Pipelines with a total length of several million kilometers worldwide supply us with oil, gas and water. However, the seamless and continuous monitoring of this supply network is a challenge for operators: if pipelines are damaged, for example by construction work, agriculture and forestry or earth movements such as landslides, there is a risk to people and the environment.

A key safety aspect to prevent this is the so-called layer thickness. According to legal requirements, underground gas and oil pipelines must be covered with at least one meter of soil. The pipelines are usually flown over by helicopter every two to four weeks and inspected from the air. However, these visual inspections are expensive. In addition, the experts on board lack the "X-ray vision" they would need to recognize whether the ground cover is sufficiently high.

"We are currently developing a new, much more efficient inspection method for systematic and automated layer thickness measurement from the air - with the help of a drone," explains Claudia Windisch from Siemens Corporate Technology. "To do this, we are combining mobile, copter-borne sensor technology and intelligent image data analysis."

Equipped with a color and a near-infrared (NIR) camera, the drone flies along an underground gas pipeline along a predefined route and photographs the surface of the earth. In future, the images will be transmitted directly to the Siemens pipeline analysis platform via a web interface. The data is used to create a three-dimensional surface model of the pipeline route, which is georeferenced. This means that spatial information is assigned to the data.

"We achieve an accuracy of around ten centimetres when measuring the layer thickness. Our method therefore delivers very high-quality results," says Windisch. According to the expert, the Siemens system enables objective and repeatable analysis.

The flood of information...

...is reduced to relevant events thanks to the automated analysis of image and process data. The analyzed data is then made available to the pipeline operator either via a web interface or directly in a SCADA system.

Pipeline operators lose over 10 million euros per year due to leaks, damage or illegal tapping. If a leak occurs, it can often take days for the maintenance team to find the cause in areas that are difficult to access. "Automated 3D image analysis, as we are developing it, can help our customers to monitor pipes better and more cost-effectively. If the system detects deviations at an early stage, operators can initiate countermeasures more quickly to avoid potential damage or even failures," emphasizes Mike Liepe, Head of the Siemens Oil and Gas Pipeline Solutions business unit.

How it works in practice

The drones are stationed along the pipelines at the bottom valve stations, which are 30 to 50 km apart. If the process control system reports a deviation via permanently installed sensors, the nearest drone flies autonomously along the section in question. The recorded data is reported back to the control system, which analyzes the event and suggests countermeasures.

"We are supplementing the existing SCADA system with automated analysis of image and process data. Anomalies and defects are displayed in the real 3D environment. This additional information and the visual processing of the data in SCADA make it easier to identify faults more quickly and initiate measures," explains software expert Daniel Schall from Corporate Technology.

It will take another year or so for the researchers to complete their work. Further test flights are planned in Germany and Austria, during which different sensor systems and various missiles, such as light aircraft, will be tested.

Threats for building and overhead line monitoring

The technology should then not only be used for monitoring oil and gas pipelines, but also for other industrial applications, for example in buildings. Large office or industrial buildings have a high demand for heating energy. It should also remain in the building and not be lost through poorly insulated areas of the façade or roof. With the help of drones, Siemens researchers can fly close to buildings and inspect roofs and façades in detail.

The cameras on the small aircraft take infrared images of the heat emitted by the buildings and combine these with photos in the visible range of the spectrum. A three-dimensional model is calculated from this data: Hot spots, i.e. areas where heat is being lost, can thus be localized from the air. The advantage for the operators of an office or industrial complex is obvious: the heat leaks can be sealed, the environment is relieved and costs are saved.

Another possible area of application for drones is overhead lines. These are not only available for trains, but also for trucks that are operated electrically in an environmentally friendly way - such as the eHighway from Siemens. Here, drones have an optimal view of the lines from above. The copters fly over the lines and take photos at regular intervals, which are later analyzed on the computer. Whether foreign objects on the wires, sagging cables or defective pylons: thanks to pre-programmed parameters, the software recognizes exactly if and when damage is imminent. In this way, failures can be prevented in advance.