基于改进人工蜂群算法的传感器温度补偿

摘要：针对红外气体传感器在工作时外界温度对测量精度影响较大的问题，提出一种基于自适应人工蜂群-BP人工神经网络(AABC - BP)温度补偿方法。原始人工蜂群算法在运算过程中容易随着迭代次数增加而丢失优质解,降低解的稳定性，引入自适应人工蜂群优化算法提高的算法的稳定性。通过不同测试函数对自适应人工蜂群算法进行性能测试对比，结果表明自适应人工蜂群算法全局搜索能力强、计算精度高且计算过程稳定。利用自适应人工蜂群算法对BP神经网络的阈值和权值进行优化。实验结果表明(AABC - BP)混合算法对红外气体传感器的温度补偿误差在5%以内。

Sensor Temperature Compensating Based on Improved ABC Algorithm

Abstract: For the influence of temperature on the measuring precision of infrared gas sensor, a method of gas sensor temperature compensation is proposed based on Adaptive Artificial Bee Colony Algorithm-Back Propagation Neural Network (AABC - BP). During computational processes, Artificial Bee Colony (ABC) algorithm is easy to lose high quality solutions as the number of iterations increases and to reduce the stability of the solutions. Therefore, AABC algorithm is introduced to improve the stability of the solutions. According to performance testing of AABC algorithm with different benchmark test functions, performance testing results show that the AABC algorithm is with strong global searching ability, high accuracy and good stability. The threshold value and weight of BP neural network are optimized by the AABC algorithm. Experimental results show that the AABC - BP algorithm can control the error less than 5%.