0　Introduction

Welding temperature field plays an important role in the welding process. Accurate and reliable measurement of temperature field during arc welding process is quite challenging due to the severe environment. However, in some cases, obtaining the accurate range of temperature field is necessary. Recently, China offshore oil Engineering company (COOEC) is plagued by this problem. The PPG PITT-CHAR XP flame retardant system has been used by COOEC to preventing the thermal softening of steel in the high temperature. But the fireproof paint near the welded joints should be reserved prior to welding. Otherwise, the fireproof paint will be burned by the high temperature during arc welding process. The degradation temperature of PPG PITT-CHAR XP fire retardant paint is 80 ℃ and it is costly. Therefore, getting the accurate range of 80 ℃ temperature field is necessary.

Recently, the model of ‘theory- numerical simulation- production’ has been developed and the scientist and quantitative analysis could be achieved based on the application of welding numerical simulation technology. Measurement of temperature field is costly and time-consuming. Hence, the methods of numerical simulation are often been employed to analyze the temperature field. In this study, a thermal analysis has been carried out using the software of SYSWELD to predict the temperature field distribution of typical welded joints under the given welding parameters, material and thickness. The material thermal physical properties, non-linear relationship between material latent heat of phase change and temperature is also considered in the analysis[1]. Then the results of numerical simulation will be verified by experimental results.

1　Experimental procedure

1.1　The building of FEM

Three-dimensional numerical models were developed based on the entity structure with the 3-D analysis software SYSWELD. There were about one hundred models created. These models were not exactly same as the entity structure. For example, one of the typical entity structure diagrams was shown in Fig.1. Two H-steel plates and a stiffened plate were welded together. This typical entity structure could simplified into two typical welded joints which were butt joints(Fig.2) and T-joints(Fig.3). In the welding process, the next group of multi-pass welds would not start until the last group of multi-pass welds were cooling enough. So the influence of temperature field on each other could be ignored. Hence, simplifying complex construction into simple construction is reasonable. In order to investigate the effect of groove type on temperature field, three kinds of butt-welding models with same thickness were developed. And the groove shapes are K-shape, V-shape and X-shape, as shown in Fig.4a, Fig.4b, Fig.4c.

Fig.1　The typical entity structure

Fig.2　The FEM of butt joint (mm)

Fig.3　The FEM of T-joint (mm)

Fig.4　The different groove shapes of 18 mm butt joints (a) The K-shape butt joint (b) The V-shape butt joint (c) The X-shape butt joint

1.2　Heat source and calibration the heat source

Shieldedmetal arc welding (SMAW) and flux cored arc welding (FCAW) are both used in the welding process without preheating and heat treatment. The welding parameters are shown in the Table 1. S355J2G3 is selected as the material for numerical simulation and its physical properties were showed in the Table 2. Double ellipsoid model is selected to be the heat source model [2]. The model uses the following equation to define the volumetric heat flux (q) inside the front and rear regions of the heat source, where these regions are denoted by the subscripts f and r, respectively:

马云曾经在公开演讲中表示，自己对阿里的CEO张勇说，不超过200亿元人民币的投资你就决定，你拍板，不用告诉我。这句话中国重汽济南区的经理张超的解读是，一是马云对张勇这个接班人很放心，二是阿里200亿元以下的投资太多，马云根本顾不过来。

(1)

where a, b, c describe the dimensions of the heat source, v the welding velocity, t the time, τ a lag factor defining the position of the heat source at t = 0, ff, fr defines the fraction of the heat deposited in the front and rear regions, respectively. Q is the power input from the welding heat source, that is:

Q=ηUI

(2)

The 3D dynamic numerical simulation for multi-pass welding temperature field was carried out. In general, about one hundred kinds of welded joints which were simplified from the typical entity structures were simulated by SYSWELD. Then the range of temperature field above 80 ℃ was measured with the help of SYSWELD. There were many more results than I have shown here in Fig.5.

