论文部分内容阅读
遗传算法常规二进制编码在求解高维最优化问题时，因模型参数量太大会使其二进制码占用巨大的内存，并需进行大量的译码工作，甚至有可能出现影响遗传算法实现的问题。为此，我们提出了一种实用有效的隐形二进制优化编码方案。当遗传算法应用于高维反演问题时，这种编码方法不仅能将参数占用的内存减少到最低的限度，而且同常规的二进制编码法相比，还能几倍地减少参数译码的计算工作量。
Genetic Algorithm Conventional Binary Coding In solving high-dimensional optimization problems, the large amount of model parameters causes the binary code to occupy a large amount of memory, which requires a lot of decoding work and may even affect the implementation of genetic algorithms. To this end, we propose a practical and effective invisible binary optimization coding scheme. When the genetic algorithm is applied to high-dimensional inversion problems, this coding method can not only reduce the memory occupied by the parameters to a minimum, but also reduce the computation of parameter decoding several times compared with the conventional binary encoding method the amount.