Remote monitoring services for thermal and renewable facilities

Particularly useful for generation asset owners who don’t want to hire control engineers, have limited access to them, or cannot justify the cost, based on the asset size.

One of our engineers can monitor multiple assets from multiple owners, achieving significant benefits of scale.

Services fall into three broad categories:

  • Monitoring for anomaly detection.

  • Monitoring for maintenance optimization.

  • Monitoring for performance improvement and cost reduction.

BKO Services will work with any client-side hardware or software, bringing you the benefits that come from monitoring by skilled engineers, analysis using Machine Learning, and reporting from data stored in Modern Real-Time databases, and Cloud technologies.

Monitoring for Anomaly detection

Our engineers perform anomaly detection using in-house trained predictive models to provide early warning alerts and diagnostic guidance to our customers.


Detecting anomalies of a machine in an industrial setting is an active problem that is being tackled by researchers and engineers till date. In a complex industry setting such as power plants, aviation etc., anomaly detection is critical to raise alarms beforehand to prevent significant damages and mishaps. Most of the active research in data-driven approaches for anomaly detection is based on using neural networks. However, most of the neural network based approaches deal with data coming from a single source. Here, we make an attempt to solve the challenge of including information from multiple data sources of a rotating equipment by proposing a novel approach of using multimodal neural network based autoencoder. These detected anomalies generally tend to be more accurate and robust than anomalies detected single data source as latent representations capture inter-dependencies between different parts of the system through those multiple data sources.

  • Moreover, to better facilitate Subject Matter Experts (SMEs) in annotating potential events picked up by anomaly detection models, we have built a web-based annotation tool.

  • Web application-based annotation tool developed to annotate events detected by anomaly detection algorithm.

A provision shall also be provided for selected users to annotate the event types that are detected through the anomaly detection models. Fig 1 is a sample schematic of the time-series annotation process. This allows the Client’s SME to troubleshoot and label events. A Fault detection model can then be trained to learn the event types when such events occur again. This interface provides SEEQ visualizations of actual vs predicted readings and the difference between the readings for the tags that differ the most. It also provides MHM vibration data for convenience.

Maintenance Optimization

AI based Optimization for Gas Turbine Maintenance Scheduling

Challenges to Establish Predictive Maintenance

  • Provide optimal windows for operational duration.

  • Provide novel operational insights for equipment.

  • Proactively identify potential failure modes and time to failure before planned maintenance windows.

  • Improve decision-making by increasing visibility into equipment health, including predictions of performance degradation and/or equipment failure.

Our Approach

Our approach is divided into 2 stages:

RUL Algorithm

Remaining Useful Life (RUL) algorithm monitors degradation for each fault type and outputs a damage variable that indicates the degradation over time / cycles of operation. Based on this damage variable, an estimate is given for number of cycles / hours of operation before a fault threshold is reached.

Optimal Maintenance Scheduling

The RUL estimations are then overlain with equipment run hour forecasting algorithms. This would properly place the maintenance window when outages are more likely needed and when the operational and financial consequences of such outages are minimal.

Why Predictive Maintenance?

  • Reduction in maintenance costs.

  • Increased service life of equipment.

  • Increased production and higher financial returns.

  • Reduction in machine failures.

  • Reduced downtime for repairs.

  • Substantially improve the overall operation of turbines.

At BKO Services, we combine the power of IIOT data with some of the most advanced machine learning algorithms to tackle these challenges and play a pivotal role in transforming your maintenance strategy from Run-to-Failure / Preventive to a much efficient and much more robust Predictive strategy.

Performance Improvement and Cost Reduction

Modern Power Plants are very efficient so are the de-regulated markets they operate in. With all that efficiency, complexity has come along for the ride. In a deregulated power market, most plants can offer their generation capabilities and their costs to a clearing house called Day Ahead Market, DAM.To address the modeling complexity, we created a power generation marketing and planning system for each plant centered around machine learning model representations of the plants.The end result was a robust process that improved the enterprise financial goals through optimized power plant dispatching and generation planning.


Our approach was to create a two- step solution.

  • Use Machine Learning and Thermodynamics Simulations to create a digital twin representation of each plant capabilities.

  • Build an interface for plant operators to forecast future generation capabilities based on local weather forecast.

  • Results are then shared with energy dispatchers to monetize those capabilities.