Reducing Emissions, Carbon Footprint with Edge Computing

Over the past year, we have seen an accelerated pace of low-carbon technology adoption across the energy value chain

In fact, 2019 was the highest year in terms of energy investment since 2013, with a total of $3.95 billion in investments and 447 total deals. So, what is driving the trend?

Carbon-heavy industries, including oil and gas, manufacturing, and transportation have traditionally been challenged to reduce emissions while maintaining productivity and profitability. However, due to increasing regulatory and industry pressure to reduce carbon emissions, as well as recent market volatility and its long-term implications, these industries have been put under a microscope. Now, new technology has become available which can substantially lower emissions while increasing factory efficiency.

Digitization, driven by the adoption of Industrial IoT, has resulted in billions of devices in countless factories and industrial environments generating unfathomable amounts of data. IoT solutions are considered diagnostic tools that enable operators to root cause issues, albeit this is still often reactive. As these solutions have evolved, industries are beginning to realize the value of edge computing, especially as it pertains to powering clean-tech applications. Typically, devices send the data to a data center for processing, increasing response times, and bandwidth usage. What if the bulk of computing became concentrated on the edge of the IoT network? How would edge computing impact CO2 emissions and decarbonization initiatives?

IoT and Edge Computing

In edge computing, the data from sensors and devices is processed at the edge, where data is being generated. The data never has to leave the network to provide insights (although you can certainly still send it to other endpoints for further analysis). This approach reduces latency and puts far less strain on network bandwidth, speeding up time to insights to react proactively and prevent downtime, lower CO2 emissions, and improve worker safety. Speed counts in environments where change can happen in seconds, and edge computing answers that need. This is a powerful realization that low latency response times can translate to paused production or halted assembly lines based on live data from the field. And so we can do more, to proactively address decarbonization, and we can – using edge computing powered by machine learning and AI.

Reducing Emissions

Specifically, oil and gas and industrial manufacturing organizations are feeling the pinch of regulations to reduce CO2 emissions. Decarbonization initiatives have been easier to accomplish in industries like commercial and residential buildings. But now, that focus is on industries that have a much harder time reducing emissions.

Oil and gas facilities burn hazardous gases to avoid releasing them directly into the atmosphere. These measures were meant to reduce the emission of greenhouse gases, but the Environmental Protection Agency cites some facilities are using flare burning to bypass proper pollution controls in non-emergency situations. This behavior puts the industry in a difficult place, especially since monitoring flare stacks is currently a dangerous and manual activity.

Edge computing, powered by machine learning, can provide visual and audio monitoring of flare stacks to circumvent illegal flares while eliminating the need for on-site personnel at the stack. Sensor fusion can correlate the flare state with compressor audio to give operators a clear picture of current events. Operators can then use this data to monitor flares and ensure they are only happening in emergency situations, increasing regulatory compliance, and decreasing the emission of greenhouse gases.

Another example where emission is a key challenge is in the steel industry. Steelmaking is one of the world’s most prolific sources of CO2 emissions, counting for 7%-9% of greenhouse gas emissions globally. In the manufacturing process, iron ore is heated to a temperature that releases CO2 directly into the environment. Production of one ton of steel can release up to two tons of CO2. Global steel production totaled 143.3 metric tons in February 2020, and of that, approximately 3% was scrap. That 3% represents 4.3 metric tons of scrap and 8.6 tons of CO2 in just one month.
Edge computing can help drastically reduce scrap by using real-time measurement data and machine learning models to determine the quality of the manufactured steel and ensure the product meets standards. Then, it signals about any quality issues before that steel becomes scrapped. This, in turn, results in optimal energy used for production and reduces wasteful emission.

Edge Computing for IoT is a Force for Sustainability

Global greenhouse gas emissions, of which a large portion is CO2 emissions, in the buildings and power markets increased by ~20% between 1990 and 2014, while emissions from industrial sectors increased by ~70%. Over the past few decades, technological breakthroughs played a significant role in forging efficient decarbonization practices for buildings and power sectors, says McKinsey & Company. Yet for industrials, these efforts (or lack thereof) have been less successful. However, due to demographic and resource fluctuations, like rapid urbanization and growing constraints on critical resources, such as copper and zinc, industrial players should reconsider existing strategies.

The manufacturing efficiencies introduced by edge computing can reduce emissions of greenhouse gases, even in industries where they are notoriously difficult to curb. The applications are endless and include fleet and transportation, as well as any industrial endeavor that is traditionally a source of CO2 emissions. In summary, stakeholders in the energy market should be paying close attention to edge computing and how it can be utilized moving forward.


About the Author

Ramya Ravichandar is Vice President of Products at FogHorn. FogHorn brings a rare combination of technical expertise in real time analytics, machine learning and AI, combined with valuable experience in Industrial IoT. With over a decade of experience in edge computing, she continues to be excited about disrupting traditional industries to build a more sustainable planet. Ravichandar has a Ph.D. in computer science from Virginia Polytechnic Institute and University.