作者：吴新忠, 耿柯, 陈昌

作者单位：中国矿业大学信息与控制工程学院，江苏 徐州 221008

关键词：风压传感器;温度补偿;蝗虫优化算法;径向基神经网络

摘要：

针对矿用风压传感器受环境温度影响导致风压测量不准确的问题，提出基于改进蝗虫算法优化径向基神经网络(IGOA-RBF)的误差补偿方法，来消除环境温度影响所带来的零点漂移和灵敏度漂移。首先建立基于RBF神经网络的温度补偿模型，利用蝗虫算法(GOA)对RBF的网络初始权值、激活函数的数据中心及扩展常数进行优化，来提高模型的补偿精度；进一步地，分别利用佳点集进行种群初始化、非线性自适应参数策略平衡搜索能力来改善GOA寻优质量；最后通过实验数据对补偿模型进行验证。结果表明，灵敏度温度漂移系数和零点温度漂移系数均降低一个数量级，且经IGOA-RBF补偿后的最大相对误差比经RBF补偿后的降低1.4%，测量精度明显提高。

Research on temperature compensation of mine wind pressure sensor based on IGOA-RBF neural network
WU Xinzhong, GENG Ke, CHEN Chang
College of Information and Control Engineering, China University of Mining and Technology, Xuzhou 221008, China
Abstract: Aiming at the problem that the mine wind pressure sensor is not accurate due to the influence of environmental temperature, an error compensation method based on improved grasshopper algorithm and radial basis function neural network (IGOA-RBF) is proposed to eliminate the zero drift and sensitivity drift caused by environmental temperature. Firstly, a temperature compensation model based on RBF neural network is established. The initial weight of RBF network, the data center and the expansion constant of activation function are optimized by using GOA algorithm to improve the compensation accuracy of the model. Furthermore, the optimal point set is used to initialize the population and the nonlinear adaptive parameter strategy is used to balance the search ability to improve the quality of GOA optimization. Finally, the compensation model is verified by the experimental data. The results show that the sensitivity temperature drift coefficient and zero temperature drift coefficient are reduced by an order of magnitude, and the maximum relative error compensated by IGOA-RBF algorithm is reduced by 1.4% compared with that compensated by RBF algorithm, and the measurement accuracy is significantly improved.
Keywords: wind pressure sensor;temperature compensation;grasshopper optimization algorithm;radial basis function neural network
2021, 47(6):137-143  收稿日期: 2020-11-13;收到修改稿日期: 2021-01-11
基金项目: 国家重点研发项目（2018YFC0808100）
作者简介: 吴新忠(1976-),男,江苏徐州市人,副教授,博士,主要研究方向为机电设备的状态监测、面向应用的智能控制、检测与传感器技术、矿山智能通风等
参考文献
[1] 周福宝, 魏连江, 夏同强, 等. 矿井智能通风原理、关键技术及其初步实现[J]. 煤炭学报, 2020, 45(6): 2225-223
[2] 李金娥, 王丽, 薛斌. 温湿度对煤矿井下电气设备的影响[J]. 煤矿机械, 2014, 35(9): 76-78
[3] 党瑞荣, 张宏伟, 宋楠, 等. 高温高压井下压力传感器的补偿与校正[J]. 仪器仪表学报, 2016, 37(4): 737-743
[4] WU T S, CHEN S Y, WU P. A high precision software compensation algorithm for silicon piezoresistive pressure sensor[J]. Chinese Journal of Electronics, 2019, 28(4): 748-753
[5] 郭志君, 卢文科, 左锋, 等. 基于NSGA-Ⅱ& BP的应变片式压力传感器温度补偿研究[J]. 中国测试, 2020, 46(6): 72-77
[6] 崔萌洁, 卢文科, 左锋. 基于PSO-BP模型的扩散硅压力传感器温度补偿[J]. 中国测试, 2019, 45(11): 95-100+125
[7] 张小俊, 张明路, 李小慧. 基于RBF神经网络的电化学CO气体传感器的温度补偿[J]. 传感技术学报, 2009, 22(1): 11-14
[8] ALAA T, HOUSSEIN E H, AHMED M M. MOGOA algorithm for constrained and unconstrained multi-objective optimization problems[J]. Appl Intell, 2018, 48: 2268-2283
[9] HEIDARI A A, FARIS H, ALJARAH I, et al. An efficient hybrid multilayer perceptron neural network with grasshopper optimization[J]. Soft Computing, 2019, 23(17): 7941-7958
[10] LIANG H B, CHEN H F, LU Y J. Research on sensor error compensation of comprehensive logging unit based on machine learning[J]. Journal of Intelligent & Fuzzy Systems, 2019, 37(3): 3113-3123
[11] SAREMI S, MIRJALILI S, LEWIS A. Grasshopper optimization algorithm: Theory and Application[J]. Advances in Engineering Software, 2017, 105: 30-47
[12] 何庆, 林杰, 徐航. 混合柯西变异和均匀分布的蝗虫优化算法[J/OL]. 控制与决策:1-10[2021-01-10].https://doi.org/10.13195/j.kzyjc.2019.1609.
[13] 徐松金, 龙文. 嵌入遗传算子的改进灰狼优化算法[J]. 兰州理工大学学报, 2016, 42(4): 102-108
[14] 李洋州, 顾磊. 基于曲线自适应和模拟退火的蝗虫优化算法[J]. 计算机应用研究, 2019, 36(12): 3637-3643
[15] ZHANG J H, WU Y S, LIU Q Q, et al. Research on high-precision, low cost piezoresistive MEMS-array pressure transmitters based on genetic wavelet neural networks for meteorological measurements[J]. Micromachines, 2015, 6(5): 554-573