A new approach to measure the stress level in tunnel linings of the Italian highway networks

In the framework of ongoing design activities focused on the rehabilitation works of almost 600 highway tunnels of Autostrade per l'Italia's network, a smart analysis method which mixes historical calculation techniques with innovative standards has been developed. The key point of the method is the reliability of measurements regarding the actual stress level acting in existing linings, since it directly involves tunnels stability during the construction stages as well as tunnels' strength and resilience in the long term. The experimental measurement of the stress level in the tunnel linings typically leads to a set of uncertainties and issues: the need to define standardized survey procedures which can be carried out under objective conditions, significance level/surrounding conditions/validity of measurements and, finally, how to use measurements for design purposes. Following the statistical analysis related to a large database consisting of over 1,500 results of standard flat-jack tests, an experimental stage ranging from the depth of lining cutting and to the collection of strain measurements in the surrounding lining intrados by means of highly dynamic and sensitive sensors has been developed. This experimental stage has provided the main elements to assess the reliability of flat-jack test results and, in particular, to assess whether its actual representativeness is limited to the stress level acting on the lining intrados side or - and, in case, how - it can be extended to the whole lining thickness. The single flat-jack test consists of making a cut in a structural element and then applying a known pressure to the cut surfaces to restore the initial condition of the body. The analysis aims at standardizing the test procedure, i.e. defining the number and magnitude of the pressure increase with a flat-jack as a function of the initial and closing measurement of the cut made in the lining. In order to obtain feedback on the measured tension states, the simple single flat-jack test was integrated by the measurement of the deformation of the lining around the cut, both during the cutting phase and during the phases of increase and decrease of the jack pressure. Assuming that the furthest ends of the strain gauges from the cut are fixed and unaffected by the deformation of the lining induced by the saw, at each cutting step, the strain gauges ends which are closer to the affected area will extend towards the cut.