top of page
Writer's pictureTheCyberDiplomat LLC

The use of Artificial Intelligence in SCADA Security - A case of prioritising Cyberdiplomacy


Engineers working on an AI enabled SCADA System
Engineers working on an AI enabled SCADA System

The evolution of SCADA systems has helped plants minimize errors that can lead to malfunctions or regulatory permit violations, improving efficiency and helping reduce costs for energy and supplies. The systems are integrated with real-time data that helps make operators do their jobs more efficiently and safely. Operators are accustomed to using these tools and reacting to the presented data, avoiding problems, saving costs, and delivering better performance.


But new advances in AI are reshaping the future of water and wastewater operations in the critical infrastructure sector. 


There are many differences between traditional SCADA systems and SCADA systems with AI and ML integration. With AI technology, digital monitoring and data input help anticipate problems before they occur. The technology can also provide advice or guidance on solving a problem before it becomes more serious or operates more efficiently. For example, most operations have control loops where limited operational adjustments are made from a few data points. In the water treatment plant, chlorine feed where flow and residual measurements are used to adjust feed to maintain a set residual.


SCADA Security


Imagine extending this single-point idea across key plant processes like breakpoint chlorination and nitrification, nutrient removal at a wastewater plant, changing the biological treatment time due to an approaching winter season, or all of these. With AI technology, SCADA data can become real-time operating advice across these complex operations or take action without operator intervention.


Beyond operations, maintenance issues are anticipated before they occur, and corrective measures are identified. In many cases, the sensors can also automatically make adjustments in the plant's existing instrumentation to correct any issues they have detected.


This advanced predictive maintenance type sets AI apart from traditional SCADA systems.


One example of this is happening right now in Tacoma, Washington, where the city is actively seeking to achieve several objectives with AI, including:

  • Detect real-time changes in effluent flow and propose adjustments to improve water quality

  • Propose operational adjustments to reduce operating costs (i.e., reductions in energy or chemical usage)

  • Ensure the city's plants remain in permit compliance

  • Integrate AI with existing computerized control systems

 

Smaller-scale AI pilot programs have already been implemented at other locations in the U.S., including the wastewater treatment plant in Pigeon Forge, TN, where operators recently installed strategically-placed AI sensors to gauge dissolved oxygen levels. The sensors, capable of monitoring chemical levels and adjusting as needed, have helped reduce energy costs at the plant by as much as 20 percent.


Advanced countries using AI/ML in SCADA operations 


U.S. municipalities looking to incorporate AI into their water treatment systems have been encouraged by examples where advanced AI systems have been used successfully in similar plants. In Cuxhaven, Germany, sophisticated AI sensors were installed in the city's wastewater treatment plant in 2017 to help reduce energy costs associated with aeration. This has led to a 30 percent reduction in energy usage at the plant, leading to cost-savings for the municipality and contributing to the country's climate goals.


In Denmark, where water and wastewater treatment accounts for two percent of the country's electrical consumption (the same as in the U.S.), AI has been incorporated into several plants, combining data gathered through multiple sensors with statistical analysis and process knowledge, which enables the system to adapt to changing conditions automatically. This effort is expected to contribute to greater plant performance and reduced errors.


Successful applications of AI begin with defining the objectives. Are you seeking better performance around a single unit process, like using oxygen at an aeration basin or disinfection residuals, or a larger scale use? Are you seeking only guidance from the operator or direction and control?


From there, integrating AI into existing water and wastewater treatment plants requires a careful analysis of each plant's components to identify the optimal way to install sensor probes, which provide real-time data of how the plant is functioning and then make corresponding adjustments while alerting operators to potential problems.



Engineers working on an AI enabled SCADA System on a Water Treatment Plant
Engineers working on an AI enabled SCADA System Water Treatment Plant

These are steps that an experienced water and wastewater operations partner can assist municipalities with.


For municipalities considering an AI solution:

  • Updating existing plant instrumentation and updating existing digital platforms, such as SCADA computer data systems, to align with AI technology

  • Installing automated probes in key components of plant equipment so they are strategically located to ensure accurate readings

  • Making sure probes are routinely cleaned, inspected, calibrated, and replaced as necessary

Operators must also be cautious about leaving their updated systems vulnerable to cybersecurity attacks, ensuring they operate independently from open Internet platforms.


Because AI offers enormous potential for cost-savings and enhanced environmental sustainability, U.S. municipalities are already seeking to invest in these new AI technologies.


Micro-analysis of the technical advantage 


AI and ML for data analysis

One of the main benefits of AI and ML for HMI and SCADA systems is that they can help analyze large and complex data sets from multiple sources and sensors and extract valuable insights and patterns. For example, AI and ML can help detect anomalies, faults, and deviations in the process variables and alert the operators or trigger corrective actions. AI and ML can also help optimize the process parameters, such as temperature, pressure, flow, and speed, and improve product quality, energy efficiency, and safety.


AI and ML for cybersecurity

Another benefit of AI and ML for HMI and SCADA systems is that they can help enhance their cybersecurity and resilience against cyberattacks, such as malware, ransomware, denial-of-service, and data breaches. For example, AI and ML can help monitor network traffic and system behavior and identify and block suspicious or malicious activities. AI and ML can also help encrypt and protect the data and the communication channels and prevent unauthorized access or tampering.


AI and ML for maintenance

AI and ML for HMI and SCADA systems can help reduce maintenance costs and downtime and increase the system's availability and reliability. For example, AI and ML can help implement predictive maintenance, which uses data and models to forecast the condition and performance of the equipment and schedule the optimal time and frequency for maintenance. AI and ML can also help diagnose and troubleshoot system issues and provide recommendations or solutions.


AI and ML for usability

A fourth benefit of AI and ML for HMI and SCADA systems is that they can help improve the operators' and managers' usability and user experience. For example, AI and ML can help design and customize the HMI interface and provide intuitive and interactive visualizations, dashboards, and reports. AI and ML can also help enable voice and gesture control, natural language processing, and chatbots and provide feedback and guidance to the users.


Threats and Conclusion


Despite the benefits of AI and ML for HMI and SCADA systems, some challenges and limitations need to be addressed. For example, AI and ML require high-quality and relevant data, which may not always be available or accessible. AI and ML also require computational resources, such as processing power, storage, and bandwidth, which may only sometimes be sufficient or compatible. AI and ML also raise ethical and legal issues, such as privacy, accountability, transparency, and trust, which must be considered and regulated. Cyberdiplomacy needs to be prioritized so as to enable information sharing and implemnettation of regulations and standards.


We at CyberDiplomat Work on these areas of research and development. We work towards a future for cybersecurity for these SCADA systems through partnerships and practice. If you would like to work with us or collaborate in our initiative please get in touch by emailing us at info@thecyberdiplomat.com



2 views0 comments

Comments


bottom of page