基于改进人工鱼群算法的LED混光方法

LED Color Mixing Design Based on Improved Artificial Fish Swarm Algorithm

为寻求高LED颗粒利用率和高光效,引入了新LED组合进行混光,提出利用改进人工鱼群算法优化LED光通量配比,并设计了一款可调光调色筒灯来验证算法的可行性。引入相对基强化学习算法优化人工鱼种群;引入柯西分布和正态分布函数作为自适应算子,优化视域和步长。用积分球测试经优化后的灯具,数据表明灯具具有很高的颗粒利用率和光效。结果表明本文提出方法切实可行,具有很高的实用价值。

A novel design of LED mixing color based on improved artificial fish swarm algorithm(AFSA) was proposed. In order to improve convergence rate and accuracy of AFSA,oppositionbased reinforcement learning was utilized to optimize solution space of AFSA,Normal and Cauchy distribution function were employed as adaptive weight operators to optimize vision field and stepsize. The white,amber and blue LEDs were chosen to design LED lamp samples. LED proportions of 4 000 K color temperature among those LEDs were optimized by AFSA and new AFSA in the Matlab. Simulations indicate that improved AFSA has the higher convergence rate and accuracy. After LED lamps were tested by integrating sphere,the results show that the difference between the actual value and simulation calculation value is tiny,and the improved AFSA is effective. The improved AFSA provides a new efficient method for the calculation of LED proportions. Compared with the traditional method,the new method not only saves a significant amount of time,but also achieves higher luminous efficacy for lamps. All these show that the new method is effective and has high practical value.