Via analyzing the data, we found that maximum range of welding temperature field above 80 ℃ increasing with the increasing of weld layers at first and it inclined to stable value later. But the experimental results became larger and larger as the increasing of weld layers and didn't reach a plateau as shown in Fig.6. In order to figure out reasons, the maximum interpass temperature of each welding process was compared with the simulation value[4]. They were very similar in trend. The maximum interpass temperature kept increasing with the increasing of 80 ℃ welding temperature field. When the temperature inclined to a stable value, the range of temperature got to a stable value, too. Due to welding interpass temperature decided by welding interval time , the time interval was a certain value in the simulation. While the interval time always changed in realistic welding process. Therefore, the simulation results could show good laws in some cases and the experimental results could not. Fig.6 and Fig.7 show that there was a big difference between simulation results and experimental results. It was caused by workers' unstandardized operation. Interpass temperature should not high than 180 ℃ in the stand-ards, while the experimental value has reached to 215 ℃. Standardized operation should receive enough attention in practical project. And this suggestion has been adopted by COOEC. Up to now, large economic benefits have been obtained.

Different groove corresponds to different bead layout, so it can be foreseen that the groove type has a significant effect on temperature history[5]. Three models with different groove types (V-groove, K-groove and X-groove) were developed in this study. The dimension of each weld plate is 300 mm 300 mm 12 mm.

(3)

The range of 80 ℃ temperature field was measured by IR. Then experimental results are compared with the simulation results. Measured value is 37 mm in a single weld. And simulation value is 35 mm. The relative error is less than 6%.

Table 1　Welding parameters

LayersWeldingprocessDiameter/mmBrandType&polarityCurrent/AArcvoltage/VGastypeTravelspeed/(mm·min-1)Heatinput/(kJ·mm-1)Interpasstemperature/℃RootSMAW3．2LB-52UDC(+)10020—432．8<200FillFCAW-G1．2TWE-711NiDC(+)20027CO22001．62<200CapFCAW-G1．2TWE-711NiDC(+)19026CO22201．35<200

Table 2　The physical properties of S355J2G3

Temperature/℃Specificheat/(J·kg-1·K-1)Thermalconductivity/(W·mm-1)Density/(kg·m-3)04300．04678151005000．04678153005800．04378154006100．04178156007100．03578158008650．024781515006300．0327290

2　Results and discussions

2.1　The law of temperature field distribution

where η denotes arc efficiency (in this study, 0.8 was used), U the voltage of the weld, I the current of the weld.

Heat source calibration makes the simplified structure have the same boundary conditions with original structure. So it is very important to numerical simulation.Advanced technique of infrared radiation (IR) is a convenient, non-contact and fast response method making the base metal temperature measurements[3]. During welding, convection and radiation coexist. The coefficient of natural convection is about 5-30 W/(m2·K), and the radiation's coefficient is calculated by the formula recommended by SYSWELD software. The total heat transfer coefficient h0 is equal to the sum of convection and radiation coefficient considering fireproof paint and can be expressed as follows:

Fig.5　The range of 80 ℃ temperature field of different welded joints (a) Temperature distribution of T- joint (b) Temperature distribution of butt joint

Fig.6　Maximum range of 80 ℃ temperature field after every welding process

Fig.7　Maximum interpass temperature after every welding process

2.2　Groove type

本文基于通州城市副中心的建设背景，对农村公路在“城市副中心”背景下的规划设计进行了问题研究探讨，为以后规划副中心周边的农村公路进行了问题引导，同时结合通州区农村公路规划进行案例阐述，可作为今后其他副中心城市以及发达地区农村公路网规划问题的参考.

On the base of specified total quantity of heat, effects of the region of heat flux distribution on the temperature distributions were discussed[6]. Via analyzing the data, we found that the range of temperature field was linked to ∑α (∑α=α1+α2+…+αn). A sketch map was made to help explain this issue, asshown in Fig.8. Shaded area represented the region of heat flux distribution. The ∑α of V-groove joint was 55°.The ∑α of K-groove joint and X-groove joint were 90° and 110°.This result proved that the smaller the sum of angle was, the more heat it would accumulate. The bigger the region of heat flux distribution was, the larger range of temperature field it would get[7]. So the X-groove joint got the biggest range of temperature field above 80 ℃ under the same condition.

