Come join us to tackle real-world challenges and create a sustainable future together. Let’s connect and work together to bring your ideas to life through our Research and Development Programs. IVA’s R&D Innovation Ecosystem provides a platform for visionaries and domain experts to connect and collaborate globally, leveraging people, resources, and tools to drive innovation. This is a space for those passionate about transforming the everyday and breaking down borders to accelerate progress worldwide. Get in touch and let’s tackle real-world problems together
Implementing a real-time pipeline for inferential sensors by collecting sensor data and processing it in real-time including modules such as failed sensor detection logic, noise detection, synchronization of modules, AI/ML module, and documentation and demo. These modules work together to ensure that accurate data is obtained, analyzed, and used to generate insights and recommendations in real-time. The documentation and demo provide a means of showcasing the pipeline and demonstrating its functionality
This system provides a centralized repository for storing engine-run data, enabling easy access and management of large volumes of strain sensor data. Process automation streamlines data processing, improving efficiency and accuracy. By consolidating data in a single location, the system simplifies data management and enables better data analysis. The solution can be customized to meet specific requirements and is designed to support a wide range of engines and data sources
IVA is engaging with DRDO (Defense Research and Development Organization) to remove noise from high-frequency strain data. Typically, noise removal algorithms are applied to remove unwanted signals from the data, but the standard algorithms that are presently being used are not effective for the specific strain sensors. Therefore, specialized algorithms are being developed for removing unwanted signals from high-frequency strain data, without compromising the accuracy of the data. This is likely to involve training models on large amounts of data to learn the unique characteristics of the strain sensors being used by DRDO and analyzing the characteristics of the strain sensors, such as their sensitivity, frequency response, and noise sources.
Failures in sensors of Turbo jet engines can have serious safety implications. By detecting failures in sensors automatically, potential safety hazards can be identified early and addressed before they become more serious. By addressing problems early, the likelihood of more significant failures that can lead to expensive repairs and downtime is reduced. The process of researching and developing an algorithm for the automatic detection of failed sensors requires a combination of data analysis, machine learning, and software development skills, to automatically detect sensor failures in real-time, based on the identified patterns and behaviours.
In trying to make predictions using machine learning, there are a variety of different models that can be used, such as linear regression, decision trees, neural networks, ensemble models and more. Each of these models has its own strengths and weaknesses and may perform better or worse depending on the specific task and data being analyzed.
IVA is developing digital twins of human organs, which can analyze healthcare practices, monitor disease and wellness, and integrate individual patient data with physiology and immunology to predict medical events. These 3D images are tailored to each patient and created using machine learning algorithms trained on a massive patient database. Initially, the IVA will create digital heart models to assess cardiovascular risks.