**1. Introduction**

A sudden valve closure has always created complex conditions capable of causing major problems and damages to the water conveyance systems. So far, many methods have been proposed and used to overcome these problems, e.g., surge tanks, air vessels and different types of valves. Of those methods listed, air vessels have been used more frequently with better application because of their economic and the stability advantages. An air vessel is capable of controlling both negative and positive pressures and usually, the outflow action from an air vessel to the pipe has less head loss than the inflow into the air vessel to control the negative pressure [1]. Most of the previous and recent studies in air vessel sizing have investigated the entrance connection situation effect. Plenty of charts and graphs are presented by these studies for the air vessel sizing [2–6]. However, most of them are rarely used for sizing purpose because of the simplification assumptions made for their preparation. Instead, a trial and error approach for checking the minimum and maximum pressure in the pipe is used. By means of throttling the entrance in the air vessel, smaller air vessel sizes can be achieved. Based on this concept, Martino and Fontana [5] proposed a method for sizing air vessels based on optimizing the throttling. Nevertheless, the confined air in a hydraulic system can amplify the pressure peaks from a transient action and make the transient situation more complex due to the compressibility of the air. Hence, there is actually a grea<sup>t</sup> necessity for studies about air pocket (AP) behavior during a fast-transient occurrence that can provide grea<sup>t</sup> knowledge in the primary sizing of air vessels and more reliable final decisions in water pipe systems' behavior and safety. Hatcher and Vasconcelos [7] studied the behavior of an air pocket inside a pipe using an experimental approach, giving useful information to fill the gap between numerical models and real situations. They declared that few experimental studies focused on the effect of parameters related to the geometry of the system as well as air and flow conditions.

Several researchers have worked on the behavior of air in the rapid filling subject or entrapped air bubbles along a pipeline [8–10]. There are some studies that investigate the thermodynamic prediction and behavior of an AP under slow-transient which consider the heat transfer process that is negligible in fast-transient cases [1]. Usually, it is assumed that fast-transient thermodynamic behavior obeys the polytropic equation, while in experimental works, it was shown that this equation is not obeyed for a special range of AP sizes and Reynolds number (Re) [11]. For that reason, more analysis of AP behavior, specifically for small APs, seems to be required.

This present work studies the behavior of an AP in a sophisticated experimental apparatus for different major governing variables. In this study, the outflow way at the air vessel entrance is closed using a check valve (CV) to store the surge pressure from a fast-transient action induced by a water hammer (WH) event. It helps to study the compression phase of an AP as in real conditions. Furthermore, the expansion phase investigation is also provided by means of a ball valve (BV) at the lower level of the air vessel. Placing a CV in the air vessel entrance allows the examination of a micro-CAES (compressed air energy storage) system for storing the surge pressure in small-scale (SS) size. A traditional CAES system stores energy from a cheap and sustainable source inside underground caverns which is a major limitation of this technique. Energy is stored as pressure in compressed air in low-peak hours and the stored energy is recovered during high-peak hours. SS-CAES systems are defined in the net power range of 10 kW to 10 MW [12]. SS-CAES systems may present good alternatives for common generators without using electricity [13]. A water-compensated micro-CAES or SS-CAES system is the focus of this study to examine the storage ability of confined AP within a compressed air vessel (CAV). The size of the air vessel needs exact investigation, in this case affecting the applicability and efficiency of the system. Mostly, the cylindrical shape is more economic and practical for air vessels. To achieve the most effective air vessel size, the most important parameters are the maximum pressure and the AP volume fraction within the air vessel. In a previous study, Kiam and Favrat [14] proposed a SS-CAES system with constant air pressure using a column of water. In a review work, Budt et al. [15] presented a wide review of CAES systems from past to future, addressing the main challenges for their development. One challenge that they indicate appears in the lack of appropriate tools for the detailed simulation of CAES systems [15]. Experimental works can provide valuable background for similar challenges by examining various parameters, relating them together and presenting dimensionless variables.

Understanding the behavior of a confined air pocket (AP) under different events of WH seems to give substantial knowledge for future experimental studies or real field designs. In this study, a well-equipped experimental apparatus was used to test and measure AP reaction under three WH events' conditions, namely the occurrence of one WH (1WH), five WHs (5WH) and nine WHs (9WH). In this study, the stored amount of pressure inside the AP for different flow conditions has been assessed. Also, the ability of WH to relocate water to another place with a similar action to the old ram pumps was also examined. To find the behavior of the system as a WH protection vessel, the effect of each parameter on the pressure and the velocity changes along the pipe system has also been investigated. The major considered parameters in this research are constituted by flow velocity, which in the presented graphs is related with Reynolds number (Re); volume fraction ratio (VFR) of air; the storage amount of pressure within each AP; and the relocated volume of water. The VFR is defined as the ratio of air volume to water volume inside the CAV, in percentage.