Temperature field was investigated based on SYSWELD. In this part, a thermal cycle curve was taken as the heat source. And the same heat source (thermal cycle curve) was used in these three models. The purpose was to ensure total quantity of heat being similar. At last, maximum range of welding temperature field above 80 ℃ was measured. The measured values were shown in Table 3. From the Table 3, we found that X-groove model got the biggest range of temperature field, then the K-groove model. And the V-groove model got the smallest range of temperature field.

所采用的仪器操作条件如表1。各元素测定同位素为56Fe、68Zn、75As、118Sn、121Sb、208Pb、209Bi。

要说钛金属在腕表的使用上是很普遍的，比如很早就有作表壳的，Richard Miller理查德米尔用钛合金制作夹板。但用来作钟表核心擒纵装置的摆轮轴的还真没有。原因就是摆轮轴尖要有很高的硬度，之前的钛合金都达不到这样的硬度。带过钛金表壳的表的朋友，都知道相比不锈钢表壳钛金属表壳相对更软一些，更容易留下划痕。

Table 3　Maximum range of welding temperature field above 80 ℃

WeldingprocessLengthof80℃temperaturefieldwithV⁃groove/mmLengthof80℃temperaturefieldwithK⁃groove/mmLengthof80℃temperaturefieldwithX⁃groove/mm16583852118127130315916216041771801825182195199

Fig.8　The distribution of heat fluxin different groove shapes (a)Region of heat flux distribution of K-groove joint (b)Region of heat flux distribution of V-groove joint

3　Conclusions

(1) In welding process, the range of temperature field above 80 ℃ keeps increasing until the maximum interpass temperature inclines to stable value. At this point, the range of temperature field also becomes a stable value.

1906年12月，汪优游、朱双云、王幻身等在上海创立开明演剧会。“是年（1906年）江皖患水，饿殍载道。双云、优游辈，谋有以赈之者，乃发起开明演剧会，演剧助赈。”[2]541907年2月，开明演剧会在上海县城举行赈灾公演，历时5天，受到上海市民的欢迎，成为清末民初上海学生演剧活动成功的范例之一。[2]52在汪优游撰写的《我的俳优生活》一文中也有开明演剧会的详细记录：

(2) Time interval between layers plays an important role in affecting the temperature field. If you want to get the certain range of temperature field, specific time interval between layers should be set.

(3) The bigger the region of heat flux distribution was, the more heat it would accumulate and the larger range of temperature field it would get.

References

[1]　Li W S, Bai S W, Yan C Y, et al. Numerical simulation of temperature field for 9Ni alloy steel welding. Electric Welding Machine, 2009, 39(1): 1-4. (in Chinese)

[2]　Ravisankar A, Velaga S K, Rajput G, et al. Influence of welding speed and power on residual stress during gas tungsten arc welding (GTAW) of thin sections with constant heat input: A study using numerical simulation and experimental validation. Journal of Manufacturing Processes, 2014, 16(2):200-211.

[3]　Zhou G T, Zhao Z Y, Liang G L, et al. Investigation on the welding temperature field measurement in gas tungsten arc welding. Materials Science Forum, 2011, 704-705:786-789.

[4]　Tong L, Kang Q, Wang L, et al.Numerical simulation of temperature field in multi-pass compound groove weld for high strength pipeline steel X80. International Heat Transfer Conference, Washington, DC, AUG 08-13, 2010. USA: ASME, 2010：891-893.

[5]　Ye Y H, Cai J P, Jiang X, et al. Influence of groove type on welding-induced residual stress, deformation and width of sensitization region in a SUS304 steel butt welded joint. Advances in Engineering Software, 2015, 86:39-48.

[6]　Stokes-Griffin C M, Compston P. An inverse model for optimisation of laser heat flux distributions in an automated laser tape placement process for carbon-fibre/PEEK. Composites Part A Applied Science & Manufacturing, 2016, 88:190-197.

[7]　Wang J H, Chen C. Computer system of welding heat transfer with different groove types. Transactions of The China Welding Institution, 1990, 11(1):57-64.(in Chines)