内外缺乏对生物质气化过程参数优化问题的研究,在实际气化过程中燃气品质难以保证从而对燃用气设备产生了不利的影响,降低了燃气的利用价值。为此建立了一种能适应生物质(竹子)气化过程的支持向量机模型用于预测生物质气化气组分、气体热值及气体产率等气化指标。在此模型基础上,采用MOPSO算法寻找最优控制参数当量比ER和气化温度T,使得气体热值和气体产率两个目标折中并在一定程度上都趋近于最大化。通过生物质料竹子为例的计算验证,得到了满意的结果,即在保证气化指标的同时可得到一组最优的控制参数。

The final purpose of the biomaas gasification progress is to obtain high-quality flammable gas as much as possible. At present, it is seldom to research the biomas,s gasification of parameters optimization problems. It is difficult to guarantee the quality of flammable gas in the practice of gasification process while it has a negative impact on the burning equipments by gas, reducing the utility value of flammable gas. Therefore, this paper establishes an SVM model which adapts to the biomass （bamboo） gasification progress so as to predict gas component of biomass, Lower Heating Value （heat value）, gas yield and other indicators of biomass gasification. On the basis of this model, MOPSO algorithm is used to find the yield optimal control parameters （ER and T） while the heat value of gas and gas yield compromise are relatively close to maximize to a certain extent. Through the biological material the paper verifies the calculation of bamboo as an example. The results show that gasification gas targets can be guaranteed at the same time a group of the best available control parameters can be got.