JPH0333557B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for manufacturing an independent rectangular tank installed in a liquefied gas tanker for loading LNG, LPG, etc.

As shown in Figure 1, a liquefied gas tanker has an independent rectangular tank 2 in a hull 1 and a rolling chock 3.
It is supported by a support 4 with a gap between it and the inner surface of the shell 1, so that when LNG, LPG, etc. are loaded into the independent rectangular tank 2, contraction and movement due to low temperatures can be tolerated. The independent rectangular tank 2 containing the low-temperature liquid has a structure that has sufficient strength against deformation stress during contraction.

This independent rectangular tank 2 is made of an aluminum alloy, and as shown in FIG. 2, for example, a large number of tank members 5 such as brackets, ribs, shapes, and skin plates are assembled inside the tank 2, and the joints of each tank member 5 are welded. is being constructed. This type of liquefied gas tanker, which falls under Type B in the IMO Code, is required by the IMO Code to conduct a detailed fatigue strength analysis. This B
The definition of a type tank is that the safety of the tank has been confirmed through accurate analysis and experimentation, and the tank is unlikely to undergo major destruction at any one time.For this reason, emphasis is placed on fatigue analysis and fracture analysis. An independent rectangular tank has a bone structure with many welded tank parts inside, and has a high unsteady constant order, so complete analysis has been difficult, but recently computers have made it relatively easy to analyze. . However, fillet welds and the like are inevitably used frequently in bone structures, and because the weld shape is irregular, it is necessary to conduct fatigue tests on each welded part individually. Conventionally, this fatigue strength analysis of welded joints has been performed based on the nominal stress of the member, and for each type of welded joint, when stress is repeatedly applied, the relationship between the number of times it will fatigue and break and that stress is shown. The fatigue strength of the welded joint with respect to the nominal stress was determined based on the S-N curve obtained through experiments. However, since the fatigue strength differs depending on the shape of the member, the shape of the joint, or the shape of the weld bead, various S-N curves are required, which requires many tests. in fact,
It is impossible to cover all the joints appearing in an independent rectangular tank with the S-N curve determined for each test.

The present inventors found that the fatigue strength of welded joint members is influenced by the shape of the welded joint, metallurgical properties, residual stress, and other factors. The inventors have discovered that this is the stress concentration factor of the fatigue crack occurrence area, particularly the stress concentration factor of the weld toe, and have thus arrived at the present invention.

An object of the present invention is to easily perform fatigue analysis of various welded joints of the above-mentioned independent rectangular tank, and to
The present invention provides a method for manufacturing an independent rectangular tank for a liquefied gas tanker, which allows for easy fatigue-resistant design of various tank members.

Consists of numerous aluminum alloy tank components,
In addition, in the manufacturing method of an independent rectangular tank for a liquefied gas tanker in which the joints between the members are welded, the stress concentration coefficient Kt at the weld toe is calculated for the nominal stress generated in the plate material at the welded joint between the members. The local stress at the weld toe is determined from the function of the flank angle θ, toe radius ρ, and bead height h, and the local stress at the weld toe is determined from the stress concentration coefficient Kt, and a fatigue fracture curve is created based on the local stress. The method is characterized in that the allowable local stress of the corner welded joint is determined from the fatigue fracture curve, and the flank angle θ, toe radius ρ, and bead height h are controlled and welded so that the stress concentration factor Kt is 3.0 or less. A manufacturing method for an independent rectangular tank for a liquefied gas tanker, in which the fatigue fracture curve of a welded joint including the base metal is determined based on the local stress calculated from the stress concentration factor.
By creating an N curve, designing the fatigue strength of each welded joint based on this, and controlling the weld shape of the welded joint, an independent rectangular tank whose fatigue strength can be easily analyzed can be manufactured.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment of the method for manufacturing an independent rectangular tank for a liquefied gas tanker according to the present invention will be described below with reference to the accompanying drawings.

