Data Acquisition Strategy Research And Implementation Of The Vibration Signal Analyzer

Collecting data of the rotating machinery and realizing real-time monitoring and fault diagnosis is an important guarantee of safety production in related field. In the paper, fault monitoring technology on rotating machinery is studied, a portable vibration signal analyzer used for operation monitoring of steam turbine is developed, and the data acquisition strategy is researched and implemented.In view of the need to collect24road vibration signals in real time, the design scheme of distributed processing with multiprocessor is established. The scheme is based on embedded technology, and divides the analyzer into data processing unit and data acquisition unit. The data acquisition unit is responsible for collecting vibration signal and controlling gain, regulating, multi-channel synchronous acquisition and processing; the data processing unit is used for data calculation, analysis and storage. The high-speed data transmission between the two units is realized by using shared memory. According to the requirement of vibration monitoring that the equipment running in the high-speed stable state and the unstable state, two signal acquisition strategies are established, including fixed frequency sampling and multiple frequency sampling under the control of key-phase signal. The paper mainly discusses the development of data acquisition unit. On the design of the hardware, Bently3300series sensors are used for vibration signal acquisition, the ARM processor LPC1768that kernel is Cortex-M3is used as the core device, and the circuits of signal conditioning, A/D conversion, key-phase signal plastic and shared memory are designed. On the design of the software, Keil-uVision4is used as development environment and the two sampling methods are programmed. On the vibration signal filtering algorithm, in view of the actual situation that the vibration signal processing needs strict phase, filtering algorithm based on oversampling ADC is proposed, and it can guarantee the linear phase of filter processing.The vibration signal analyzer designed in the paper can operate steadily and meet the requirements of the indicators after being tested. Theory research and equipment development practice show that the design of distributed and collaborative processing with multiple processors based on embedded technology can evenly distribute load and meet the demand of real-time and portability, the methods that fixed frequency sampling and multiple frequency sampling under the control of key-phase signal can realize flexible configuration between signal collecting and signal processing, and the filtering algorithm based on oversampling ADC can not only ensure the linear phase but also remove noise well, meeting the need of the filtering of vibration signal.