Laboratory Investigation of the Dynamic Response of a Prestressed Composite Steel Cylindrical Tank Subjected to Horizontal Loading

Abstract

In this article, a laboratory investigation on prestressed composite steel cylindrical tanks is performed at different prestressing parameter values (coil span, thickness, and prestressing force). Natural vibration frequencies of a traditional tank and a prestressed composite tank were determined at different filling levels. The research results reveal that prestressing using a steel wire strand has a positive effect in terms of the value of the attenuation coefficient where, when comparing a traditional tank with a coiled tank with a coil span equal to a = 3d, the attenuation coefficient changes in a positive direction by 22.9%; whereas, when comparing a traditional tank with that with a coil span equal to a = d, then the positive effect reaches up to 33%. The value of the attenuation coefficient of a half-filled tank shows that prestressing improves the attenuation coefficient at a coil span equal to a = 3d and up to 8.7%, and with a coil span equal to a = d, up to 26%. Conversely, in the analyses of the tank specimen filled up to the maximum level, the attenuation coefficient changes in a positive direction with a coil span equal to a = 3d, up to 15%, and when accounting for a coil span equal to a = d, up to 35%. In general, the effect of the use of prestressing in terms of the attenuation coefficient shows a positive trend between a percentage range of 8.7 and 35%, depending on the liquid filling conditions, and the vibration amplitudes decrease in a percentage range of 3.8–20%, also depending on the coil span and filling conditions of the tank. The obtained laboratory results positively expand the investigations performed within this research field. As a result, the corresponding findings can be used for the construction and design phases of vertical steel cylindrical tanks.

1. Introduction

The seismic resistance and durability of vertical steel cylindrical storage tanks have always been of particular interest to numerous scholars. Such objects belong to structures of increased responsibility, the destruction of which can lead to consequences on the environment and economy [1,2,3,4]. Given the context concerning the durability of steel tanks, several experimental and theoretical works have been conducted [5,6,7,8], whereby special attention has been paid to scientific solutions aimed at improving the seismic resistance of steel tanks, where, at present, it is possible to use prestressing methods to tune the necessary seismic indicators [9]. These experimental and theoretical works represent an urgent need in the field. Furthermore, since the methods aimed at studying the stress state of structures are wide-ranging at present, they are used in construction [10,11,12,13]. Particularly, in the work of Hosseinzadeh et al. [14], the authors analyzed 161 existing tanks in a refinery complex classified into 24 groups, studying them using both API650-2008 rules and numerical finite element (FE) models. The failure modes and dynamic characteristics of the studied structures were calculated using FE analyses and compared with a set of code requirements. The results demonstrated that, in some cases, there were some deficiencies within the code requirements, such as sliding bottom, elephantine leg arching, squishing, and lifting, which required further evaluations. Seismic effects on steel tanks with different wall thicknesses also require special attention. In fact, the obtained results emphasize the importance of considering seismic effects when designing similar steel structures [15]. Particularly, the main objective of the study implemented by Korkmaz et al. [16] was to evaluate the seismic resistance of Turkish industrial facilities, especially storage tanks, in terms of their seismic resistance. The design of their tank specimen was modeled as a single unit with concentrated mass and spring systems. Moreover, the performance was evaluated on the basis of various earthquake data by using a nonlinear time history analysis. After analyzing the time history, the fragility analysis provided a probabilistic seismic estimation of the tank specimen. With regard to the structure of the specimen, the results obtained via the analyses were evaluated and compared. Such a study identified the vulnerability of storage tanks in Turkey and identified a probabilistic risk based on the results of the analyses. Hamdan [17] instead provided an overview of the behavior and design guidelines for earthquake-prone cylindrical steel liquid storage tanks, where the primary goal was to identify areas in which current design guidelines needed further developments. Field observations during past earthquakes were presented, which were then used, together with FE analyses and published experimental results, to assess the accuracy of current design recommendations, with particular emphasis on EUROCODE 8 [18]. Various phenomena, such as the sloshing and required freeboard, shear of base, and overturning moment applied at the base of the tank, as well as the bending strength of the tanks, were considered in such a work. Furthermore, Chang and Lin [19] considered 242 storage tank accidents that occurred at industrial facilities over the past 40 years. A fishbone diagram was used to analyze the causes leading to the accidents. Corrective actions were also provided to help practical engineers to deal with similar situations in the future. The results showed that 74% of the accidents occurred at refineries, oil terminals, or storage facilities. Fires and explosions accounted for 85% of the accidents. There were 80 accidents (33%) caused by lightning and 72 (30%) caused by human errors, including poor operation and maintenance. Other causes were identified as equipment malfunctions, sabotage, cracks and ruptures, pipeline leaks and ruptures, static electricity, open flames, etc. Most of these accidents could have been avoided if the suitable engineering approaches had been properly applied.