First, the stress concentration coefficient Kt will be explained using a cross-welded joint as an example in FIG.

In FIG. 3, it is assumed that 6 is a vertical plate and 7 is a horizontal plate, and a fillet weld 8 is made between the vertical plate 6 and the horizontal plate 7. In this case, the reinforcement angle (hereinafter referred to as flank angle) of the weld 8 is θ, the bead height is h, the heed toe radius of the toe 9 of the weld 8 is ρ, and the thickness of the horizontal plate 7 is t.

Assuming that a nominal stress σ _N is now applied to the horizontal plate 7, the stress distribution will be as shown at 10. Although the nominal stress σ _N remains at the unwelded horizontal plate 7, the toe 9
Stress concentration occurs as the distance approaches , and approximately the maximum local stress σ _L acts on the toe 9. Therefore, if the stress concentration coefficient at this toe 9 is Kt, Kt is expressed by the following formula.

K _t =σ _L /σ _N (1) Moreover, this Kt is expressed by the following formula as a function of the above-mentioned flank angle θ, toe radius ρ, and bead height h.

K _t = [1+f(θ){g(ρ)-1}]C(a/t)
...(2) Here, f(θ) is the influence of flank angle, g(ρ)
is a function of the influence of the radius of the weld toe, and C(a/t) is a function of the influence of the presence of an unwelded portion, and is expressed by the following formulas.

flank angle; Toe radius; g (ρ) = α _t・g _t (ρ) + α _b・g _b (ρ)……(4
) Here, gt(ρ) is given by the following formula in case of tensile load.

g _t (ρ)=1+β _t {(h/ρ)・1/2.8(W/t)−
2}0.65...(5) Here, βt takes the following value depending on the shape of the welded joint. Cross joint 2.2, butt joint 2.0, T joint 1.0.

In addition, gb (ρ) in the case of bending load is given by the following formula, regardless of the joint shape.

g _b (ρ)=1+β _b・{tanh ^1/2 (2tp/t+2h
+2ρ/t)} × [tanh{t+2h/t-1 ^1/4 /1-ρ/1}
] {0.13+0.65 (1-ρ/t) ⁴ / (ρ/t) ^1/3 }...
(6) Here, βb takes the following value depending on the shape of the welded joint. Butt joint = 1.5, T joint = 1.9, other joints = 1.0.

Unwelded area; C (a/t) = 1 + 0.64 (a/t) ²
/(2hp/t)−0.12(a/t) ⁴ /(2hp/t) ² …(7
) Here, W in equations (3) and (5) is used depending on the type of welded joint as follows.

Cross joint; W = (t+4h) + 0.3 (tp + 2hp) Butt joint; W = t + 2h + 0.6hp T joint; W = (t + 2h) + 0.3 (tp + 2hp) ... (8) However , (2) to (8), where θ is the flank angle, ρ is the toe radius, t is the plate thickness of the loaded member, tp is the plate thickness of the unloaded member, h is the bead height (leg length), hp is the weld leg length, a is the length of the unwelded part, αt is the tensile stress coefficient at the toe (1 for tensile load, 0 for bending load), αb is the bending stress coefficient at the toe (For bending load = 1, for tensile load = 0).

As a result of looking at the stress concentration coefficient Kt in various joints using Kt obtained from equations (2) to (8) above, flank angle θ, toe radius ρ, and bead height h as parameters,
It was found that the influence of the flank angle θ and bead height h is relatively small, and that the stress concentration coefficient Kt changes mainly depending on the size of the toe radius ρ. Using this result as an example for a T-shaped joint and a cruciform joint, the fourth
As shown in Fig. 5.

FIG. 4 shows the relationship between the toe radius/plate thickness and stress concentration factor in a T-joint.

From the graph in FIG. 4, it can be seen that if the plate thickness t is constant, the stress concentration coefficient Kt increases as the toe radius ρ becomes smaller. When the toe radius ρ is small, the stress concentration coefficient changes greatly depending on the flank angle θ (100 to 170°). It can be seen that there are fewer changes.

