• Structural Engineering and Mechanics
• /
• 제81권6호
• /
• pp.665-675
• /
• 2022

The safety problems of giant hydraulic structures such as dams caused by terrorist attacks, earthquakes, and wars often have an important impact on a country's economy and people's livelihood. For the national defense department, timely and effective assessment of damage to or impending damage to dams and other structures is an important issue related to the safety of people's lives and property. In the field of damage assessment and vulnerability analysis, it is usually necessary to give the damage assessment results within a few minutes to determine the physical damage (crack length, crater size, etc.) and functional damage (decreased power generation capacity, dam stability descent, etc.), so that other defense and security departments can take corresponding measures to control potential other hazards. Although traditional numerical calculation methods can accurately calculate the crack length and crater size under certain combat conditions, it usually takes a long time and is not suitable for rapid damage assessment. In order to solve similar problems, this article combines simulation calculation methods with machine learning technology interdisciplinary. First, the common concrete gravity dam shape was selected as the simulation calculation object, and XFEM (Extended Finite Element Method) was used to simulate and calculate 19 cracks with different initial positions. Then, an LSTM (Long-Short Term Memory) machine learning model was established. 15 crack paths were selected as the training set and others were set for test. At last, the LSTM model was trained by the training set, and the prediction results on the crack path were compared with the test set. The results show that this method can be used to predict the crack propagation path rapidly and accurately. In general, this article explores the application of machine learning related technologies in the field of mechanics. It has broad application prospects in the fields of damage assessment and vulnerability analysis.

• 한국해양공학회지
• /
• 제5권2호
• /
• pp.207-207
• /
• 1991

The crack length measurements by electrical potential(EP) method for 1% Cr-Mo-V and 12%Cr steel of 0.5T-CT specimen were performed at $500^{\circ}C, 600^{\circ}C 700^{\circ}C$, and an applicability of stress intensity factor($K_I$), net section stress($\sigma_{net}$), $C^*$-ingegral and $C_t$ parameter was studied to measure creep crack growth rate(da/dt) with side groove and without side groove under static and cyclic loading condition. The experimental result could be summarized as follows: 1) Crack measurement by EP method was available and coincided with the Johnson,s analytical equation. 2) da/dt by $K_I$ and $\sigma_{net}$ was not adequate because of the wide scatter band according to load and temperature, but $C^*$-integral, except for transition region, was adequate. 3) $C_t$ parameter showed the best fitted line through total creep region without relating with both temperature and load condition. 4) Under the cyclic loading condition, $C_t$ parameter was proper to extimate da/dt. And it was shown that da/dt for 1% Cr-Mo V steel under the static condition(R=1) was 1.16 times faster than the case under cyclic loading(R=0), and for 12% Cr steel, 1.43 times.

• 한국해양공학회지
• /
• 제5권2호
• /
• pp.67-75
• /
• 1991

The crack length measurements by electrical potential(EP) method for 1% Cr-Mo-V and 12%Cr steel of 0.5T-CT specimen were performed at $500^{\circ}C, 600^{\circ}C 700^{\circ}C$, and an applicability of stress intensity factor($K_I$), net section stress($\sigma_{net}$), $C^*$-ingegral and $C_t$ parameter was studied to measure creep crack growth rate(da/dt) with side groove and without side groove under static and cyclic loading condition. The experimental result could be summarized as follows: 1) Crack measurement by EP method was available and coincided with the Johnson, s analytical equation. 2) da/dt by $K_I$ and $\sigma_{net}$ was not adequate because of the wide scatter band according to load and temperature, but $C^*$-integral, except for transition region, was adequate. 3) $C_t$ parameter showed the best fitted line through total creep region without relating with both temperature and load condition. 4) Under the cyclic loading condition, $C_t$ parameter was proper to extimate da/dt. And it was shown that da/dt for 1% Cr-Mo V steel under the static condition(R=1) was 1.16 times faster than the case under cyclic loading(R=0), and for 12% Cr steel, 1.43 times.

• Smart Structures and Systems
• /
• 제18권3호
• /
• pp.569-584
• /
• 2016

This paper presents a method for stochastic modelling of fatigue crack growth and optimising inspection and maintenance strategy for the structural members of steel bridges. The fatigue crack evolution is considered as a stochastic process with uncertainties, and the Gamma process is adopted to simulate the propagation of fatigue crack in steel bridge members. From the stochastic modelling for fatigue crack growth, the probability of failure caused by fatigue is predicted over the service life of steel bridge members. The remaining fatigue life of steel bridge members is determined by comparing the fatigue crack length with its predetermined threshold. Furthermore, the probability of detection is adopted to consider the uncertainties in detecting fatigue crack by using existing damage detection techniques. A multi-objective optimisation problem is proposed and solved by a genetic algorithm to determine the optimised inspection and maintenance strategy for the fatigue affected steel bridge members. The optimised strategy is achieved by minimizing the life-cycle cost, including the inspection, maintenance and failure costs, and maximizing the service life after necessary intervention. The number of intervention during the service life is also taken into account to investigate the relationship between the service life and the cost for maintenance. The results from numerical examples show that the proposed method can provide a useful approach for cost-effective inspection and maintenance strategy for fatigue affected steel bridges.

• 한국세라믹학회지
• /
• 제35권8호
• /
• pp.850-856
• /
• 1998

The dynaamic fatigue behavior of alumina ceramics was observed at room temperature using four-point bending method. Dynamic fatigue fracture strength was observed as function of down speed and notch length. The crack growth exponent of the specimens was calculated from the fracture strength and lifetime in dynamic fatigue test. After loading the stresses in the range of 0% to 105% compared with the average in-ert strength the value of residual fracture strength was measured for unnotched and 0.5mm notched speci-mens at the 0.001 and 0.0005 mm/min down speed respectively. After the 95% stress of the average inert strength was applied repeatedly the value of rsidual fracture strength was measured for 0.5mm notched specimens at the 0.001 and 0.0005 mm/min down speed respectively. The material constant A was found to be almost the same and not to depend on the loading mode or the down speed for unnotched and notched specimen. The value of fracture strength with time calculated from the constants n and A was in good agreement with the measured value.

