Artificial Intelligence in Infrastructure Inspection (by Parviz Soroushian)

 

Traditional bridge and road inspection methods are time-consuming and expensive, requiring a lot of coordination, such as traffic control, and they put personnel in danger. They involve labor-intensive techniques like visual inspection, thermal imaging, ground penetrating radar, and acoustic inspection, all of which often necessitate complex traffic lane-closure management, additional labor hours, expensive equipment, and can potentially place workers in unsafe environments.

The United States alone has close to 600,000 bridges dispersed across the roadway network. Federal regulations mandate that each bridge that has a length of 20 feet or greater must be inspected at least once every 24 months.

Drones offer a faster, safer, and more cost-effective approach. They can access areas that are difficult or dangerous for humans, such as under bridges or along train tracks. Studies indicate significant cost reductions. Drones can collect data from a variety of angles, aiding in identifying problems that would not be visible from a single vantage point. They capture significantly more thorough inspection data and provide high-definition photos from limited and inaccessible areas, such as beneath bridges and along beams and girders. Drone inspections can reduce the environmental impact of bridge and road inspections by reducing the need for traffic control and other measures that can disrupt traffic and pollute the air.

Drone inspection of infrastructure systems allows for data collection beyond capturing images using camera alone. Drones enable integration of high-resolution optical camera with infrared (IR) camera. The IR camera is crucial because it can provide more details to the interior structural damages of a bridge or a road surface than an optical camera, which is more suitable for inspecting damages on the surface of a bridge or a road.

Drones can be equipped with a minicomputer that runs Machine Learning algorithms enabling autonomous drone navigation, image capture of the bridge or road structure, and analysis of the images. When damage is detected, the location coordinates are saved. Machine learning models are used to detect defects from subscribed image topics. These features enable the drone to self-operate and carry out the inspection process using AI algorithms, reducing the need for human operation.

The machine learning capabilities that can be used for defect detection include deep learning models. Deep Convolutional Neural Network (DCNN) models can be employed for image classification. Two main types of models are utilized:

·       Classification Deep Neural Network: Can be used on the drone itself for quickly analyzing photos and classifying the photos into one of any number of groups, such as faulty roadways or okay roadways. These models perform a preliminary analysis.

·       Region-Based Convolutional Neural Network (RCNN): Can be used for ground station analysis due to higher processing power requirements. These models can localize, identify, classify, and bound objects inside the photo itself. They prioritize images flagged as faulty roadways by the classification models.

Sample images of potholes and manholes should be collected using both visible light and thermal cameras to train the neural networks. The images undergo feature extraction through multiple convolution and pooling layers.

The regulatory considerations relevant to AI-enabled drone inspection of infrastructure systems encompass drone licensing, airspace restrictions, and bridge inspection regulations.

Engineering Skills and Artificial Intelligence (by Parviz Soroushian)

 

This writing highlights a critical skills gap regarding the development and deployment of AI systems. AI is not solely about machine learning algorithms, but a complex interplay of software, sophisticated electronics, connectivity, machines, and infrastructure. AI can realize its full potential in Engineering applications if this broader AI systems engineering skill shortage is addressed. There is an urgent need for collaborative efforts between industry, government, and academia to create relevant educational and training programs, and for upskilling the current workforce to meet the demands of the AI era.

AI is a revolutionary technology with the potential for profound impact on the economy, society, and environment. One should make a critical distinction between focusing solely on machine learning algorithms and the broader AI ecosystem. AI systems combine algorithms with sophisticated electronics, pervasive connectivity, electro-mechanical modules and machines, and physical infrastructure. A complete view of AI should include an understanding of the end-to-end systems that integrate AI. There is a serious gap between the skills needed for the future and those currently available. While there is a focus on training and development within the broader AI skill sets, a gap has been noted in specifically training for AI systems.

