[en] One of the main objectives of the PLANCK mission is to perform a full-sky cluster survey based on the Sunyaev-Zel'dovich (SZ) effect, which leads to the question of how such a survey would be affected by cosmological models with a different history of structure formation than LCDM. To answer this question, I developed a fast semi-analytic approach for simulating full-sky maps of the Compton-y parameter, ready to be fed into a realistic simulation pipeline. I also implemented a filter and detection pipeline based on spherical multi-frequency matched filters, that was used to study the expected SZ cluster sample of PLANCK. It turned out that realistic samples will comprise 1000 clusters at low rate of contamination, significantly lower than originally anticipated. Driven by wrong estimates of the impact of early dark energy models on structure formation, we studied the spherical collapse model in dark energy model, finding that models with varying equation-of-state have a negligible impact on the structure formation. Yet, the different expansion history for the different models can be detected via volume effects, when counting objects in a known volume. Furthermore, it turned out that the different expansion history strongly affects the angular SZ power spectra for the various models, making them an interesting tool to distinguish and constrain alternative cosmologies. (orig.)