• 대한토목학회논문집
• /
• 제7권2호
• /
• pp.99-106
• /
• 1987

용접(鎔接) 강구조물(鋼構造物)의 피로균열성장거동(疲勞龜裂成長擧動)에 관(關)한 용접잔류응력(鎔擬殘留應力)의 거동(擧動)을 파악(把握)키 위하여 열처리(熱處理)를 실시(實施)하고 피로균열성장(疲勞龜裂成長) 및 검토(檢討)한 결과(結果) 용접부(鎔接部)는 경화(硬化)되어 늑성(勒性)이 감소(減少)되므로 소재부(素材部)보다 피로균열성장율(疲勞龜裂成長率)이 높았으며 이를 열처리(熱處理)한 결과(結果) 경도(硬度)나 저항력(抵抗力)이 증가(增加)되어 개선(改善)되었으며 $650^{\circ}C$ 정도(程度)에서 가장 이상적(理想的) 이었다. Elber 식(式)은 Paris-Erdogan 식(式)에 비(比)해 균열(龜裂)닫힘현상(現狀)을 고려(考廬)하므로 더 낮은 균열성장율(龜裂成長率)을 가지며 각(各) 열처리(熱處理)에 의한 곡선(曲線)들이 서로 비슷하므로 잔류응력(殘留應力)의 효과(効果)를 고려(考慮)할 수 없었다. 균열(龜裂)길이에 따른 개구하중(開口荷重)은 경도(硬度)가 높을수록 큰 값을 가지나 열처리(熱處理) 후(後)는 경도(硬度)값의 저하(低下)로 피로균열(疲勞龜裂) 성장율(成長率)이 낮아지며 흡수(吸收)에너지와 파괴늑성(破壞勒性)을 증대(增大)시켜 균열저지능력(龜裂沮止能力)을 가져 파괴(破壞)의 위험성(危險性)을 감소(減少)시킨다.

• 한국원자력학회:학술대회논문집
• /
• 한국원자력학회 1998년도 춘계학술발표회논문집(2)
• /
• pp.336-343
• /
• 1998

An iteration method has been developed for determining crack growth and fracture resistance cure (J-R curve) from the load versus load-line displacement record only. In this method, the hardening curve, the load versus displacement curve at a given crack length, is assumed to be a power-law function, where the exponent varies with the crack length. The exponent is determined by an iterative calculation method with the assumption that the exponent varies linearly with the load-line displacement. The proposed method was applied to the static J-R tests using compact tension(CT) specimens, a three-point bend (TPB) specimen, and a cracked round bar (CRB) specimen as well as it was applied to the quasi-dynamic J-R tests using CT specimens. The J-R curves determined by the proposed method were compared with those obtained by the conventional testing methodologies. The results showed that the J-R curves could be determined directly by the proposed iteration method with sufficient accuracy in the specimens from SA508, SA533, and SA516 pressure vessel steels and SA312 Type 347 stainless steel.

• 대한조선학회지
• /
• 제27권3호
• /
• pp.100-106
• /
• 1990

본(本) 연구(硏究)에서는 304스테인리스강을 사용한 평활재와 작은 인공피트재의 고온 피로시험(疲勞試驗)과 크리프 시험(試驗)을 $593^{\circ}C$의 대기중에서 실시하고, 표면의 분포균열을 실온에서 레프리카법으로 연속적으로 관찰하고, 표면균열의 발생, 성장, 합체 거동을 관측하여, 균열 밀도(密度) 길이의 분포등의 정량적인 통계적(統計的) 성질(性質)에 대하여 검토하는 동시에, 파괴역학적으로 시험편의 수명을 지배하는 주균열의 성장거동 및 수명예측을 실시한 것이다. 또 실온의 결과와 고온피로(疲勞) 및 크리프 경우의 미소 분포균열의 통계적(統計的) 결과를 비교 검토하였다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2007년도 춘계학술대회A
• /
• pp.1673-1678
• /
• 2007

The use of flip-chip type electronic package offers numerous advantages such as reduced thickness, improved environmental compatibility, and downed cost. Despite numerous benefits, flip-chip type packages bare several reliability problems. The most critical issue among them is their electrical performance deterioration upon consecutive thermal cycles attributed to gradual delamination growth through chip and adhesive film interface induced by CTE mismatch driven shear and peel stresses. The electronic package in use is heated continuously by itself. When the crack at a weak site of the electronic package occurs, thermal deformationon the chip side is changed. Therefore, we can measure these micro deformations by using Moire interferometry and find out the crack length.

• 한국자동차공학회논문집
• /
• 제4권4호
• /
• pp.171-178
• /
• 1996

DCB(pure mode I) and CLS(mixed mode) tests were performed to investigate the effect of fracture mode on the interlaminar fracture of composite laminate. Mode I critical strain energy release rate was found to be $133J/m^2$ from the DCB test and total strain energy release rate decreased from $1, 270J/m^2$ as thickness ratio(tl/t) varied from 0.333 to 0.667 from the crease from the CLS test. Crack length had no effect on the total strain energy release rate and load was almost constant during the crack growth of the specimen which had the specific thickness ratio. Crack initiated when the stress of the strap ply reached constant stress $42kgf/mm^2$ which was found to be independent of the thickness ratio.