Technology

TECHNOLOGY OVERVIEW

LYTT LYVE™ is our patented, cloud-based data storage, management and processing platform.

It combines industry-leading, real-time data streaming technology with our pattern recognition algorithms to intelligently extract and classify the events and features from sensor data that matter most to our customers.

We turn your data into tangible business insights within seconds, and send them anywhere in the world via a web-based, customised dashboard.

LYTT LYVE™

Cloud-based event detection and pattern recognition for real-time surveillance and optimization.

LYTT LYVE™ CAPABILITIES

Edge analytics

Streaming, storing, and processing terabytes of data used to be expensive and time consuming. Not anymore. Our edge computing solution intelligently extracts the key features from sources like distributed acoustic sensing (DAS) and distributed temperature sensing (DTS) data streams within seconds, reducing the volume of data by 1,000 times without compromising quality.

Data management

Our cloud-based data management tools make it quicker and easier to process large volumes of data, such as DAS and DTS. Our systems can also be integrated with other contextual data sets to show you the bigger picture. And since data is streamed straight to the cloud, you can access our tools and carry out real-time analysis from any location worldwide with an internet connection.

Continuous machine learning

This is where the magic happens! Machine learning is the process of teaching computers to use data to make accurate predictions. At LYTT, we use a unique combination of machine learning and physics models to continuously transform raw sensor and time series data into actionable insights in a matter of seconds.

spotLYTT™ visualization

spotLYTT™ is our user-friendly data visualization dashboard. It takes the data gathered by our patented, field-proven analytics tools and turns it into contextualized insights via a single, intuitive dashboard. It gives you the freedom to visualize and analyze relevant data in real-time for faster, more informed decision making.

Platform management

We know how precious your data is. That’s why our cloud-based platform comes with industry-leading security features, giving you the flexibility to grant access to multiple users anywhere in the world, safely.

HYBRID ANALYTICS

Our pattern recognition and visualization tools draw on machine learning analytics, similar to those used in music recognition applications, to identify and extract relevant features from terabytes of data.

Just like a song has an individual, unique pattern, different fluids– such as water, oil and gas – have unique acoustic fingerprints. But in complex, high-value businesses, it’s not enough to simply ‘see’ the fingerprint, operators need to know why they’re seeing it in order to make informed choices.

That’s where our tried-and-tested hybrid analytics approach comes in.

LYTT combines the power of acoustic pattern recognition technology used in the music industry with knowledge inherent in physics-driven models to deliver significant value to the oil and gas sector and beyond.

We use physics models to teach our models the impact a set of known variables will have on the data. This helps identify hidden data patterns created by fingerprints under different conditions. In other words, we don’t just ‘see’ the patterns, we understand ‘why’ we’re seeing them.

It is unlike anything else on the market today.

PAPERS

As thought leaders, LYTT has published numerous papers highlighting our subject matter expertise and demonstrating our innovative and proven technology in action.

Flow diagnostics in high rate gas condensate well using distributed fiber-optic sensing and its validation with conventional production log

This paper demonstrates an innovative approach to processing Distributed Fiber Optic Sensing (DFOS) deployed to a gas condensate well.

Paper number: SPE-205435-MS

Well integrity diagnostics using acoustic event classification on distributed acoustic sensing data

The paper describes the case of a critical injection North Sea well that developed sustained casing pressure in the B-annulus and required an investigation to understand the leak origin.

Paper Number: OTC-30930-MS

Real-time water injection monitoring with distributed fiber optics using physics-informed machine learning

In the case study outlined in this paper, an innovative cloud-based injection monitoring application is deployed that uses DAS and DTS and overlays physics-informed machine learning models to deliver real-time visibility.

Paper Number: OTC-30982-MS

Production optimization of sanding horizontal wells using a DAS sand monitoring system

In the case study outlined in this paper, the oil production rate was increased by 25% using a predictive pressure optimization workflow to minimize sand production in a horizontal, sand prone well.

Paper Number: SPWLA-2021-0001

Production optimization using a 24/7 distributed fiber optic DFO sensing based multiphase inflow profiling capability

This paper summarizes the main findings from the first successful deployment of a continuous multiphase inflow profiling application that uses innovative signal processing techniques and machine learning that leverage distributed fiber optic data to identify the phase and rate of the inflow along the wellbore during production

Paper Number: SPE-201543-MS

Using distributed fiber optic sensing to recover well integrity and restore production

The paper describes a case study of a well where persistent sustained casing pressure (SCP) in the A-annulus resulted in the well being shut in for a period of almost three years.

Paper Number: SPE-204450-MS

20 years of downhole surveillance history

Within a comprehensive overview of the evolution of downhole surveillance in the customer's field, DAS sand detection is discussed as a new, real-time downhole sand monitoring technology.

Paper Number: SPWLA-2018-V59N4A6

Application of distributed acoustic sensing DAS technology in identification and remediation of sand producing zones in OHGP completion

DAS sand detection surveillance in the customer's field delivers substantial production and safety benefits.

Paper Number: SPE-188991-MS

Downhole sand ingress detection using fibre-optic distributed acoustic sensors

A new DAS solution identifies real-time downhole sand ingress locations across a reservoir.

Paper Number: SPE-183329-MS

Well integrity flow detection using novel acoustic pattern recognition algorithms

This paper describes the application of fiber optic technology to identify and differentiate multiple downhole well integrity events such as tubing leaks, casing leaks, flow behind casing and overburden integrity.

Paper Number: SPE-203447-MS

Raising recovery rates in real-time using in-well fiber optic sensing

Learn how LYTT’s unique technology is turning in-well fiber optic sensing data into actionable insights, making well production safer, faster and more efficient.

SEE ALL PAPERS

FIBER OPTIC COMPANION TECHNOLOGIES

Distributed acoustic sensing

Distributed acoustic sensing (DAS) technology essentially turns a fiber optic cable into a long microphone, making it possible to track different sounds, such as oil, gas or sand moving in a well from the reservoir all the way to the surface.

However, DAS produces an extraordinary amount of data. For example, a single hydrocarbon well can generate 100 million data points every second. That’s like streaming 1,000 full movies every hour. Until now, it has been difficult for operators to convert this data into useful insights in a timely manner. But our data analytics solutions are changing all that.

Distributed temperature sensing

Distributed temperature sensing (DTS) uses fiber optic cable to continuously record temperature. DTS is particularly useful in certain operational environments where temperature can affect operational decisions.

DTS has proven challenging to operators, as it has traditionally taken them a long time to interpret. Our data analytics tools and solutions make light work of this, collecting, processing and analysing DTS in real time.