往复式压缩机是流程工业安全生产的关键机组,由于缺乏有效的安全监控和故障预知手段,往复式压缩机存在故障率高、安全事故频发的特点.为有效降低往复式压缩机故障停机时间并减少安全事故,在对压缩机运动部件结构、功能和故障机理分析的基础上,针对往复式压缩机振动激励源多和故障关联性强的特点,开发了基于多传感器信息融合和正向推理的往复式压缩机智能诊断专家系统,通过提取敏感特征参数并建立和故障类型相关的独立诊断规则,实现了自动故障诊断.建立的往复式压缩机智能诊断专家系统已应用于国内多家石油炼化企业.实践证明:往复式压缩机智能诊断专家系统在机组异常时能够自动报警并给出故障诊断结论,提高了设备预知维修水平,保证了往复式压缩机运行的安全性、可靠性.
Due to the lack of effective means of security monitoring and fault prediction, reciprocating compressor, which is the fundamentul units in process industry, has the characteristics of high failure rate and high frequently safety accidents. Reciprocating compressor has multi-vibration excitation sources and strong fault correlation, in order to reduce the downtime and accidents of reciprocating compressors effectively, an intelligent Diagnostic Expert System (IDES) is developed based on muhi-sensor information fusion and forward reasoning after the analysis of the structure, function and failure mechanism of reciprocating compressors. By extracting sensitive characteristic parameters and building the independent fault diagnosis rules related to fault type, automatic fault diagnosis is achieved. The established IDES has been applied in some domestic oil refining enterprises,the applications show that intelligent ularm and auto fault diagnosis are achieved, the predictive maintenance level is improved, and the safety and reliability of reciprocating compressors have dramatically secured.