New Deep Learning IDS SPARK & SAD Boost SCADA Cybersecurity for Solar Plants

Researchers have developed two new deep learning-based intrusion detection systems designed to improve cybersecurity for SCADA networks, according to pv magazine . These systems are particularly relevant for large-scale solar power plants, where SCADA oversees energy generation, monitors panel performance, optimizes output, identifies faults, and maintains operations.

The research team noted that current cybersecurity frameworks are often inadequate for SCADA systems, as they struggle with the complexity and evolving nature of modern cyber threats. Many existing methods rely on signature-based detection, which requires prior knowledge of attack patterns and fails to identify novel intrusions or zero-day exploits. Deep learning methods were considered to address this, as they can process large data volumes, identify complex patterns, and enable more proactive threat detection.

The new approach integrates two systems named the Spike Encoding Adaptive Regulation Kernel and the Scented Alpine Descent algorithm. By combining their strengths, the method is reported to improve spike-threshold accuracy while enhancing adaptability and robustness under dynamic conditions. The SPARK model uses adaptive spike encoding, dynamically adjusting thresholds based on input signal characteristics with advanced statistical methods. This improves sensitivity to changes and integrates temporal and spatial features for better information encoding, providing context-aware thresholding unlike traditional fixed methods.

The SAD algorithm complements SPARK with an optimization strategy inspired by olfactory and Levy flight behavior. This is said to enable efficient exploration of solution spaces, avoid local minima, and ensure optimal threshold selection. The hybrid approach can dynamically adjust and optimize spike thresholds simultaneously. Scientists indicated the SPARK model is well-suited for SCADA and IoT systems due to its scalability, real-time adaptability, efficient data handling, and lightweight design that reduces computational overhead and false positives. The SAD algorithm focuses on improving detection accuracy while maintaining computational efficiency, and its anomaly scoring mechanism can add another layer of detection running parallel to SPARK for a more fine-grained analysis of attack patterns.

Evaluation involved multiple benchmark datasets capturing diverse industrial environments, attack types, and operational conditions. These datasets included time-series sensor data, actuator commands, and labeled attack scenarios such as denial-of-service, distributed denial-of-service, malware, and injection attacks. Standardized procedures and visualization tools were used to enable comparison and provide insights into model behavior.

The SPARK model was found to consistently demonstrate superior performance, achieving high accuracy, precision, and recall across datasets. It outperformed traditional machine learning and deep learning approaches in detecting diverse intrusion types. The researchers stated that the SPARK and SAD models represent an advancement in modern intrusion detection, designed to provide improved detection capabilities and operational efficiency for more resilient security solutions in industrial control and IoT networks.

The novel intrusion detection systems were presented in a paper published in the International Journal of Critical Infrastructure Protection . The research team was composed of academics from Leeds Beckett University in the United Kingdom and King Abdulaziz University in Saudi Arabia.

Making Data-Driven Decisions to Grow Your Business

A Quick Overview of Market Performance

Understanding the Current State of The Market and its Prospects

Finding New Products to Diversify Your Business

Choosing the Best Countries to Establish Your Sustainable Supply Chain

Choosing the Best Countries to Boost Your Export

The Latest Trends and Insights into The Industry

The Largest Import Supplying Countries

The Largest Destinations for Exports

The Largest Producers on The Market and Their Profiles

The Largest Markets And Their Profiles

Instant access. No credit card needed.

Online access to 2M+ reports, dashboards, and tables. Trusted by Fortune 500 teams.