Prevent Downtime for Rotating Machine in The Petrochemical Industry
Most petrochemical plants have been operarting for more than ten to twenty years. The high cost of machine replacement and maintenance often results in delays or a failure of timely machine refurbishment. Among all possible public safety issues involving such plants, mechanical failure is one of the most important. Accidents can be caused by a faulty pump, compressor, or a valve that leaks flammable fuel, or exhaust circulation might activate an explosion. After this, highly corrosive and poisonous raw material as well as products can cause severe damage to people and the environment.
The petrochemical industry is mostly long-term continuous production, which cannot readily be stopped for maintenance or repair. Intensive manual patrol and inspection is time consuming. The results of the inspection often rely on personal experience, so there is no guarantee that every patrol can always accurately verify whether machines are operating normally. To ensure steady and reliable operation of motors, pumps, compressors, blowers, etc, a PHM (Prognostic and Health Management)-embedded predictive maintenance system is needed to address the following issues:
- For continuous production, any unexpected downtime will result in widespread and complex impact on the entire production line and require lengthy restoration of production, not to mention quality issues of the batch product.
- Maintenance people may often need to spend a great deal of time traveling around to get confirmation and information in an urgent situation in facilities where machines are widely spread.
- There are many different facility machines, such as motors, pumps, decelerators, compressors, blowers, which supply water, electricity, gas, materials and so on in the factory. It would be very difficult to ensure that all the patrol personnel would be familiar enough with all these different machines to be capable of making accurate judgment about their real condition.
Advantech iFactory has developed the PHM/Rotating Machine SRP, a comprehensive SW/HW integration system that addresses such industrial application scenarios, which includes pre-trained machine diagnosis model. After installing the wireless sensors on-site and collecting data for 1 to 2 weeks, an AI health model can be quickly built for sensing and issuing accurate pre-warnings when the machine health condition reaches an early-stage degradation standard.
Furthermore, the AI Model will also provide machine degradation level monitoring and 7-day degradation trend prediction by comparing the level of degradation against healthy condition. Professional experience can be translated into AI indicators via objective and accurate observations, so on-site maintenance personnel can react early.
PHM/Rotating Machine SRP can be simply implemented on-site by collecting machine healthy data while maintaining normal production, and this complies with the requirements of industries with continuous production where any arbitrary stoppage of the line would be unacceptable. In addition to keeping on-site personnel informed about machine condition, transparent trending of machine health conditions can also help O&M personnel prepare maintenance-related work in advance. This reduces the waste of manpower by reducing repetitive patrol and manual inspection. Professional technicians may focus on their dedicated work and continue to play key roles in corporate development and transformation.
“Production and safety risks due to inaccurate machine maintenance are key concerns of the petrochemical industry. Advantech iFactory developes a PHM predictive maintenance solution for rotating facility machine which ensures stable production with low incident occurrence. The whole plant is operating efficiently and safe.”
– a Senior Manager in an eminent petrochemical factory.