Abstract: In order to solve the problem that the eigenvector method cannot deal with complex reaction networks and that the equal time method is easy to appear negative and imaginary numbers, the reaction network kinetic model was improved. Firstly, the concentration of components was linearly independent solutions of component concentration to ensure the existence of the inverse matrix. Secondly, the real concentration was appropriately transformed into a virtual concentration to simplify the expression of concentration-time equation. Finally, the model algorithm was improved by using the concept of equal time. The results showed that in the improved model, the equilibrium concentration of components was determined only by the rate coefficient, which was independent of the initial concentration. With the decrease of the distribution entropy of rate coefficient, the middle of concentration-time curve of the components moved up, but the initial and terminal values did not change, and the improved algorithm made the calculation of the rate constant go on smoothly. The calculation results could be more reasonable by adjusting the value of △t in the matrix decomposition equation. When △t = 0.5, the mean deviation between the calculated concentration and the experimental value was the smallest, which showed that the new model made up for the shortcomings of the equal time method and the improved algorithm was effective and feasible.

Key words: reaction network, reaction kinetics, rate coefficient, eigenvector method, equal time method