Startup Projects
Energy Technology

IoT Enabled Anti-Theft Technology for Emerging Markets 

​

Project “ninakaw” is an anti-theft full-system project for medium voltage customers. The system uses leading-edge IoT technology to gathers all the elements of power use measurement through digital meters into a single device that connects to the provider through a remote management system using cellular communication networks.

​

The devices are intelligent modules with diverse functionalities, and once a day, they transmit accurate consumption information to the provider for efficient remote management of supplies, disruptions, and reactivations. Artificial Intelligence (AI) and Machine Learning (ML) models are used in the control center to identify unusual patterns relative to customer profiles located in similar areas. This data is also used to anticipate consumer behavior and predict which customers are likely to have informal connections to the power grid. This information can then be used to curb such connections and cut waste as well as identify potential theft.

Artificial Intelligence System for Prevention of Power Outages by Predicting Asset Failures

​

Power grids in the developing markets of the world are constantly underprepared for events that cause outages. These same power grids are vital to the economies of these developing countries and are used to deliver electricity from powerplants to factories, hospitals, offices, homes, schools, and other vital facilities. However, given the state of these power grids, they have many vulnerabilities. Natural disasters, faulty maintenance, human error, equipment failure, and sabotage are all common reasons for power outages in these developing countries.

​

Project “auxilium” utilizes Deep Learning (DL) systems to make data-based predictions about upcoming malfunctions of power grids in these developing countries. Data is collected from various industrial IoT sensors, automatically analyzed by Machine Learning (ML) and Deep Learning algorithms, and then used by operators to take immediate steps for outage and blackout prevention. Moreover, Project “auxilium” also helps with optimizing supply according to seasonal demand, which can significantly reduce maintenance costs.

​

Project “auxilium” is an Artificial Intelligent (AI) system to prevent power outages by predicting asset failures weeks in advance. By showing which assets are likely to fail within a specific timeframe, given the possibility of a certain event, Project”auxilium” lets utility companies plan repairs and safely correct issues before a power outage occurs. This allows a utility company to reduce customer outages, improve customer satisfaction, reduce overhead and maintenance costs, and improve public and employee safety. When unplanned outages occur, utility companies must make emergency repairs – install new equipment, dispatch crews and control traffic, and coordinate logistics – all with increased time and expense.

​

With Project “auxilium,” utility companies can predict and plan repairs, reduce customer service disruptions and give customers advanced notice about the planned improvements and outages. By providing advanced warning of disruption, utility companies can ensure energy reliability to mission-critical services such as hospitals, fire stations, and rescue services.

​

In real-time, Project “auxilium” collects and analyses high-frequency power quality data to identify the warning signs of an imminent failure of distribution system components. Project “auxilium” then applies Machine Learning techniques and algorithms to combine historical high-frequency sub-cycle data with data from operational systems, including distribution connectivity models, to identify the failures.

Optimized Geothermal Plant Operations Through Deep-Learning 

​

Project “influunt” enables the effective and efficient management of geothermal facilities through the use of purpose-built and off-the-shelf IoT-enabled devices that can be deployed at various points throughout the facility to facilitate sensing and data collection. Data from such points as the fluid path of the plant, steam water gathering systems, separators, turbines, condensers, etc. can be collected to improve the plant’s operations. The plant operation data collected from the various sources and IoT devices is encrypted and standardized for future use and predictions. 

The encryption and standardization process ensures the security, privacy, and usability of the data for future operations. The encrypted information is then stored in a Cloud repository. The data from the Cloud repository can be analyzed for the realistic modeling of the plant processes, accessible and timely updating, efficient problem solving, and continuous improvement using feedback mechanisms and cybernetics. Project “influunt” has several modules that worked together to optimize plant operations while at the same time managing risk, providing descriptive analytics, diagnostic analysis, predictive analytics, prescriptive analytics, and detective analytics.

The data collected from various IoT devices, cameras, and sensors are utilized to detect cracks and other potential defects in plant equipment. Artificial Intelligence (AI) and Machine Learning (ML) techniques are used for predictive maintenance to prevent future emergency situations and reduce dangers. Analyzing the historical, comprehensive, accurate, and live data from multiple sources helps in effective visualization and studying the various processes in the power plant. It also helps in future comparisons to analyze the rate of energy generation, cost analysis, appropriate pricing, and consumption statistics. 

This additionally helps to prescribe the best solutions and modes of operation based on the inputs. Machine Learning techniques like regression analysis, correlations, and data discovery are used to identify the reasons for the behavioral aspects of various processes and faults. In addition, causation and correlation are used for the near past data analysis that helps in predicting future behavior, biases, trends, etc. Big data analytics is used for the predictive diagnostic of anomalies for predicting any inappropriate behavior and processes ahead of time. This helps to lower the rate of problem occurrence at the plant, reduces the outages, monitor the generation costs, and detect potential issues and problems in the functional processes during the normal functioning of the power plant. 

The data patterns from plant processes are analyzed to prescribe control measures. Machine learning algorithms such as Latent Dirichlet Analysis (LDA), Singular Value Decomposition (SVD), and Principal Component Analysis (PCA) are used to automate the decisions and recommendations. In addition, detective analysis techniques utilizing Deep Learning models such as Recurrent Neural Networks (RNN), Generative Adversarial Networks (GAN), and Radial Basis Function Networks (RBFNs) are used to identify inappropriate values and behavioral patterns in the plant’s functioning so that such anomalies can be detected, eliminated, or rectified.

​

Grid Usage Modelling

​

Project "naustrdomi" is a Deep Learning (DL) system built on cybernetic models that help identify and predict grid usage patterns. IoT technology is used to gather data over time and build predictive analysis through the use of advanced mathematical models such as expectation-maximization algorithms utilizing an iterative refinement approach -  employing both k-means and Gaussian mixture modelling.