Based on the inquiry accuracy, inquiry completion and the response time of the combination and operation optimization of the multi-independent research engine, combined with the feature of the research engine operation, the multi-independent research engine is put in to dynamic optimization. Borrowed the thought of niches, an improved niche genetic algorithm of the multi-objective assembled optimized operation is put forward by combining the technology of niches and the genetic algorithm. This algorithm enables each individual to have local optimization-oriented operation ensures the diversity of the group, strengthens the capability of local research and restrains the prematurity. When many subobjects cannot reach optimization at the same time, this algorithm, in use of individual comprehensive sufficiency can enforce weighting each objective function harmonize and optimize each research engine combination in order to fine a good solution to the combination and operation sequence of research engine. Results of the experiments prove that this Algorithm can improve the efficiency of research engine operation and it surpasses the common inquiry technique in the inquiry accuracy and calculation speed.