INTEGRATION OF MATHEMATICAL-MODELING AND KNOWLEDGE-BASED SYSTEMS FOR SIMULATIONS OF BIOCHEMICAL PROCESSES

In this work, a hybrid modelling scheme that integrates available math ematical description with artificial intelligence techniques was devel oped to simulate complex biochemical processes. The hybrid model featu res the parallel arrangement of a numeric module and a knowledge-based module. The numeric module was composed of 44 process variables and 1 45 model parameters, which involves algorithmic calculation of the dif ferential equations for the metabolic behaviour of the biochemical pro cesses. The knowledge-based module was designed to carry out frame-bas ed inference to determine the cause of the modelling discrepancy using the relationship between the derivation of the process variables and the model parameters. Two types of frames (e.g., 8 frames for the vari able error information and 14 frames for the parameter adjustment) wer e built to support the decision-making process. This novel hybrid mode lling method was evaluated by comparing tire predictions of a hybrid m odel with conventional mathematical modelling. The result has shown th at by using the proposed scheme, the discrepancy between the model pre dictions and the experimental data can be reduced, and that the predic tive strength of the simulation ran be enhanced.