多头组合秤的算法研究分析

Algorithm and Analysis of Multi-head Combination Weigher

目的为研究不同类型算法对提高多头组合秤称重精度和称重速度的影响。方法把动态规划算法和遗传算法应用于多头组合秤,通过对比不同数量的称量斗,2种算法在组合成功率和组合时间上的优劣,分析出最适合多头组合秤的算法。结果动态规划算法的组合时间明显少于遗传算法的组合时间。遗传算法的组合成功率一直保持着96%以上的高水平;动态规划算法在称量斗数量少于10时的组合成功率低于90%,但随称量斗数量的增加而明显提高。结论在少于20个称量斗的称重系统中采用遗传算法,虽然花费时间稍多,但是可以获得高组合成功率;在大于20个称量斗的称重系统中采用动态规划算法,既花费时间少,也能获得高组合成功率。

The work aims to study the influence of different types of algorithms on improving the weighing accuracy and speed of multi-head combination weigher.The dynamic programming algorithm and genetic algorithm were applied to the multi-head combination weigher.By comparing the merits and demerits of the two algorithms in combination success rate and combination time when there were different quantities of weighing weighers,the most suitable algorithm for the multi-head combination weigher was selected.The combination time of dynamic programming algorithm was significantly less than that of genetic algorithm.In addition,the combination success rate of genetic algorithm was kept at a high level of more than 96%and the combination success rate of dynamic programming algorithm was less than 90%when the number of weighing weighers was less than 10,but the success rate rapidly improved with the increase of the number of weighing weighers.The genetic algorithm can be used in the weighing system with weighing weighers less than 20,which takes a little more time but can get a high combination success rate.On the contrary,the dynamic programming algorithm can be used in the weighing system with more than 20 weighing weighers,which not only takes less time but also gets a high combination success rate.