Abstract: Based on the evolutionary epistemology idea, an algorithm called UOP-KEA for solving unconstraint optimization problems utilizing knowledge evolution principle is proposed in this study. The main idea of this algorithm can be described as follows. Firstly, an initial knowledge base is formed. The next work is to inherit excellent knowledge individuals by inheritance operator, produce new knowledge individuals by innovation operator, update knowledge base by update operator and accordingly the knowledge evolution is realized. At last, the problem’s optimal solution can be gained from the optimal knowledge individual. Experiments were taken on optimization of unconstraint nonlinear test functions. The successful experimental results show that this algorithm is feasible and valid. The algorithm can search the global optimal solution with less population and less reiteration. The global convergence speed and the global optimal solution quality are all satisfactory.