Further, similarly to the graph of FIG. 5, it can be seen that if the plate thickness t is constant, the stress concentration factor increases as the toe radius ρ becomes smaller.

In the above, by determining the stress concentration coefficient Kt, the local response σ _L at the upper end portion 9 can be determined from the above equation (1). In other words, the local stress σ _L can be found from the formula below.

σ _L =K _t・σ _N Therefore, for example, when creating an S-N curve when the nominal stress σ _N is repeatedly applied to the horizontal plate 7 of the cruciform joint shown in FIG. 3 until it breaks, If the S-N curve is drawn using the local stress σ _L instead of the nominal stress σ _N , a unified S-N diagram can be obtained. This S-N diagram is shown in FIG. In the figure, the vertical axis shows the local stress σ _L at the toe within the stress range, and the graph plots the points at which the butt joint, cruciform joint, and T-joint break when each stress is repeatedly applied. The straight line shown indicates the probability of survival among the number of test specimens.

Since this S-N curve is drawn based on the local stress σ _L , the fatigue strength of each joint can be easily determined.

The stress concentration factor Kt is normally distributed over a wide range if welding is performed without control. For this reason, for example, Kt value 7.0
, the local stress σ _L on the toe is the nominal stress σ _N
It becomes 7 times larger than that, and its fatigue strength decreases. Therefore, in order to increase the fatigue strength, it is necessary to design the plate to have a large thickness and a large cross-sectional area to reduce the overall stress applied to the plate. That is, when welding joints, if welding is performed without control, the welding cost will be lower, but on the other hand, the Kt value will tend to increase, so thick plates must be used, and the material value will increase.

In the present invention, the welding shape of each welded joint of an independent rectangular tank is kept constant in advance, each joint is welded, and the stress concentration coefficient distribution at each toe is analyzed. If the stress concentration coefficient is 3.0 or less, the tank is welded. We confirmed that it can be designed safely.

Figure 7 shows the distribution of Kt values when T-joints are welded under the same conditions.The total number of Kt value measurements is assumed to be 100%, and the frequency is shown for each Kt value. It is something. In this Figure 7, the stress concentration factor Kt under the same welding conditions has a maximum frequency (14%) around 1.5, and the distribution range is as follows:
Distributed from 1.13 to 2.38. Also, based on the distribution curve in Figure 7, the probability for each Kt value was integrated from the maximum value to the minimum value of Kt as shown in Figure 7, and the probability for each Kt value was shown in the graph in Figure 8. .

In the figure, the dotted line l shows the graph when welding is carried out without control, n shows the average Kt value, the dashed-dotted curve m shows the graph integrating the frequency distribution in Figure 7, and o shows the average Kt value. It shows the value.

As shown by the dotted line l, if welding is performed without control, Kt
The values are distributed over a wide range from the maximum to the minimum, but as shown by curve m, by welding under constant welding conditions, the maximum stress concentration factor Ktmax appears to be 3.0.
We can see below that the probability is definitely zero.

Therefore, if each joint is designed for welding so that the stress concentration factor Kt is 3.0 or less, a safe fatigue strength can be achieved against the nominal stress.

From the above, the maximum nominal stress σ _N and Kt on the base metal
The local stress σ _L is calculated from the value, and the S-N curve (fatigue fracture curve) in Figure 6 is used to determine whether the local stress σ _L is within the allowable stress. θ, toe radius ρ, and bead height h) are determined to be appropriate, and the Kt value is controlled so that it falls within the allowable local stress.