In Zhao et al.’s study [20], based on a simplified scaled model of generic thin-walled cylindrical containers, experiments were conducted to simulate a side impact by using a drop hammer impact tester. Consequently, typical deformations and failure modes of the structure were obtained. After processing the experimental data to obtain a load–strain curve and an energy absorption curve, the curves were analyzed to study the dynamic response and characteristics of cylindrical containers upon impact, as well as the effect of the container length and impact velocity on the dynamic response of the structure. Ansys LS-DYNA software [21] was then applied to simulate the exposure procedure and its dynamic response. The effect of the impact velocity, length-to-radius ratio, and radius-to-thickness ratio on the dynamic performance as well as structural failure modes were investigated using FE parametric analyses combined with experimental results. Similar works on full-scale structures were also developed [22,23,24,25,26].

Increasing the strength of the tank structure with the use of intermediate rings, with due consideration of the geometric dimensions of the structure, was additionally studied [27]. Steel wire strands and composite elements were also used as coils to improve the seismic resistance of the tanks [28]. However, these studies were only performed through the use of modeling simulations, i.e., without any experimental ones using full-scale specimens or corresponding models of the structures [29,30,31]. Several studies were also executed on traditional steel tanks subjected to dynamic loading [32] using numerical approaches [33,34] as well as analytical simulations [35]. In addition, it should be noted that existing foreign and national standards and codes mainly describe design measures and solutions for the design of steel vertical cylindrical tanks, taking seismic loads into account, where the existing methods used to calculate the total hydrodynamic pressure on inertial and convective parameters are presented in the form of a resultant, which represents a rough approximation. Additionally, within the existing calculation methods, there are no constructive solutions to effectively strengthen shell walls in existing reservoirs, which, in turn, can favorably influence the horizontal seismic loading [18,36,37,38,39,40].

The aforementioned literature review showed that the vibration frequencies of a prestressed steel vertical cylindrical tank for oil products required additional studies related to the use of prestressing as a method of protection against dynamic loading, in which a complete coverage of the problem was required by considering different filling levels. Based on the foregoing ideas, the goal of this laboratory investigation is to identify the features of the operation of the main dynamic characteristics of the specimen of a prestressed composite vertical steel tank during free vibrations with the determination of the amplitude and logarithmic decrement of its vibration. This study also aims to establish the actual work, participation in the oscillatory process of the structure wall, and the influence of the design parameters of prestressing on the overall work of the prestressed tank model. The results presented in this article can be used for the design and construction phases of new vertical steel tanks, as well as for strengthening existing ones to extend their service life. Such measures will have a positive impact on the economic indicator and will provide an opportunity to avoid an environmental catastrophe in the event of a spill of petroleum products [1,2,3,4]. At the same time, the obtained findings presented in this article can positively complement our previous related work [35] and can be used by scientific and technical workers as well as being applied in educational training.

2. Materials and Methods

The methodology used to conduct the laboratory investigation was chosen on the basis of a series of rules that consisted of reproducing the operational, technological, geometric, and design parameters with the maximum creation of real conditions through appropriate test equipment and measuring instruments as well as appropriate modern processing software systems. Specifically, the research object was to determine a model of a vertical steel cylindrical tank for oil and oil products created on the basis of the similarity criteria of the specimen and a full-scale object, where steel sheets were butt-joined by welding via weld grinding. Particularly, the geometric dimensions of the specimen of the prestressed composite steel vertical cylindrical tank were equal to: diameter, D = 919 mm; height, h = 596 mm; wall thickness, t = 1 mm; and coil (steel wire strand) diameter, d = 2 mm. The fragments were composed of steel sheets according to GOST 16523-70 [41]. The chemical compositions of the steel grade of the tank and coil were instead determined by laser analyses (elementary laser analyzer called “Elanik”). The results for the spectral analyses of the steel materials, according to GOST 28033-89 [42], are described in

Table 1. Chemical composition of the steel grade of the tank.

