钢芯输送带为煤矿主要运输设备，由于钢芯输送带长期、高负荷运行，内部钢丝绳芯和接头会产生损伤，最终导致横向断带事故发生。传统的钢芯输送带缺陷检测方法存在检测效率低、评定缺陷难等不足。磁检测为一种先进的钢芯输送带缺陷在线检测方法，可以有效预防断带事故发生。因此，研制一种基于高灵敏度磁阻传感器的钢芯输送带缺陷在线检测系统对煤矿安全生产具有重要的现实意义。 针对钢芯输送带较宽和磁阻传感器在钢芯输送带宽度方向检测范围有限的问题，本文采用模块化设计思想，借助弱磁检测技术、计算机技术、信号处理技术和通讯技术等，设计了一种基于多磁阻传感器的钢芯输送带缺陷检测系统总体方案，该方案实现了钢芯输送带缺陷的在线检测，较大地提高了检测效率。 在硬件设计方面，系统以微处理器STM32F103为核心，研发的硬件电路主要包括传感器检测电路、放大电路、信号采集电路和通讯电路。针对钢芯输送带缺陷磁阻传感器检测信号比较微弱问题，采用高性能仪表放大器AD623设计了弱磁信号放大电路，实现了缺陷弱磁信号有效放大。设计了一种置位/复位电路，有效地提高了磁阻传感器抗强磁场干扰能力。设计了基于SPI总线的多磁阻传感器信号采集电路，多片A/D芯片TLC2543通过SPI总线与微处理器STM32F103通讯，实现了多磁阻传感器信号高速、准确采集。 在软件设计方面，采用模块化设计思想，开发的软件系统主要包括信号采集模块、信号处理模块和信号传输模块。针对钢芯输送带缺陷实时采集、存储和降噪问题，运用双缓冲技术实现了弱磁检测缺陷数据的实时存储和快速传输，并采用平均值软件滤波方法对采集信号进行滤波，保证了缺陷检测数据的可靠性。针对多磁阻传感器大量数据高效传输问题，采用以太网的TCP/IP协议，实现了钢芯输送带缺陷弱磁信号高速、稳定传输。 为了验证设计的基于磁阻传感器的钢芯输送带缺陷检测系统可行性，运用建立的弱磁检测系统对钢芯输送带内部钢丝绳芯断绳缺陷和接头位移缺陷进行检测试验，实验结果表明：磁阻传感检测系统实现了钢芯输送带断绳和接头位移缺陷信号快速、准确采集，并根据试验分析得出了传感器安装个数和安装位置。

Presently, steel cord belt conveyor is main transportation equipment of most coal mines, often works in an environment of long-term and high-load, which will lead to the damage of the internal steel cord and connector, even the accident of transverse broken belt. The traditional detection method in steel core conveyer belt has some disadvantages, such as low inspective efficiency, is difficult to evaluate defects etc. As an advanced defect detection method of steel core conveyer, magnetic detection can effectively prevent the broken belt accicent.As a result, Designing a system of real-time online detection of defects for steel core conveyer belt based on high sensitivity magnetoresistive sensor has great importance signification. Aimed at the problem of magnetoresistive sensor has a limited measuring range in the direction of steel core conveyor belt width and steel core conveyor belt has a longer width, according to the idea of module integration, the detection system in steel core conveyer belt has been developed by weak magnetic detection technology, computer technology, signal processing and communication technologies etc. The system can realize online detection of the steel core conveyer belt and greatly improve the detection efficiency. In hardware development, based on the microprocessor STM32F103, the hardware circuit mainly involves Sensor circuit, amplifying circuit, signal acquisition circuit and communication circuit. Aimed at the problem of steel-cored conveyer belt magnetic singnal from the magnetic resistance sensor is weak, a high performance instrumentation amplifier AD623 is proposed to amplify the weakening magnetic signal. A set/reset circuit is designed, which effectively improve anti-strong magnetic interference of the magnetoresistive sensor. A multi-magnetoresistive sensor signal acquisition circuit is designed based on SPI bus, and a TLC2543 with multi-chip A/D chip communicate with a microprocessor STM32F103 via SPI bus, which make the multi-speed, accurate acquisition of the magnetoresistive sensor signal become possible. In software development, according to the idea of module integration, which includes signal acquisition program module, signal processing module and signal transmission module. Aimed at the detection problem of real-time collection, storage and noise reduction in the steel cord conveyor belts, the application of D-buffering is used to realize real time storage of testing defects data and the average software filtering method to filter the signal, which ensures the reliability of the detection data. Aimed at the detection problem of large amounts of data transmission rate and transmission distance. The ethernet transmission technology is analyzed and a network communication module is established. To verify the feasibility of the steel core conveyer belt defect detection system based on magnetoresistive sensor, the weak magnetic detection system is established to test the defects in steel cord of the internal steel cord and displacement defect in joint, which indicates that the system achieves quickly and accurately capture defect singnals in broken rope of steel core conveyer and fittings, and the number of installation and mounting positions of the sensor are also analyzed by experiment.