The mission of this group is to bring together utility professionals in the power industry who are in the thick of the digital utility transformation. 


Combining AI and Video for Optimal Security

image credit: © Frui |
Julian Jackson's picture
Staff Writer, Energy Central, BrightGreen PR

Julian Jackson is a writer whose interests encompass business and technology, cryptocurrencies, energy and the environment, as well as photography and film. His portfolio is here:...

  • Member since 2020
  • 503 items added with 187,303 views
  • Dec 3, 2021

Within the cybersecurity sector of the utilities industry is the increasing awareness that companies will need to employ a wide variety of methods to keep the bad actors at bay. As infrastructure is big and potentially vulnerable, there will be a place for a traditional security watchman doing his or her rounds. However at the other end of the spectrum, we will need to use the most advanced technology – after all, the bad guys certainly will.

Protecting installations, pipelines, cables, vehicles, and communications systems is mission-critical. Some equipment, like substations or pipelines, for example, can be in remote locations. Others, a large power plant say, might be in the middle of an urban conurbation or beside a busy river. Those will have different monitoring and security issues. But both could come under physical attack. It might be from vandals, criminals, or even a terrorist group or malevolent state actor.

A physical and/or cyber-attack on any important location could damage operations, and perhaps even force a utility to shut down to fix its systems. That's why having real-time monitoring of installations using wide-spectrum cameras to capture thermal, heat and night-vision data is essential. Modern AI video analytics in real-time will ensure that all potential threats, as well as false alarms, can be categorized and remediated.

By using thermal, infrared and night-vision cameras for real-time monitoring, combined with AI interpretation and alerts, security can be optimized, and legitimate activities differentiated from a imminent threat. The company security team will be alerted immediately if there is a security breath, vandalism or theft in progress. Warnings can also be sent to administrators or other personnel depending on the level and progress of the incursion. If necessary the installation can be locked down, rendered temporarily inoperable, and backups activated. This will ensure that damage or power disruption is minimized.

A further development of these AI systems is the monitoring of the condition of equipment, as well as adverse weather effects, for example, flooding at a remote site. Combining AI and video analytics can improve preventative and condition-based maintenance. Seeing the condition of remote machinery and equipment in real-time can save operators and plant maintenance teams significant time and reduce costs.

By analyzing the raw video feeds from numerous cameras, then utilizing unstructured machine learning algorithms, organizations can start to build new mathematical models that assess risk, site stability and machinery reliability. In future it will be possible to predict when a given asset or machine will suffer a fault, using leading indicators machine learning models may have created. The outcome is likely to ensure that machinery has an extended life-span, which will lower costs – a valuable end in itself. Combining real-time data and analytics and machine learning models can help predict which types of machinery or equipment is most vulnerable to theft or vandalism. Having these predictive insights, companies can operate more powerful deterrence strategies to protect their assets.

Combining AI, effective multi-spectrum cameras, and delivering real-time monitoring with video analytics as well as a continual feed of data from physical monitoring of equipment can reduce false alarms and ensure the broadest security coverage possible. The investment of time and money into video analytics and AI is comparatively small, compared to the assets at risk and will help ensure that the worst-case scenarios are fully preventable.


No discussions yet. Start a discussion below.

Julian Jackson's picture
Thank Julian for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »