Wireless condition monitoring with smartmesh

One of the barriers that has prevented the adoption of conventional monitoring systems has been the complexity and cost associated with running wiring for signals and power around industrial plants.

Although point-to-point wireless systems do offer one potential solution, end users have been wary of deploying them, unconvinced that they would perform reliably enough in a factory environment.

Now, the Smartmesh wireless-communications technology developed by Hayward, California-based Dust Networks claims to have overcome the limitations of earlier systems while delivering the data quality and access of permanent, fully wired condition-monitoring equipment.
The communications protocol driving Dust Networks’ Smartmesh wireless system is called the Time Synchronized Mesh Networking Protocol (TSMP). It supports self-organising networks of wireless devices, called ’motes’, that stay synchronised to each other and communicate in timeslots in a similar way to TDM (Time-Division Multiplexing) systems.

Such communication allows the motes to be extremely low power, as they are only powered up when they are scheduled to communicate with one another. The protocol is also designed to operate reliably in a noisy environment, which might typically be found on a plant or a factory floor.

Every node in the network has the intelligence to discover and establish paths with its neighbours

A key feature of wireless-sensor networks built on the TSMP concept is that the wireless mesh automatically and continuously optimises its own performance. Every node in the network has the intelligence to discover its neighbours, measure radio-frequency (RF) signal strength, acquire synchronisation and channel-hopping information, and then establish paths and links with its neighbours.

To help design engineers build industrial-networking monitoring systems around its Mesh Network concept, Dust Networks has developed a set of off-the-shelf Smartmesh products.

These include a low-power IEEE 802.15.4-compliant wireless transceiver called the Mote-on-Chip, as well as a module based on it that allows developers to build remote nodes that can be interfaced to a variety of sensors. These sensors then take measurements from assets on the factory floor and the results are wirelessly transmitted over the Smartmesh network.

The company also offers an embedded network manager that can be designed into its customers’ gateways and controllers, where it can dynamically provide network optimisation and intelligent routing of data.

The Dust Networks approach now appears to be finding favour with some of the larger suppliers of industrial monitoring equipment and has recently been incorporated into GE Energy’s Bently Nevada Essential Insight mesh wireless machine-condition monitoring system. ’We chose Dust Networks’ Smartmesh solution because its low power enables us to deliver a reliable system powered solely by batteries, or via energy harvested from the machinery vibration, which dramatically cuts the cost of deployment. This allows our customers to deploy more intelligent monitoring precisely where it is needed,’ said Sean Coyle, product line manager, Bently Nevada Asset Condition Monitoring at GE Energy.

Dust Networks’ wireless-networking products are embedded in the three main components of GE’s Essential Insight mesh solution: in a wSIM node, a wireless-mesh network sensor that supports vibration and temperature analysis; in a Repeater, a mesh-only node that extends the coverage of the network; and in a Manager Gateway, a node that serves as the network manager and communicates with other networks in the process environment.

According to Coyle, in a typical system, the wSIM node would be located by the side of an asset in a plant that required monitoring, such as a pump or a motor. There it would acquire temperature or vibration data from a mix of transducers, after which it would process the signals before transmitting them onto the wireless-mesh network through the Dust Networks Mote-on-Chip - the wireless transceiver built into the wSIM module.
’Up to four vibration or temperature sensors can be linked up to each WSim node at the present time, but in the futurewe anticipate additional measurement types to support a variety of asset-condition monitoring applications,’ said Coyle.

“The low power of Smartmesh allows us to deliver a reliable system powered solely by batteries”

SEAN COYLE, GE ENERGY

All the wSIM network nodes are managed and monitored by the Dust Network Controller (DNC), an embedded system used to oversee network wireless functions such as security policies, intrusion prevention, RF management, load balancing and network optimisation. The DNC itself has been embedded by the GE engineers into the company’s Manager Gateway, which connects through a wired Ethernet connection directly to the plant network.

The GE Manager Gateway manages all the different wireless WSims, as well as the connectivity and reliability of the network. It also transmits the data that is coming onto the network to our System 1 optimisation and diagnostic software that is commonly used in industrial plants today to monitor a variety of assets, including the plant’s critical assets,’ Coyle said.

The third element in the system, the Repeater, is similar to the WSim node with regards to the DN2510 Mote-on-Chip wireless technology it uses, but it has no sensors connected to it, nor does it perform any signal-processing functions. It simply acts as another transport node within the wireless network. ’Once the system is up and running, intelligent wireless routing based on the dynamic conditions in the network allows a node to find a route to the other nodes within the mesh, providing resiliency against interference and helping ensure high network capacity. This makes sure that RF environments cannot affect network stability and peak network performance is maintained,’ said Ross Yu, senior product manager at Dust Networks.

As for the wireless range supported by the GE wireless system, Coyle said that distances between nodes of up 100m in good line-of-sight conditions can be achieved. However, he added that his customers are provided with more conservative guidelines that state interoperability at a 50m range to ensure that the network performs reliably.

The nature of the wireless mesh means that any issues relating to the failure of a particular node are accounted for. Any failure in a portion of the network due to equipment malfunction or damage is compensated for by the self-healing nature of the mesh network itself. Even if large portions of the network fail due to a catastrophic event, the remaining functional nodes continue to transmit monitoring data to the plant systems.

’Because the new technology addresses robustness and reliability, we are now seeing much more interest from our customers in deploying wireless,’ said Coyle, ’and the data acquired by the wireless systems can now be brought into the same condition-monitoring systems that are acquiring data by the more traditional wired methods from a plant’s critical assets.’