Chaos optimization algorithm has the features of
ergodicity, randomness and inherent regularity. It can traverse
all of the state by its own laws and won't repeat[5] .Chaos
optimization algorithm can avoid falling into local minima,
has more advantages than the random search , and is easy to
jump out of local optimal solution[6]. In paper [7], it used the
chaos optimization algorithm for power system economic load
dispatch, compared with genetic algorithm, and proved the
superiority of chaos optimization algorithm. In paper [8], it
first proposed the mutative scale chaos optimization
algorithm, and verified the effectiveness of the proposed
algorithm through numerical examples. In paper [9], it
improved mutative scale chaos optimization algorithm,
applied the improved algorithm in the example of large-scale
power system economic load dispatch, and achieved better
results compared with other algorithms.
This paper studies the influences of initial value, the
variable space narrowing coefficient and the "secondary
search" adjustLnent coefficient of the mutative scale chaos
optimization algorithm, and proposes improvements. Through
some numerical examples, this paper verifies the effectiveness
of the improvements. Combined with the characteristics of the
micro-sources, this paper establishes a microgrid optimization
run mathematical model, and applies the improved algorithm
to optimize. The simulation certifies that the desired result is
available.