Where the Canary in the Coalmine Meets Aware-able Innovation

Our CapTech consultants always try to meet our clients where they are, and this has even meant in an oil field. Yet one particular connection went awry when loud alarms sounded from a gas detection unit (a required piece of safety equipment on well sites) in the middle of a conversation. Dangerous gas was present, and a safer location was needed. However safer locations are sometimes hard to find, and gas meters often continue their alarms without giving any direction of gas levels, gas locations, or where a safer place might be.

When the most recent CapTech Innovation Challenge presented a focus of “Aware-ables,” or using existing wearable technologies in new ways, our CapTech Denver team knew exactly what we wanted to do. We created a novel gas sensor that meets the needs of this client in the field and provides additional opportunities to improve the safety of any workers in potentially hazardous gas environments.

The Challenge

Three major issues exist with current gas-detection technology.

1. Gas detectors provide no assistance in determining proximity to dangerous concentrations of gas beyond being in an “active” or “inactive” state.

2. Field workers are not consistently utilizing safety equipment per regulation.

3. There are no records of dangerous conditions beyond anecdotal reports from field workers.

To meet these challenges, our team designed and built a prototype gas sensor with an abundance of new features that improve upon status quo devices like those employed by our client. Our new sensor not only detects gases, but also records the concentration of those gases, logs the location of the gas reading, and writes both metrics to an Azure Cosmos database. We also developed a reporting and visualization layer on top of our data storage solution to facilitate in-depth analysis of the newly collected data.

How Needs Were Met

Improved safety, compliance auditing, and reporting are all essential pieces of the gas-detection equation.

First, the safety of personnel on site is dramatically improved by our device. The buzzer on our gas sensor will increase in volume as the concentration of a dangerous gas increases and will decrease as the gas dissipates. This variable volume output helps clients avoid and exit dangerous locations more easily than current devices. Second, the device uploads and stores gas readings, which improves the ability of users to determine if field workers are both wearing their equipment and visiting all assigned well sites. Finally, reporting will improve time to remediation as the reporting layer will facilitate locating gas leaks and identifying the level of danger presented by each leak.

We utilized two pieces of software on the Raspberry Pi platform, and we used Python to collect and ship our data up to Azure. There were three classes we used to gather and ship data. Two were modified versions of code provided by the GPS and gas sensor manufacturers, and the third was a custom class that leverages Azure connection classes. Using these three classes, our device runs two scripts on boot, one gathering GPS data and the other logging detected gas levels. Both scripts collect a set amount of data points and then ship the data in a batch up to Azure Cosmos containers, where the two readings are joined on a time attribute.

Data in the Azure Cosmos containers can be accessed for reporting and analysis purposes, and the data created by the sensor allows for visualizations based on geography, time, and gas levels. This allows us to create granular heat maps that show where the device encountered gas levels out of a safe range. This visualization coupled with more devices would give operators passive maps of their well sites, allowing for potential leaks to be spotted and fixed before they became more severe. Additionally, we can show when gas spikes happen visually, which is a valuable tool when correlating different activities on well sites.

In the future, these visualizations would be able to detail different device interactions as well as more granular gas composition. We would also be able to build in safety controls that would monitor for noncompliance activities such as not wearing a gas sensor (leaving it in the car) as well as providing alerts in case devices became immobile during use.

Lessons Learned

Our prototype device has obvious room for improvement before becoming a full-fledged production model. For example, a GPS sensor that accesses more satellites simultaneously would provide more accurate and reliable location data; gas sensors used in the field must detect a minimum of four gases to provide safety to workers; a pump and fan system would ensure exposure to a constantly refreshed sample, which would greatly increase functionality and accuracy; and the case our device is housed in also needs improvement. However, the possibilities of a new innovation – a new canary in the coalmine for the digital age – can greatly improve safety conditions in a space where it is desperately needed.

CapTech consultants meet our clients almost anywhere, and this is one more instance of an unexpected connection with dramatic possibilities.