Energy & Utilities
Case Studies
Newsfor Energy & Utilities
A large european integrated electric power company is predicting, diagnosing and reducing equipment failures in conventional power plants with machine learning
A large european integrated electric power company implemented C3 IoT''s C3 Predictive Maintenance solution to achieve more accurate predictions of equipment failure and maintenance needs. The technology uses advanced machine learning-based algorithms to monitor instrument signals, track failure modes and detect anomalies in equipment components. The company''s 2,640 megawatt conventional coal-fired power plant benefited from the implementation at it improved prognostic lead time and flexibility in scheduling of maintenance tasks and reduced ununplanned, emergency maintenance tasks.
Linked use case: Predict problems and recommend proactive maintenance for power generation and supporting equipment
Australian Renewable Energy Agency improved accuracy of solar energy predictions by 31% using machine learning methods based on a distributed network
Australian Renewable Energy Agency investigates the ability of machine learning models to predict the output of photovoltaic (solar) energy at different timeframes for a network of sites. Three machine learning methods are investigated along with a baseline. Each performs differently depending on the time interval at which they provide predictions. However, basing predictions on a distributed network versus single site, as used in other methods, resulted in increased accuracy of 9-31% over existing methods.
Linked use case: Automate aircraft piloting
BP reduces methane emissions by 74% and increases production volume by 20% by optimising oilfield well valve functions with AI modelling
BP is trialling a modelling system which uses current and historic oilfield data to build a simulation and test the effects opening and closing valves at different production sites on gas emissions, which has resulted in decreased methane coming from the vents by 74%, increased production volume of 20% and overall costs decrease of 22%.
Linked use case: Optimise extraction plans based on data including drilling samples and performance at historic and comparable sites
Baltimore Gas and Electric generated $2.8 million in economic benefit from identifying fraud and unbilled energy usage with machine learning
Baltimore Gas and Electric Company (BGE) is leveraging machine learning to identify and tackle unbilled energy usage. In doing so, the company has generated $2.8m in economic benefit and is expecting its annual economic benefit to reach $20 million. The company is using C3 IOT''s solutions, such as C3 Revenue Protection™ and C3 AMI Operations™ to improve the operation of its advanced metering infrastructure (AMI) network.
Linked use case: Identify potential fraud from utility consumers
Bonanza Creek Energy avoids production shutdown and fines by anticipating when certain emissions are going to occur using a machine learning system
Bonanza Creek Energy has implemented a system that anticipates increasing risk of volatile organic compounds and emissions being released by monitoring real-time data from each of its production locations.
Linked use case: Predict and support mitigation of unplanned downtime
DeepMind increases value of wind power by 20% by predicting supply 36 hours in advance
Researchers from DeepMind and Google develop a neural network machine learning system to better predict availability of wind power 36 hours in the future. This is based on weather forecasts and historic turbine data, allowing for better grid scheduling of wind power supply a day in advance.
Linked use case: Optimise performance and positioning of wind turbines
EDF Energy is testing automatic recognition of the figures on meter readings achieving 79% accuracy
EDF has been testing AI for automatic character recognition. It aims at using the technology to read and process the figures on meter readings, an otherwise manual procedure. Until now, the company''s optical character recognition system has managed to recognise meters'' digits with a 79 percent accuracy. EDF also aims at using machine learning to extract important information and patterns of its customers energy consumption based on the collected usage data.
Linked use case: Identify potential fraud from utility consumers
EDF Energy matches physical components of nuclear power stations with digital instruction manuals using deep learning
EDF Energy''s team in France has developed deep learning models, which are capable of labeling components on the inside of power stations in order to digitally match the physical components to digital instruction manuals.
Linked use case: Automate image labelling of components to support complicated infrastructure management
EDF Energy wishes to monitor power station conditions in real time and predict maintenance requirements using machine learning
During the 2017 AI Summit in London, David Ferguson, head of digital innovation at EDF, unveiled that the company is planning on using AI for real time condition monitoring as well as for predictive maintenance to optimise operations and increase efficiency of its nuclear power stations.
Linked use case: Predict maintenance requirements
Enedis reduces high-tension electrical grid outage with predictive maintenance using supervised learning
French power supplier Enedis prevents outages throughout its network of more than 400,000 generators using DCbrain''s Deep Flow Engine. The technology integrates historical data of the topology and pre-mature ageing of assets with machine learning. As a result the company can use the technology to predict where maintenance is required and avoid a failure.
Linked use case: Predict maintenance requirements
Enel Green Power North America and Raptor Maps streamline solar facilities’ faults detection using machine learning
Enel Green Power North America has partnered with Raptor Maps to offer a solution for real-time identification of maintenance requirements for solar facilities. The solution is based on Raptor Maps’ machine learning post-inspection analysis system, Raptor Solar™, which will then be embedded into EGPNA’s drone hardware. The collaboration aims to bring optimise the detection of repairs needed and decrease the time needed to detect them from days to just hours.
Linked use case: Predict maintenance requirements
Enel improved the average energy recovered per non-technical loss inspection by 70% in Italy and more than 300% in Spain using machine learning
Enel leverages C3 IoT to identify electricity theft (non-technical loss) and recover unbilled energy. The solution applies machine learning and analytics to calculate the probability of fraud for each customer meter using data from seven Enel source systems. The company managed to improve the average energy recovered per inspection in both Italy and Spain.
