Investigating water pressure in the plumbing systems of Victorian houses

What we did and why

The BPC supported research undertaken by Deakin University to explore how pressure monitoring, data analytics and artificial intelligence can be used to improve the performance, safety and reliability of residential plumbing systems.

Residential plumbing pipe systems are largely hidden from view, making it difficult to identify emerging problems before they result in leaks, failures, property damage or costly repairs. This research examined whether high-frequency pressure monitoring and advanced analytics could provide new insights into plumbing pipe system behaviour, identify pressure-related risks, and support improved system design and maintenance.

The research involved installing pressure monitoring devices in nine occupied detached dwellings (Class 1 buildings). The devices continuously recorded pressure data from both cold and heated water services, capturing normal water use as well as transient pressure events, such as water hammer. The data was then analysed using event detection algorithms and machine learning techniques to identify patterns, classify events and assess system performance.

The research demonstrated the potential for pressure monitoring and intelligent data analytics to provide a practical, data-driven approach to monitoring residential plumbing pipe systems.

By supporting this research, the BPC is helping to build the evidence base for more resilient plumbing systems and identify opportunities to improve plumbing design, maintenance practices and regulatory frameworks.

The research was undertaken by Dr James Gong and Brendan Josey, and was supported by the BPC through a research grant awarded in 2024. The project was completed in 2026.

What we found

Pressure monitoring can identify normal and abnormal system behaviour

The research showed that high-frequency pressure data can be used to detect fixture (tap) and appliance operation and identify repeatable patterns of system behaviour. Machine learning techniques successfully grouped similar pressure events and demonstrated the potential for future anomaly detection, fault identification and condition monitoring.

These findings provide a strong proof of concept for data-driven monitoring of residential plumbing systems.

Excessive pressure was common in several homes

Some plumbing systems were regularly subjected to pressures exceeding the 500 kPa maximum pressure specified in AS/NZS 3500.

Excessive pressures were caused by a combination of:

• elevated incoming mains pressures
• pressure increases associated with heated water storage systems, and

• localised pressure build-up from trapped transient pressure events.

These findings indicate that pressure-related risks are not solely influenced by the water supply network and can also be generated or amplified within the plumbing system itself.

Water hammer events can be reliably detected

Transient pressure events, commonly known as water hammer, can be reliably identified using high-frequency pressure monitoring.

This capability creates opportunities for earlier identification of conditions that may contribute to component wear, leaks and system failures.

Common plumbing components may contribute to pressure risks

A significant finding was the identification of trapped transient pressures associated with some mini-stop taps containing check valves.

These components can prevent pressure from dissipating normally, allowing pressure build-up within connected braided flexible hoses.

Because braided flexible hoses are widely used in modern homes and are recognised as a vulnerable plumbing component, the researchers recommend further investigation into the prevalence and implications of this issue.

Performance varied across different plumbing systems

The research found that plumbing system characteristics influence the effectiveness of pressure-based monitoring.

Pressure signals were generally easier to distinguish in homes with metallic pipework. Systems containing plastic pipework tended to dampen pressure waves, making event classification more challenging.

This highlights the need for additional research before a universally applicable monitoring approach can be developed.

Report recommendations

The research identified several opportunities to improve residential plumbing performance, safety and regulatory outcomes.

Expand smart plumbing monitoring and research

Further monitoring across a larger and more diverse range of homes is recommended to strengthen the evidence base and improve the reliability of machine learning models.

Improve pressure management in residential systems

Greater attention should be given to managing pressure within residential plumbing systems, including:

• incoming mains pressure control
• pressure increases associated with heated water systems, and
• management of transient pressure events.

Investigate pressure-related component risks

Further testing is recommended to better understand the role of:

• heated water storage systems
• pressure control devices
• check-valve-equipped mini-stop taps, and
• braided flexible hoses.

The findings may inform future improvements to plumbing products, installation practices and regulatory requirements.

What difference this made

This research demonstrates how pressure monitoring, machine learning and artificial intelligence can be applied to residential plumbing systems to improve understanding of system performance and pressure-related risks.

The findings establish a foundation for future technologies that may support:

• real-time system monitoring
• early fault detection
• proactive maintenance
• improved consumer protection, and
• evidence-based policy and regulatory decision-making.

The project also identified potential risks associated with common plumbing system components, providing valuable evidence to inform industry guidance, standards development and regulatory policy.

Download the report

Read the full report: Smart Plumbing Systems for Households: Event characterisation and anomaly detection through the monitoring and analysis of transient pressure.

Related research

This research complements the BPC’s broader program of research focused on improving system performance, reducing consumer harm, and supporting innovation across the building and plumbing sectors.