Developing a Real-Time Dynamic Discrete-Time Bayesian Network-Based Model for Assessing Security Risks in Offshore Wind Farms

Description

Project Description (max 700 words):

This research project aims to develop a sophisticated and dynamic model based on Discrete-Time Bayesian Networks (DTBNs) to assess security risks in offshore wind farms (OWFs). The project will focus on incorporating real-time data, such as Vessel Monitoring Data and predicted weather data, to enhance the accuracy and responsiveness of the model. By considering the dynamic nature of security threats and the evolving conditions in OWFs, the model will provide timely risk assessments and enable proactive security measures. The project will address specific scenarios, such as the threat of attacks on offshore substation platforms by remote-controlled kamikaze drones launched from passing ships. Ultimately, this project aims to enhance the security and resilience of OWFs by leveraging advanced data analysis techniques and real-time risk assessment capabilities.

FreedomGRESS is a young and visionary international company dedicated to providing advanced solutions for the offshore wind energy (OWE) sector. Our primary goal is to optimize operations and enhance the security of critical infrastructure in offshore wind farms.  As a company, we draw inspiration from cutting-edge scientific research and aim to translate it into practical applications for the real sector. Our solutions are designed to address the complex challenges of OWE and contribute to the sustainable growth of the industry.

One of our core areas of expertise lies in quantitatively assessing security risks in offshore wind farms. By leveraging the power of artificial intelligence, we are developing advanced technological solutions that enable comprehensive risk analysis and assessment. Our clients should benefit from early threat detection, timely responses, and the prevention of deliberate incidents, safeguarding their critical assets such as offshore substation platforms or pipelines/cables.

The primary aim of this project is to develop a real-time dynamic Discrete-Time Bayesian Network (DTBN)-based model for assessing security risks in offshore wind farms (OWFs). The project aims to address the following key objectives:

  1. Enhanced Risk Assessment: The project seeks to advance the current methodologies for security risk assessment in OWFs by leveraging the capabilities of DTBNs. By integrating real-time data, including Vessel Monitoring Data (such as vessel location, speed, course, direction, type of vessel, affiliation, port of origin and port of destination), predicted weather data (including wave height), and additional information related to the vessel's cargo and type, the model will provide a comprehensive and accurate assessment of security risks.
  2. Timely Threat Detection: The project aims to develop algorithms and techniques to enable the early detection of potential security threats in OWFs. By analysing real-time data streams, including information from port authorities, CCTV footage, and communication with relevant authorities, the model will identify deviations from vessel trajectories, potential proximity of vessels for launching drones or using other means of attack, and potential detection of unfamiliar or unidentified vessels. This information, combined with socio-political statistics and other relevant data, will contribute to timely threat detection and proactive risk mitigation.
  3. Adaptive Risk Management: The project aims to create a flexible and adaptive risk management framework for OWFs. The DTBN-based model will be designed to adapt to new information and data sources, including encrypted and anonymized (incl. video) data, as well as open-source information that may impact the emergence of security incidents. This adaptive approach will enhance the resilience and security of OWFs by continuously monitoring the current situation and incorporating policy, socio-political, and statistical data for situation assessment and risk evaluation.

 

By the completion of this project, the following outcomes are expected:

  1. Real-Time Risk Assessment Model: The project will deliver a robust and efficient DTBN-based model capable of assessing security risks in real-time. The model will integrate multiple data sources, including Vessel Monitoring Data, weather forecasts, port information, cargo details, vessel type classification, and potential attack modes based on country classification. This comprehensive model will provide OWF operators and security personnel with a powerful tool for identifying and understanding potential threats.
  2. Early Threat Detection System: The project will contribute to the development of algorithms and techniques for early threat detection in OWFs. By analysing real-time data streams, including vessel trajectories, deviations, proximity to OWFs, and the use of advanced data analysis techniques, the system will enable proactive measures to be taken to prevent security incidents, ensuring the safety and integrity of OWF infrastructure.
  3. Adaptive Risk Management Framework: The project will establish an adaptive risk management framework that continuously evaluates and updates risk assessments based on real-time data, emerging information, and evolving circumstances. This framework will empower OWF operators to make informed decisions and implement effective risk mitigation strategies, utilizing encrypted and anonymized (incl. video) data, as well as open-source information that may impact the occurrence of dangerous events.

Overall, the project aims to significantly enhance the security posture of offshore wind farms through the development and implementation of an innovative real-time DTBN-based model. By leveraging advanced data analysis, machine learning, and predictive capabilities, the project will contribute to the sustainable growth and resilience of the offshore wind energy sector, enabling proactive risk management and ensuring the safe and efficient operation of OWFs.

 

The project will adopt a multidisciplinary approach, combining advanced data science, machine learning, and Bayesian network techniques to develop the real-time dynamic Discrete-Time Bayesian Network (DTBN)-based model for assessing security risks in offshore wind farms (OWFs). The following investigations, methods, and techniques will be employed:

  1. Data Integration and Pre-processing: The project will focus on integrating diverse data sources, including Vessel Monitoring Data, predicted weather data (including wave height), port information, cargo details, vessel type classification, country classification, communication data, and socio-political statistics. Robust pre-processing techniques will be applied to ensure data quality, consistency, and compatibility.
  2. DTBN Model Development: The project will involve the development of a DTBN model tailored to the security risk assessment needs of OWFs. The model will be designed to capture the dynamic nature of OWFs, adapt to new information, and consider the uncertainty and interdependencies among variables. Bayesian inference methods will be utilized to update the model in real-time and enable accurate risk assessment.
  3. Machine Learning and Anomaly Detection: Advanced machine learning algorithms will be employed to analyse real-time data streams and detect anomalies or deviations from expected patterns. Techniques such as anomaly detection, pattern recognition, and outlier analysis will be applied to identify potential security threats, abnormal vessel behaviour, and suspicious activities.
  4. Predictive Analytics and Decision Support: The project will explore predictive analytics techniques to forecast potential security risks in OWFs. By leveraging historical data and machine learning algorithms, the model will provide proactive risk alerts and support decision-making processes for effective risk mitigation strategies.
  5. Optimization and Resource Allocation: The research will address the optimization and allocation of resources in response to identified security risks. The model will consider factors such as response time, resource availability, and cost-effectiveness to optimize the deployment of security measures and allocate resources efficiently.
  6. Validation and Evaluation: The developed model will be validated and evaluated using real-world data from OWFs. Performance metrics, including accuracy, reliability, and computational efficiency, will be assessed to ensure the model's effectiveness in practical security risk assessment scenarios.

The research focus will be on harnessing the power of DTBNs, machine learning, and predictive analytics to enable real-time risk assessment, early threat detection, and adaptive risk management in offshore wind farms. The project aims to bridge the gap between theoretical research and practical implementation, ensuring the applicability and effectiveness of the proposed model in real-world OWF security contexts.

Availability

Open to UK applicants

Funding information

Funded studentship

This project is a funded Studentship for 4 years in total and will provide UK tuition fees and maintenance at the UKRI Doctoral Stipend rate (£18,622 per annum, 2023/24 rate).

You must enter the following information:

  • Admission Term: 2023-24
  • Application Type: Research Degree (MPhil/PhD/MD) – Full time
  • Programme of Study: Electrical Engineering and Electronics – Doctor in Philosophy (PhD)

The remainder of the guidance is found in the CDT application instructions on our website.

Visit the CDT website for further funding and eligibility information.

Contact the supervisors (named above) in the first instance or visit the CDT website for Director, Student Ambassador and Centre Manager details.

Supervisors