A multi-period mixed-integer-linear-programming problem for the design of mass networks is proposed in this paper. Design of mass networks, conversion systems and storages is done so as to minimize the capital expenses and operating costs and fulfill demand in resources. This work provides a detailed model of compressed gas storage, considering pressure state inside gas tank, compression costs and compressor design, which have not been considered before in a linear model, with a discrete pressure scale. The performance of the model is assessed on a hydrogen network design problem, including renewable electricity, electrolyzer design and hydrogen storage. The share of renewable electricity use increases from 60% to 90% with hydrogen storage. Different compressed gas storage linearized models are compared with different pressure scales and the linearization error is assessed. The introduction of discrete intermediate pressures reduces the committed error on compressor power and operating costs, due to the linearization.