【摘要】 以高效液相色谱与紫外光谱联用,分离并鉴定了煤液化重质产物中非沥青质的芳烃组成,得到了液化条件对转化率、非沥青质的芳烃环分布及非沥青质量的变化规律,并从另一角度阐明了煤液化的机理。同时采用人工神经网络模拟了煤液化过程中液化条件对非沥青质中的芳烃环分布的非线性关系。结果表明,兖州煤的加氢液化中,反应温度的变化对芳烃环分布的影响较大,当反应时间为30min时,在420℃左右,小环芳烃含量最多;反应时间的影响变化规律与温度的影响相同。实验证明:在实验条件下,人工神经网络能较好地模拟反应条件与产物组成之间的关系,并能较为准确地预报在不同反应条件下芳环的变化规律。更多还原

【Abstract】 A high performance liquid chromatographic method was developed and applied to identify and quantify the distribution of aromatic rings of heavy products from coal liquefaction The effects of liquefaction conditions on THF conversion、T conversion、aromatic rings distribution and heavy oil weight were analyzed An artificial neural network (ANN) method based on feedforward algorithm was used to simulate the nonlinear process of the distribution of aromatic rings and coal liquefaction Three liquefaction variables, content of recycle solvent, reaction temperatures (375～450℃) and reaction time (30～120min) were selected as inputs, and the content of 1+2-ring and 3+4+5-ring as outputs of the network respectively The best topology was determined by the leave-one-out cross validation method, and the mean squared deviation was used as a criterion to compare the network performance with the training data The results showed that the network with a single eight-node hidden layer could represent the process with reasonable accuracy In this experiment system, reaction time affects the distribution of aromatic ring markedly, the optimum reaction temperature is about 420℃ At this liquefaction condition, the content of 1+2-ring aromatics is higher, 3+4+5-ring aromatics is lower The relation of conversion-temperature supports this result更多还原