Monitoring of railway structures of the high speed line BPL with bituminous and granular sublayers

Ballast deterioration, under dynamic loads, remains an important issue on high-speed tracks that can lead to high maintenance costs. This ballast deterioration leads to settlements. Several studies have shown that these settlements were linked to high accelerations produced in the ballast by high-speed train (HST) passages. The solution with bituminous underlayment was used since 1980s in several countries like United States, Italy, Spain, especially on high-traffic and high-speed lines (HSL). In France, the interest in this technique is recent. Following the East European HSL satisfactory behavior, a layer of asphalt concrete was used under the ballast layer of the Bretagne-Pays de la Loire (BPL) HSL. It is intended, in addition to the schedule savings and the protection of the subgrade during the construction phase, to reduce acceleration amplitudes produced at the passage of HST, to ensure moisture stability in the subgrade and thereby to decrease the maintenance costs of the tracks. BPL HSL includes 105 km of innovative track with an asphalt concrete (GB) ballast sublayer, and 77 km with a granular layer under the ballast (UGM). Out of the instrumented sections of the BPL track, 3 sections are constructed with GB subballast layer and one with a layer of UGM as a subballast layer. A total of 127 sensors that includes accelerometers, anchored displacement sensors, temperature and humidity probes, and extensometers are used. Sensors are placed at various positions and depths in the track structures. Data were first acquired during a speed up test phase, under controlled conditions, with the same train passing at speeds ranging from 160 to 352 km/h. This paper presents the different sensors used for the instrumentation as well as the acquisition system installed to collect all measurements. Data treatment and processing is explained in details. Finally, results obtained for different speeds are presented, with a focus on accelerometer and anchored displacement sensor measurements, on two sections, allowing, among others, comparisons between the response of structures with and without asphalt concrete. The role of the GB, as a subballast layer, in damping the vertical displacement of the sub ballast structure and reducing the accelerations peaks in the ballast layer for ballasted tracks is demonstrated.