Integration of multiple platforms for real-time remote model-based condition monitoring

Model-based condition monitoring has been demonstrated to have superior performance in process condition monitoring. However, it demands increased computational resources to support its more advanced intelligence. This requirement imposes difficulties in integration into an existing control system that typically has resources dedicated solely to control applications. To overcome this problem, this paper presents an integration of multiple platforms, in which tasks are allocated in different machines with different levels of software platform. Time synchronisation between the local and remote devices is implemented by means of Fieldbus network with published-subscriber architecture. A client–server arrangement is used to deal with the data communication between Matlab and Labwindows. A temporary data buffer provides variable time accuracy while using a small fraction of system resources. This approach has been realised in an electro-hydraulic control system, demonstrating the full use of existing rich software resources and the convenience of configuring the systems.