Name	CE	Fe	C	Mn	Si	Co	Ni	Cr	V	Nb	Cu	Al	Suitable for Steel
Tank (specimen)	0.087	99.476	0.006	0.398	0.015	-	0.035	0.042	0.004	0.002	0.047	-	Sv-08AA Sv-08A Sv-08

and

Table 2. Chemical composition of the steel grade of the wire strand (coil).

Name	CE	Fe	C	Mn	Si	Co	Ni	Cr	V	Nb	Cu	Al	Suitable for Steel
Wire strand (coil) material	0.206	99.247	0.098	0.491	0.086	-	0.038	0.095	0.003	0.022	0.057	-	10 ps 08 ps St1 ps

.

The mechanical properties of the steel grade of the wire strand and those of the tank specimen are shown in

Table 3. Mechanical properties of the steel grade of the wire strand (coil).

Characteristics	Steel	${\mathit{\sigma }}_{\mathit{B}}$ (MPa)	$\mathit{\delta }$	$\mathit{\psi }$	Hardness, Not Above
			%
Material of the wire strand Material of the coil and tank specimen	Sv-08	310–410	24	60	115
Material of the coil and tank specimen	St08ps	310–410	24	60	115

, where a prestressed coil wire strand with a diameter of 2.0 mm is selected in accordance with GOST 5663-79 [43].

In order to fully and reliably achieve the objectives of this laboratory investigation, three specimens of a steel vertical cylindrical tank were constructed: (1) a traditional tank without a coil; (2) a prestressed tank with a steel wire coil span equal to a = d; and (3) a prestressed tank with a steel wire coil span equal to a = 3d. A general view of the specimens of the traditional vertical steel tank and of the fabricated prestressed composite (prepared for laboratory testing) are illustrated in Figure 1, respectively.

The traditional and prestressed composite specimens of a vertical steel cylindrical tank with coil spans equal to a = d and a = 3d, respectively, with the coil’s tensile force equal to 0.2 from critical (Scr), were constructed in stationary conditions. An electrovibrodynamic stand for inducing horizontal vibrations was additionally used, the stroke of which was equal to 15 cm (Figure 2).

Schematic diagrams of the electrovibrodynamic stands utilized to induce the horizontal vibrations with and without the steel tank specimens are illustrated in Figure 3 and Figure 4.

Figure 3. Scheme of the electrovibrodynamic stand used to induce horizontal vibrations with the steel tank: 1, frame; 2, table; 3, tank specimen; 4, crank; 5, IRWD050motor reducer; 6, table support; 7, M12eyebolt; 8, M12 180 mm rigging screw; and 9, 50 × 8500 mm chain for fixing the tank during transportation (unit: mm).

Figure 3. Scheme of the electrovibrodynamic stand used to induce horizontal vibrations with the steel tank: 1, frame; 2, table; 3, tank specimen; 4, crank; 5, IRWD050motor reducer; 6, table support; 7, M12eyebolt; 8, M12 180 mm rigging screw; and 9, 50 × 8500 mm chain for fixing the tank during transportation (unit: mm).

Figure 4. Scheme of the electrovibrodynamic stand used to induce the horizontal vibrations without the steel tank. The main dimensions are listed in Table 4. (Unit: mm).

Figure 4. Scheme of the electrovibrodynamic stand used to induce the horizontal vibrations without the steel tank. The main dimensions are listed in Table 4. (Unit: mm).

Table 4. The main dimensions of the elements of the used electrovibrodynamic stand (Figure 4).

№	Name	Dimensions (mm)	Number, pcs.
1	Tube 60 mm × 30 mm × 2.0 mm	1140	4
2	Tube 60 mm × 30 mm × 2.0 mm	657	4
3	Tube 20 mm × 20 mm × 1.2 mm	1274	4
4	Plane 80 mm × 80 mm	S = 6 mm	4
5	Plane 80 mm × 80 mm	S = 6 mm	4
6	Plane 220 mm × 180 mm	S = 8 mm	1

Table 4. The main dimensions of the elements of the used electrovibrodynamic stand (Figure 4).

