| Popis: |
Due to the nature of the drilling process, there are several companies collecting data at the rig. Each company's data acquisition system applies its own time stamp to the data. Subsequent aggregation of data, for example in a data lake, relies on synchronized time stamps applied to the different data sources in order to collate the data. Unfortunately, synchronized time stamping is rarely true. This paper documents the different sources or errors in time stamping of data and provides some best practices to help mitigate some of these causes. There are many reasons for unsynchronized time stamping of data from different sources. It can be as simple as clock synchronization at the rig: each data providing or producing company has an independent clock. It can also be due to where the time stamp is applied: for example, at the data source or on data reception. Additionally, it can be due to how the time stamp is applied: at the start of the interval, the mid-point, or the end. Many of the protocols used at the well site have a high latency, mud pulse or electro-magnetic (EM) telemetry, or even WITS (Wellsite Information Transfer Standard), where the actual acquisition time may vary significantly from the time stamp. Perhaps finally, time stamping of derived data is always problematic given the unsynchronized nature of data sources. Synchronization of clocks within the data acquisition network is extremely important. The resolution of time synchronization depends on purpose: motion control for example demands high-resolution time keeping. However, for the purposes of local time stamping, synchronization to a Network Time Server with a resolution of one millisecond is sufficient. The issue is on agreeing on the common source, and agreeing on passage of the time signal through firewalls. Time stamping is a more involved matter, calling for agreement on standards and a degree of metadata transparency. The paper describes in some detail sender versus receiver time stamping, the downhole to surface time-stamp chain, and time stamping of derived data. Systems automation and interoperability at the rig site – allowing plug and play access to equipment and applications – rely on an agreed upon network synchronization scheme. Indeed, designing applications that must handle uncertain time adds considerable complexity and cost, not to mention the impact on reliability. This paper presents an ordered approach to a quite resolvable problem. |
