【作者】 朱子健；

【导师】 王鑫伟；

【作者基本信息】 南京航空航天大学 ， 测试计量技术及仪器， 2005， 博士

【摘要】 传感器技术是高科技发展的重要基础,数字化、智能化是当前传感器领域发展的主导方向。本文主要研究称重传感器,结合目前国内外称重传感器的应用现状,设计了一种基于RS485通讯的数字式智能称重传感器,硬件设计中采用了AVR24位单片机,及高度集成的CS-5532 A/D芯片,重点在对称重传感器的各种主要技术参数的误差补偿、称重系统的数据处理及故障诊断等方面作了深入细致的研究,论文的主要工作及创新性点如下:1. 系统分析了称重传感器主要误差的成因及其应用场合,在此基础上,设计开发了 基于先进数据处理方法的高性能智能称重传感器。这种传感器吸收了模糊数学、 人工智能等方面的新成果,采用合理灵活的先进数据处理方法实现传感器误差的 数字补偿,避免了传统称重传感器中繁琐且精度不高的模拟的补偿方法及传统的 数学建模的数字补偿,精度可达 OIMLC6。2. 根据称重传感器的非线性与温度误差特性,将函数连接型神经网络引入称重传感 器的补偿设计中,提出了称重传感器的非线性及温度误差的神经网络补偿并给出 了易于单片机化的实现方法。研究表明:采用所提出的补偿方法精度高且灵活可 靠且方便易行。3. 根据称重传感器的蠕变特性,在传感器的设计中引入模糊理论,创造性地提出了 称重传感器蠕变的在线模糊补偿理论并给出了易于单片机化的实现方法。研究表 明:采用提出的补偿方法,传感器的蠕变补偿简单方便,可靠性、精度好。4. 开发了基于新型传感器的智能称重系统。在分析了现有称重系统的现状并保留传 统称重系统功能的同时,增加了免标定、系统故障自诊断、不间断工作,传感器 故障预报等新功能,智能化程度及称重精度都得到极大提高。对各项功能给出了 实现的方法及实例说明,这些方法具有较高的参考意义和实用价值。5. 在智能称重系统的设计中引入灰色理论,创造性地提出了称重传感器软故障的灰 色预测理论并给出了切实可行的实现方法。通过灰色建模,对传感器的软故障进 行预报,提高系统可靠性。 本文的研究结果包括数字式智能称重传感器、一套综合智能模型及几种有效的数据处理方法,对于智能称重传感器的发展和应用有重要的参考意义。更多还原

【Abstract】 Digitalization and intelligence are hot topices of transducer research and form animportant foundation for high-tech developments. Systematic research on errorcompensation, data processing, and fault diagnosis of an intelligent, digital load celltransducer are presented in this paper. Main works and creative conclusions are presentedas follows: 1. By systematically analyzing error formation and application of load cells, a highperformance intelligent digital load cell based on advanced data processing is developed.New flexible data processing methods based on new achievements of fuzzy mathematics,artificial intelligence and Gray theory are employed in the transducer design so that thecomplicated, low accuracy modeling in traditional designs can be avoided. These newmethods are suitable for high accuracy, up to OIMLC6. 2. A function link neural network was employed according to the characteristic ofthe nonlinear error and temperature error. The neural network compensation of the aboveerrors was achieved via micro controller. Results in this paper indicate that with abovecompensation the accuracy of a transducer could be improved greatly. Because of thesimplicity, reliability and flexibility of the above compensation algorithm, it is suitable formass production of load cells. 3. Fuzzy theory was applied to load cell creep compensation according to thecharacteristic of creep error. A new concept of creep fussy on-line compensation methodis presented. The result shows that this method is simple and reliable. It works well and canbe easily implemented through hardware. 4. Based on the digital load cell, a new smart weighing system is proposed in thispaper. Some intelligent characteristics such as free calibration, self- diagnostics, run flat,load cell fault prediction, have been discussed and some are realized. These functions aresignificant for weighing applications. 5. A new kind of load cell soft fault prediction method by the application of Graytheory to weighing is discussed in this paper. By using this method the reliability of asystem has been improved and it has a potential future in the field of measurement.更多还原