№	Name	Dimensions (mm)	Number, pcs.
1	Tube 60 mm × 30 mm × 2.0 mm	1140	4
2	Tube 60 mm × 30 mm × 2.0 mm	657	4
3	Tube 20 mm × 20 mm × 1.2 mm	1274	4
4	Plane 80 mm × 80 mm	S = 6 mm	4
5	Plane 80 mm × 80 mm	S = 6 mm	4
6	Plane 220 mm × 180 mm	S = 8 mm	1

As a constraining force, an asynchronous motor, a IRWD050 reducer, with a power equal to 0.55 kW, and a 9100-Senries frequency converter were utilized to change and regulate the vibration frequency of the electrovibrodynamic stand for inducing the horizontal vibrations. During the laboratory investigation, the values of the natural frequency vibration varied between a range equal to 0.5–2.0 Hz (Figure 5).

The measuring equipment, devices, and primary voltage converters were selected by considering the expected values from the required measured test parameters, temperature and humidity of the environment, as well as on the basis of some recommendations existing in the literature [44,45,46]. As a primary transducer for measuring the natural vibration frequencies of the steel tank specimen, as well as for determining the first two amplitude values (A₁ − A₂), a LAW laser linear displacement sensor and a BLOXX-A123 controller were utilized. Test commander software and a test viewer visualizer were instead adopted as secondary converters, where the corresponding variables were set up to measure the required parameters. When testing the specimens of the steel vertical tanks, the following geometric dimensions of the shells, prestressing parameters, and dynamic performance were simulated:

−

The specimen diameter, height, and thickness;

−

The span, angle, and tensile force of the coil thread; the diameter of the coil wire strand;

−

The vibration frequency, horizontal displacements, and time of vibration.

The block diagram of the laboratory experiments of the vertical steel cylindrical tanks, simulated as subjected to horizontal seismic loading, is illustrated in Figure 6.

During the performance of the laboratory investigation, the shell fragments and stresses along the shell walls were determined at specific points and, in particular, along the circumferential and axial directions. In accordance with what is illustrated in Figure 7, the position of the LAW laser linear displacement sensors was selected at three positions on the tank specimen (specifically, at the bottom, middle, and top). The displacement sensors were installed in such a way that it was possible to simultaneously measure the dynamic indicators at three distinct locations.

The values of the attenuation coefficient (D), period of the free vibrations (T0), frequency of the free vibrations of the tank wall specimen (f0), and the index of the first two amplitudes (A₁ − A₂) are described in

Table 5. Natural frequency (NF) values of the traditional steel cylindrical tank without liquid.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	11.24	0.088	2.2924	2.2089	0.0835	0.0118
2	1.0	11.89	0.083	2.4529	2.3755	0.0774	0.0102
3	1.5	12.18	0.081	2.6246	2.5457	0.0739	0.0091
4	2.0	13.26	0.075	2.8083	2.7377	0.0706	0.0081

,

Table 6. Comparison between the natural frequency (NF) values of the traditional steel cylindrical tank half-filled with liquid.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	12.92	0.076	1.5170	1.4779	0.0391	0.0083
2	1.0	14.14	0.070	1.6231	1.5833	0.0398	0.0079
3	1.5	12.48	0.080	1.7354	1.6944	0.0411	0.0076
4	2.0	13.58	0.072	1.8568	1.8155	0.0413	0.0072

,

Table 7. Comparison between the natural frequency (NF) values of the traditional steel cylindrical tank filled up to the maximum level with a liquid.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	12.72	0.077	1.3247	1.2979	0.0269	0.0065
2	1.0	13.81	0.071	1.4174	1.3912	0.0267	0.0059
3	1.5	13.98	0.076	1.5166	1.4902	0.0264	0.0055
4	2.0	13.66	0.073	1.6227	1.5979	0.0248	0.0049

,

Table 8. Comparison between the natural frequency (NF) values of the prestressed composite steel cylindrical tank without a liquid at the coil span equal to a = 3d.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	10.01	0.098	2.0631	1.9798	0.0833	0.0131
2	1.0	10.34	0.096	2.2075	2.1274	0.0801	0.0121
3	1.5	11.11	0.088	2.3620	2.2791	0.0829	0.0113
4	2.0	11.96	0.082	2.4173	2.3479	0.0694	0.0092

,

Table 9. Comparison between the natural frequency (NF) values of the prestressed composite steel cylindrical tank half-filled with a liquid at the coil span equal to a = 3d.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	10.88	0.091	1.4259	1.3809	0.0451	0.0102
2	1.0	11.23	0.088	1.5257	1.4801	0.0456	0.0097
3	1.5	11.64	0.085	1.6324	1.5861	0.0463	0.0091
4	2.0	12.88	0.077	1.7140	1.6691	0.0449	0.0084

