Edge computing is becoming more and more important as part of IIoT. The first line of computing power in the vicinity of an asset or process can respond intelligently or coordinatedly to events and help alleviate the burden of data processing on the cloud. In systems that may have tens, hundreds, or even thousands of sensors, most of the data from these sensors may have minimal value, reporting only "normal" operating conditions. The intelligent gateway can filter this data and discard it or repackage it more efficiently and transfer it to the cloud for storage and analysis. When an event of interest occurs, the intelligent node can quickly determine the correct response, issue appropriate instructions to the connected device, and aggregate the events into reports suitable for cloud consumption. Unlike highly localized responses to signals from a small number of sensors, edge calculations are associated with more coordinated actions, evaluating data from a large number of sensors to make decisions at a higher level. For example, if excessive vibration is detected in the rig, it can be a mining application. The standard response to the error signal received from the vibration sensor may be to stop drilling, resulting in lost production and unnecessary downtime to inspect and repair the equipment. With more computing power and more historical input and storage of sensors, a more powerful edge computing device can assess the impact on the overall system and identify several possible responses and calculate the results and take the best course of action or Notify the operator of the best choice.
While the direct sensor/alarm combination does not have a larger image perspective of the edge computing device with on-board data aggregation and processing, the edge processing engine can evaluate the data received from all connected sensors and is based on predetermined priorities. In the manufacturing industry, after a product test or inspection at the end of the pipeline, the failure rate suddenly increases and it may be necessary to stop production to investigate the cause. Smart edge devices connected to all machines can coordinate this response from all devices in the line. Alternatively, the cause of the change can be identified by analyzing the sensed data from multiple machines and automatically applying the fix, or instructing the operator to correct the problem so that production can be restarted quickly and efficiently.
In addition, edge calculations can aid in predictive maintenance by comparing sensed measurements with historical data or preset thresholds to aid in digital conversion to calculate the optimal replacement time. It also intelligently manages assets deployed in geographical areas where Internet access is unreliable or poorly covered. If the gateway device is temporarily unable to connect to the cloud, the data can be stored in local memory until the connection can be re-established. The device can then automatically sync with the cloud to ensure that remote applications always have access to the full and up-to-date information.
Efficiency management is another aspect. By using sensory data to adjust and optimize settings, according to high-level energy management policies, it can be enhanced by the extra intelligence of edge devices.