**4. Concept of a Numerical Model Consistent with the In-Situ Measurement Capabilities**

The discussed models [8,9] provide very good insight into the concrete hardening process, especially for engineering applications (e.g., SHM system), where very limited data is available. The temperature of concrete and heat released during hardening are the main measurement parameters, around which hypotheses regarding the course of phenomena, mathematical theory and numerical model can be built [28,29]. Relative humidity measurements can be assessed in the surface, however, the registration of changes in concrete moisture content at any depth of the element is complicated, for example due to the construction of this type of sensor. The detection of concrete strains and microcracks could be implemented using ultrasonic methods, more and more often used successfully, but requiring extremely precise evaluation. The implementation of such an extensive system that monitors thermal, moisture and mechanical parameters at the construction site conditions, combined with numerical analysis, becomes an extremely laborious task to rival with traditional methods of thermal and mechanical properties evaluation. However, this is not an obstacle to integrate the possibility of thermal changes registration in a structure with numerical verification and prediction of temperature distribution in the time and space domain.

The thermal aspect of early age concrete should focus on the integral parameters, used by engineers, during controlling the hardening process. These parameters should have a clear technical sense, be easily measurable in the monitoring system and have their equivalents in the mathematical model. The main thermal parameters are the:


In addition, it is necessary to control the wind speed (m/s), the thickness of the element (m), type of formwork and insulation.

In the function of the above parameters, simulations of the temperature field in the concrete bridge deck were performed. For this purpose, two thermo-chemical models presented in Cervera's [8] and Martinelli's [9] papers were used. Although the mentioned models take into account only thermal and chemical aspects, its implementation for real structures is a difficult task. The discussed models simplify the problem of heat transfer omitting the heat movement by migrating moisture and assuming that the heat transport only by conduction is carried out. According to many authors, the influence of moisture diffusion on heat movement in concrete is not significant [30,31].

Prediction of temperature field of CIP concrete requires knowledge of the temperature development of concrete in adiabatic conditions. Due to the difficulties in the availability of such commercial devices, a novel method for determination of adiabatic hydration curve was proposed. To identify this curve, the concrete temperature measurements inside 150 mm cubic samples should be taken. These specimens are curing under isothermal conditions, i.e., in a water bath at constant temperature and under semi-adiabatic conditions in styrofoam containers, especially prepared for this purpose. The suggested approach and equations described in Martinelli's paper [9] allow us to plot the adiabatic hydration curve and then to identify three material parameters: κ/*n*0, *n* and *A*0/κ, necessary in the mathematical model described by Cervera et al. [8]. Parameters obtained in laboratory conditions are only a reference point for the considered concrete, because they strongly depend on the geometry and atmospheric conditions prevailing at the construction site. Therefore, in the present work, for the analyzed concrete deck of the extradosed bridge, the own functions for κ/*n*<sup>0</sup> and *n* were defined., and

The initial condition of heat transfer phenomenon does not cause many difficulties, because the initial temperature of the mixture delivered on the building site is usually known. Definitely, more complications generate the determination of boundary conditions, which constantly change under field tests (ambient temperature, wind speed, curing activities and insolation). Mainly two possibilities of heat exchange between the concrete surface and the environment deal with analyzed box bridge girder. The surface of the plate can be free or protected by a shuttering layer. In the second variant, the heat transfer coefficient on the concrete surface protected by the formwork layer (α*<sup>s</sup> f orm*) depends on the difference between the initial temperature of the mixture (*T*0) and the ambient temperature (*Tenv*). In turn, for the convective heat transfer coefficient for the free surface (α*f c*), the own pattern as a function of wind speed was proposed. The presented concept for estimation thermal changes of the concrete elements of the extradosed bridge was applied.