,

Table 10. Comparison between the natural frequency (NF) values of the prestressed composite steel cylindrical tank filled up to the maximum with a liquid at the coil span equal to a = 3d.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	12.05	0.083	1.2416	1.2093	0.0323	0.0083
2	1.0	12.08	0.082	1.3285	1.2961	0.0324	0.0078
3	1.5	13.10	0.081	1.4582	1.4263	0.0319	0.0072
4	2.0	13.17	0.076	1.5067	1.4748	0.0318	0.0068

,

Table 11. Comparison between the natural frequency (NF) values of the prestressed composite steel cylindrical tank without a liquid at the coil span equal to a = d.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	9.08	0.110	1.8774	1.7899	0.0875	0.0151
2	1.0	9.46	0.105	1.9901	1.9065	0.0836	0.0136
3	1.5	10.08	0.098	2.1294	2.0471	0.0823	0.0125
4	2.0	11.10	0.090	2.2784	2.2037	0.0747	0.0106

,

Table 12. Comparison between the natural frequency (NF) values of the prestressed composite steel cylindrical tank half-filled with a liquid at the coil span equal to a = d.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	11.38	0.101	1.2833	1.2389	0.0443	0.0112
2	1.0	12.24	0.081	1.3731	1.3279	0.0452	0.0106
3	1.5	11.28	0.088	1.4692	1.4234	0.0458	0.0101
4	2.0	13.95	0.071	1.5721	1.5253	0.0468	0.0096

and

Table 13. Comparison between the natural frequency (NF) values of the prestressed composite steel cylindrical tank filled up to the maximum with a liquid at the coil span equal to a = d.

NF№	Frequency (Hz)	Experimental Data of the Natural Frequencies of the Tank Wall (f), (Hz)	Period (T), Sec	Top Amplitude Value (A1) 10−5 m	Bottom Amplitude Value (A2) 10−5 m	Amplitude Difference (A1 − A2) 10−5 m	Attenuation Coefficient (D)
1	0.5	12.85	0.077	0.9561	0.9272	0.0289	0.0097
2	1.0	13.61	0.072	1.0231	0.9941	0.0291	0.0091
3	1.5	12.20	0.070	1.1647	1.1334	0.0313	0.0086
4	2.0	16.01	0.061	1.1922	1.1621	0.0301	0.0081

. In total, three specimens of the steel vertical cylindrical tank were tested in the laboratory. Particularly, the experimental data were obtained from the root-mean-square values of the performed measurements, according to the specific recommendations present in regulatory documents [47].

3. Results and Discussion

This laboratory investigation was executed by assuming that the traditional steel cylindrical tank was subjected to different filling levels, i.e., (1) without liquid, (2) half-filled, and (3) maximum filled. Furthermore, this investigation was also conducted by considering the prestressed composite coil with coil spans equal to a = d and a = 3d, respectively. Consequently, the values of the natural vibration frequencies of the tank specimen and those of the first two amplitude indicators (A₁ − A₂) were obtained. As a result of the obtained data, the period of the free vibrations of the tank specimen wall was calculated, whereas the logarithmic decrement (the attenuation coefficient (D)) was determined according to Equations (1)–(4), as reported as follows [48,49]. The values of the logarithmic decrement are specifically given by the following formula (the attenuation coefficient value):

$$\mathrm{D}=\frac{0.733\ast \log (\frac{A_1}{A_2})}{\sqrt{1+\left[0\right..733\ast \log (\frac{A_1}{A_2}){]}^2}}$$

(1)

The values of the period of the free vibrations are obtained as follows:

$$T_0=T_p\sqrt{1-D^2}$$

(2)

The values of the period of the operational vibrations are determined as follows:

$$T_p=\frac{T_0}{\sqrt{1-D^2}}$$

(3)

While the values of the natural frequency are achieved by the following formula:

$$F_0=1/T_0$$

(4)

Table 5, Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 12 and Table 13 list the results from this laboratory investigation regarding the determination of the natural vibration frequencies under different dynamic horizontal loadings on the specimens of traditional and prestressed composite vertical steel tanks and, moreover, by assuming different filling levels of the liquid.

