Asset Optimisation – Going Beyond the Basics

Once the basics are done well, as covered in the previous blog, the next step is to move towards advanced optimisation and ensure we can achieve every conceivable marginal gain. This requires integrating asset telemetry data with external factors such as weather data and building thermal properties. By leveraging these additional data points, businesses can implement dynamic control strategies that enhance asset efficiency beyond static changes.

In this post, we explore how data analytics can push the boundaries of asset optimisation, enabling adaptive asset management that responds to real-time conditions.

The Need for Advanced Optimisation

Traditional optimisation methods typically don’t account for external influences that affect asset performance. By interpreting multiple data streams, businesses can:

• Improve energy efficiency by dynamically adjusting settings based on real-time conditions.

• Enhance equipment life by ensuring equipment only works as hard as needed to maintain functionality.

• Increase resilience by automating the asset’s response to changing demands.

• Reduce costs by unlocking further marginal gains and responding flexibly to demand-side response events.

Leveraging External Factors

1. Weather Data

Weather conditions significantly impact asset performance, particularly for refrigeration and HVAC systems. By incorporating both real-time and forecasted weather data, we can dynamically adjust setpoints to reduce unnecessary energy consumption.

Examples:

• If a cool summer day is forecasted, HVAC systems can reduce cooling loads automatically—preventing overuse and enabling the use of free cooling.

• Refrigeration systems can adjust based on ambient temperatures to maintain efficiency while ensuring product safety—for example, through discharge pressure optimisation.

2. Building Thermal Properties

Every building has unique thermal characteristics influencing how it gains, retains, or loses heat. These properties can be inferred by analysing both temperature and energy data. Understanding them allows for more intelligent asset control.

Examples:

• Buildings with high thermal mass retain heat longer. HVAC systems can pre-cool or pre-heat spaces before peak electricity tariffs apply, reducing usage during expensive periods.

• For buildings with poor thermal performance, frequent and dynamic adjustments help maintain stable internal conditions efficiently.

Advanced Dynamic Control Strategies

1. AI-Driven Predictive Control

Predictive analytics and machine learning enable proactive adjustments to asset performance, rather than reactive responses.

Examples:

• Machine learning models can analyse historical performance and temperature trends to automatically determine optimal HVAC settings for the upcoming week, applying them autonomously.

• As discussed in previous blogs, predictive maintenance reduces downtime and eliminates energy waste by ensuring assets operate optimally and flag issues early.

2. Self-Learning Asset Control

Just as self-driving cars are trained on risk and reward, assets can be self-trained using reinforcement learning. These systems refine asset performance over time without manual intervention.

Examples:

• A self-learning HVAC system can adapt to user comfort preferences, outdoor conditions, and occupancy trends, balancing energy efficiency with occupant comfort.

• A self-learning refrigeration system can analyse usage patterns, ambient conditions, and historical data to dynamically adjust all components—optimising system-wide efficiency while safeguarding food safety.

The Future of Asset Optimisation

By moving beyond the basics and embracing data-driven dynamic control, businesses can unlock greater levels of efficiency and cost savings. Integration with external data, predictive analytics, AI/ML, and automation doesn’t just optimise energy usage—it enhances resilience to both demand-side and environmental challenges.

The future of asset optimisation lies in intelligent, adaptive systems that respond in real-time. These ensure that efficiency gains are not only achieved once, but are sustained over the long term.

By leveraging these advanced strategies, we can truly push the boundaries of asset optimisation.