Linked use case: Identify potential fraud from utility consumers
Enel is reducing operational and capital expenses by predicting maintenance and improving asset performance using machine learning
Enel implemented C3 IoT''s solution for predictive maintenance in order to avoid and reduce the occurrence of failures and faults. Using real-time network sensor data, smart meter data, asset maintenance records, and weather data the application uses machine learning an AI analytics to predict feeder failure. The company reports that it has managed to improve prediction performance.
Linked use case: Predict problems and recommend proactive maintenance for power generation and supporting equipment
ExxonMobil plans to automate hydrocarbon discovery by developing deepwater exploration robots in partnership with MIT
ExxonMobil is working with MIT to develop autonomous underwater robots which will be able to explore deep ocean floors detecting hydrocarbons for exploitation while minimising environmental damage.
Linked use case: Pilot and resupply drone independently
Gazprom Neft to optimise drilling and well completion with the use of artificial intelligence
Gazprom Neft has signed an agreement with Yandex on implementing AI in its oil and gas operations. They plan on leveraging the technology to support drilling and well completion, modelling oil-refining- and optimising other technological processes.
Linked use case: Predict problems and recommend proactive maintenance for mining, drilling and support equipment
General Electric has saved $80 million over the past few years by integrating supplier data across business units using machine learning
GE has leveraged machine learning technology from Tamr, to integrate supplier data and records across business units. The company''s goal is to identify products that are priced under different names from the same supplier in order to achieve better purchasing power and pricing. GE claims that the TAMR machine learning software has enabled GE to save $80 million over the past few years. According to Emily Galt of GE the use of Tamr has helped the company save $80 million over the past few years.
Linked use case: Capture 3rd party or internal data for price comparison and supplier relationship overview
Loka enhances HR responsiveness to employee questions and improves awareness of policies and benefits through internal chatbots
Jane is a specialized AI assisted chatbot developed by Loka to address employee queries about policies or benefits. It can also analyse employee sentiments and patterns which can be used to identify issues and take action to fix them.
Linked use case: Support employee HR questions through chatbot text functionality
MIT Department of Mechanical Engineering developed smart power outlets that distinguish dangerous electrical from benign spikes with 99.95% accuracy with machine learning
MIT’s Department of Mechanical Engineering have designed a ''smart power outlet'' to help identify the difference between harmless electrical arcs and dangerous ones for safety reasons. The supervised machine learning model was fed electrical current data, labelled as ''good'' or ''bad''. The system monitored the electric flows helping it learn to identify unique devices which posed a potential threat from their currents. The researchers claim that after sufficient training, their ''smart power outlet'' was able to identify dangerous electrical arcs with 99.95% accuracy.
Linked use case: Smart home devices manage consumer energy expenditure
National Grid is testing machine learning to automate assessment of electrical grid infrastructure using drones
National Grid has begun using machine learning to analyse aerial images taken by drones of its electrical infrastructure. The aim is to automate assessment of equipment status and damage, as well as access areas with less intrusion than by using larger aerial vehicles, such as helicopters.
Linked use case: Assess building damage from weather or natural disasters with drones and satellite delivered imagery
Orbital Insight identifies 4x more crude oil tanks in China than the official number by analysing satellite imagery
Orbital Insight has introduced its China Oil product, which estimates oil reserves in China based on oil storage tank floating roofs. This is done by using machine learning and computer vision to identify oil tankers from satellite images and then predict capacity.
Linked use case: Predict commodity price patterns based on satellite (or similar visual) data
Origin achieved 80% accuracy in identifying low production wells and $50M in savings using machine learning applications from C3 IoT
Origin has used the C3 IoT''s C3 Predictive Maintenance solution to optimise well placement and output and derive operational value from their real-time data. Furthermore data from a variety of enterprise silos were integrated into a comprehensive data scheme.
Linked use case: Predict maintenance requirements
Repsol plans partnership with Google Cloud to use machine learning to optimise crude oil refinery management
Repsol has announced a partnership with Google Cloud to leverage the latter''s machine learning software to optimise refinery management in Tarragona, Spain.
Linked use case: Optimise complex manufacturing process in real time eg determine where to dedicate resources to reduce bottlenecks and cycle time
Rice University researchers improve on the state-of-the-art for wind turbine icing detection with a CNN
Researchers from Rice University and other institutes develop an improved system for detecting icing on wind turbine blades which can affect performance and power generation. The CNN classification system, combined with anomaly detection, outperforming the state-of-the-art on real-world data.
Linked use case: Predict problems and recommend proactive maintenance for power generation and supporting equipment
SGN leverages ClickSoftware''s AI powered solution for mobile workforce optimisation
SGN announced in January 2018 that it would implement software from ClickSoftware to optimise its mobile workforce utilisation. ClickSoftware solutions for mobile workforce management use machine learning for predictive workforce intelligence.
Linked use case: Optimise routing of mobile staff
Saxnas Hydroplant increased profits by 35% after using machine vision to digitise a legacy pumping plant
Using digital cameras with machine vision technology meant that the Saxnas hydroplant was able to digitise the output from its analogue meters. This enabled a digital model that automated the unmanned plant''s configuration to optimise productivity and to respond to data on electricity prices. This drove a claimed 16% improvement in production and profitability of 35%.
Linked use case: Optimise energy scheduling and management of power plants based on energy pricing, weather and other real time data