Next, a welding method for controlling the stress concentration factor Kt within 3.0 will be explained. Figure 9 shows an example of welding a T-shaped joint. In the figure, 11 is the lower plate, 1
2 is a standing board, and the joint 1 between the lower board 11 and the standing board 12 is
3 is MIG welded using the torch nozzle 14. In this case, the position of the electrode 15 at the tip of the torch nozzle 14 is directed toward the seam 13, and the torch angle θt is maintained at a constant value (for example, 45°).
From this state, the torch nozzle 14 is oscillated left and right. This oscillation width is the vertical plate 12 and the lower plate 1.
The amplitude should be set so that sufficient welding strength can be obtained according to the thickness of the plate (usually about ±4 mm).

Normally the number of oscillations in MIG welding is 70 to 80.
times per minute, but the stress concentration factor cannot be controlled with this oscillation number. The present invention makes it possible to control the stress concentration coefficient Kt at the toe to 3.0 or less by performing oscillations at a rate of 150 to 250 times/minute. In this case, by increasing the number of oscillations, it is possible to increase the toe radius ρ at the toe, for example, the toe radius ρ can be increased.
It becomes possible to set it to 1.0 mm or more.

Furthermore, instead of controlling the oscillation number, 50% or more of helium gas can be mixed into the argon shielding gas from the torch nozzle 14, and the stress concentration coefficient Kt can be made 3.0 or less. In other words, helium gas has a higher thermal conductivity than argon gas, so the filler metal melts better.
It becomes possible to lower the Kt value.

All of this welding is performed using automatic welding materials, and it is possible to weld the mutual joints of tank members 5 in the independent rectangular tank 2 shown in FIGS. 1 and 2 with a stress concentration coefficient of 3.0 or less at the weld toe. It becomes possible. Therefore, since the Kt value of each tank member 5 can be kept to 3.0 or less, the tank member 5 can have an economical thickness commensurate with the thickness stress and also have sufficient fatigue strength.

As is clear from the above detailed description, the present invention provides the following effects.

(1) Since the stress concentration factor Kt can be determined by calculation and only one S-N curve is required, the fatigue test for each joint can be greatly simplified.

(2) The stress concentration factor Kt can be confirmed by measuring the welded shape of the actual structure, so it is highly reliable for the fatigue strength of the actually completed structure.

(3) Since the stress concentration factor Kt can be statistically understood through measurement, reliability analysis is possible in the truest sense.

[Brief explanation of the drawing]

FIG. 1 is a sectional view showing an independent rectangular tank for a liquefied gas tanker according to the present invention, FIG. 2 is a perspective view showing an independent rectangular tank, and FIG. 3 is a method for manufacturing an independent rectangular tank for a liquefied gas tanker according to the present invention. FIG. 4 is a graph showing the relationship between the toe radius and stress concentration factor in the T-shaped joint of the present invention, and FIG. A graph showing the relationship between the toe radius and the stress concentration factor in the cruciform joint part in the invention, FIG. 6 is a graph showing the S-N curve in the invention, and FIG. 7 is a graph showing the distribution of the stress concentration factor in the invention. ,
FIG. 8 is a graph of integrating the stress concentration coefficient distribution according to the present invention, and FIG. 9 is a perspective view showing an example of welding each joint in the method for manufacturing an independent rectangular tank for a liquefied gas tanker according to the present invention. be. In the figure, 2 is an independent rectangular tank, and 3 and 5 are tank members.

Claims (1)

[Claims] 1. In a manufacturing method for an independent rectangular tank for a liquefied gas tanker, which is constructed from a large number of aluminum alloy tank members and whose joints are welded, the nominal stress generated in the plates at the welded joints between the members shall be The stress concentration factor at the weld toe
Kt is determined from the function of the flank angle θ, toe radius ρ, and bead height h, and the local stress at the weld toe is determined from the stress concentration coefficient Kt, and a fatigue fracture curve is created based on the local stress. , find the allowable local stress of each weld joint from the fatigue fracture curve, and weld by controlling the flank angle θ, toe radius ρ, and bead height h so that the stress concentration factor Kt is 3.0 or less. A method for manufacturing an independent rectangular tank for a liquefied gas tanker.