The integration of AI requires a systems engineering mindset, with engineers prepared to work across different compute engines, understand algorithms, and have skills in hardware, software, and compilers. Engineers should also develop skills in digital signal processing (DSP), and other aspects of electronics. This can be accomplished through collaborative efforts between industry and academia. A closer relationship between industry and higher education can help with keeping the course contents current, and providing students with relevant experience. There are significant benefits to be realized from development of apprenticeships, especially at the degree level, as a way of bringing new talent into the AI field. Most companies are planning to up-skill and re-skill their current workforce to deal with this technological shift. Companies in the technology sector (semiconductors, electronics, sensors) are driving the evolution of AI systems. The future of AI will depend on these sectors, so focus should be placed on them.

In conclusion, the focus at the government level needs to be on AI systems and engineering skill sets, not just AI algorithms. It is crucial to adapt educational and training programs to meet the diverse skills required for AI systems, creating graduates that are ready for the industry. Greater collaboration between industry and academia is necessary to ensure training and education is relevant to current and future needs, for both those entering the workforce and those needing to up-skill.

Artificial Intelligence and Bridge Infrastructure Management (by Parviz Soroushian)

 

Bridges are essential components of civil infrastructure, requiring robust asset management to ensure safety and longevity. Traditional methods for monitoring bridge health are time-consuming, expensive, and prone to human error due to the large volume of data involved.

Artificial Intelligence (AI) offers efficient and accurate solutions for processing data from bridge monitoring systems, identifying patterns, and providing insights into structural health. Despite the potential of AI, there are still open research challenges that need to be addressed to fully utilize AI in bridge asset management.

Bridges are subject to damage from various factors including varying temperatures, road salt, heavy traffic, and abrasive forces. Structural health monitoring (SHM) systems are crucial for measuring bridge data, processing it, and assessing the health of the structure. AI algorithms can efficiently process large datasets collected by SHM systems, overcoming the limitations of manual processing. AI-powered systems can help identify patterns in data, predict potential problems, and optimize maintenance schedules, ultimately enhancing the lifecycle of bridges.

There are needs for further research in the field of AI-powered bridge asset management. This research could focus on areas such as:

·        Developing more robust and accurate AI algorithms for specific types of bridge damage.

·        Integrating AI systems with existing SHM infrastructure.

·        Addressing ethical and safety concerns related to AI-driven decision-making in bridge maintenance.

In conclusion, use of AI in bridge infrastructure management presents a promising solution to the challenges of traditional monitoring methods. By leveraging the power of AI, it is possible to create more efficient, accurate, and cost-effective systems for ensuring the safety and longevity of bridges. Continued research and development in this field will be crucial for unlocking the full potential of AI in maintaining critical infrastructure.

A Robust Process for Effective Carbon Capture (by Parviz Soroushian)

A complete replacement of fossil fuels by renewable sources of energy is not feasible in the short term. Therefore, there is a need to equip fossil fuel power plants with CO₂ capture and sequestration (CCS) capabilities in order to prevent the projected 2°C global warming by 2100. Carbon capture and sequestration would allow for continued use of fossil fuel until a deeper penetration of renewable energy sources into the grid is realized in an orderly fashion. Widespread adoption of CCS technologies would benefit from the availability of processes that lower the cost. Developments in advanced CCS materials and processes are needed in order to achieve this goal. One option is to develop a robust, economical and efficient CCS process based upon the principles of mechanochemistry.

The mechanochemical CCS process will take place in ambient condition within an energy-efficient mill incorporating a nearly dry inorganic sorbent. In the CO₂ sorption step (A), the input of mechanical energy will disturb the structure of the solid sorbent, and will drive CO₂ dissolution and deep diffusion of carbonate anions. This process offers favorable kinetics comparable to the dissolution of carbon dioxide in melts at magmatic temperatures and pressures. The disturbed structure of the sorbent and the dissolved nature of the captured CO₂ enable desorption within the same (mill) chamber at a moderately elevated temperature produced efficiently via microwave irradiation (B); rotation of the mill at a low speed benefits the uniformity of microwave exposure. The mild desorption conditions benefit the stability of the sorbent under repeated sorption-desorption cycles. The mechanochemical CCS process is robust, and can be tailored to accommodate different solid sorbents and the combustion emissions of different fossil fuels. Scale-up of the process benefits the mechanochemical effects by raising the intensity of energy input.