The processed oscillograms of the free vibrations of the steel tank specimen wall at a loading of 1.5 Hz are shown in Figure 8a–c for all cases, which, in turn, are indicated in Table 5, Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 12 and Table 13. Other values of the free vibrations of the steel tank specimen wall at loadings equal to 0.5, 1.0, and 2.0 Hz were recorded in a tabular form (Table 5, Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 12 and Table 13).

The analysis of the oscillations depicted in Figure 8a regarding the steel cylindrical tank when it was empty shows that prestressing has a positive effect on the attenuation coefficient value, where, when comparing the traditional tank to that with a coil span equal to a = 3d, the value of the attenuation coefficient changes in the positive direction by 22.9%, and when comparing the traditional tank to that with a coil span equal to a = d, the positive effect shows up to 33% (Figure 9). The value of the attenuation coefficient of the tank when it is half-filled (Figure 8b) instead shows that the prestressing application improves such a coefficient with the coil span equal to a = 3d and up to 8.7%, and with the coil span equal to a = d, up to 26% (Figure 10). Furthermore, as in the analyses depicted in Figure 8c, i.e., when the steel tank is filled up to the maximum level, the value of the attenuation coefficient changes in the positive direction with the coil span equal to a = 3d and up to 15%, and assuming the coil span equal to a = d and up to 35% (Figure 11).

Consequently, it can be stated that the use of prestressing for the design of vertical steel cylindrical tanks improved the logarithmic decrement value (attenuation coefficient value) from 8.7% up to 35% depending on the liquid filling conditions. Thus, the value of the vibration amplitude with the use of prestressing decreases by a percentage range equal to 3.8–20%, also depending on the coil span and the conditions of the filling of the tank, which represents a significant positive effect (Table 5, Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 12 and Table 13 and Figure 9, Figure 10 and Figure 11). Therefore, based on the restrictions of this work, it can also be stated that these laboratory experiments did not consider the phenomena of friction between the coil and steel tank walls and other types of fluctuations, which, in turn, will be investigated in a future study. Notably, this article was a continuation of one of our theoretical studies [35], in which a comparison of the results from the theoretical and experimental studies regarding the values of the frequencies, period, and amplitude of the oscillations was satisfactorily conducted. In addition, the findings obtained by this research can positively influence previous results and, moreover, can be adopted by scientific, technical, and design companies and institutions. Furthermore, as a result of this work, a patent [50] and utility model [51] were established.

4. Conclusions

In line with the goals of this laboratory investigation, all scheduled objectives were finally achieved. The specimens of the traditional and prestressed composite steel cylindrical tanks were constructed under stationary conditions with the spans of the steel wire strand coil equal to a = d and a = 3d, respectively, where the wire with a diameter of 2 mm was selected. Yet, under stationary conditions, an electrovibrodynamic stand for inducing the horizontal vibrations was designed, whereas the chemical composition of the steel material of the specimen walls and that of the wire strand were determined via laser analyses. The results of the analyses from the oscillograms when the tank was empty show that the prestressing has a positive effect on the attenuation coefficient value where, when comparing the traditional tank with that with the coil span equal to a = 3d, the attenuation coefficient changes in the positive direction by a percentage of 22.9%, while when comparing the traditional tank with that with of the coil span equal to a = d, the positive effect is up to 33%. The value of the attenuation coefficient when the tank was half-filled instead showed that the use of prestressing improved the attenuation coefficient with the coil span equal to a = 3d and up to 8.7% and with the coil span equal to a = d and up to 26%. Conversely, in the analyses related to the tank when it was filled up to the maximum level, the value of the attenuation coefficient changed in the positive direction with the coil span equal to a = 3d and up to 15% and with a coil span equal to a = d and up to 35%. Therefore, the use of prestressing as a method for the seismic dynamic protection of vertical steel cylindrical tanks showed a positive value, where the effect of applying prestressing in terms of the logarithmic decrement value (attenuation coefficient value) showed a positive trend between a percentage range equal to 8.7–35%, depending on the conditions of filling with a liquid. Moreover, the vibration amplitude value decreased by a percentage range equal to 3.8–20%, depending on the coil span and the conditions for filling the tank, which also represented a positive and important effect. The obtained research results can be used at the design and construction stages of new vertical steel tanks, as well as when strengthening the existing ones to extend their service life. At the same time, the obtained results presented in this article can be used by scientific and technical staff, and can be applied in educational training.