Interpretable Predicting Creep Rupture Life of Superalloys: Enhanced by Domain-Specific Knowledge.

Evaluating and understanding the effect of manufacturing processes on the creep performance in superalloys poses a significant challenge due to the intricate composition involved. This study presents a machine-learning strategy capable of evaluating the effect of the heat treatment process on the creep performance of superalloys and predicting creep rupture life with high accuracy. This approach integrates classification and regression models with domain-specific knowledge. The physical constraints lead to significantly enhanced prediction accuracy of the classification and regression models. Moreover, the heat treatment process is evaluated as the most important descriptor by integrating machine learning with superalloy creep theory. The heat treatment design of Waspaloy alloy is used as the experimental validation. The improved heat treatment leads to a significant enhancement in creep performance (5.5 times higher than the previous study). The research provides novel insights for enhancing the precision of predicting creep rupture life in superalloys, with the potential to broaden its applicability to the study of the effects of heat treatment processes on other properties. Furthermore, it offers auxiliary support for the utilization of machine learning in the design of heat treatment processes of superalloys.

Quantitative three-dimensional imaging of chemical short-range order via machine learning enhanced atom probe tomography.

Chemical short-range order (CSRO) refers to atoms of specific elements self-organising within a disordered crystalline matrix to form particular atomic neighbourhoods. CSRO is typically characterized indirectly, using volume-averaged or through projection microscopy techniques that fail to capture the three-dimensional atomistic architectures. Here, we present a machine-learning enhanced approach to break the inherent resolution limits of atom probe tomography enabling three-dimensional imaging of multiple CSROs. We showcase our approach by addressing a long-standing question encountered in body-centred-cubic Fe-Al alloys that see anomalous property changes upon heat treatment. We use it to evidence non-statistical B2-CSRO instead of the generally-expected D03-CSRO. We introduce quantitative correlations among annealing temperature, CSRO, and nano-hardness and electrical resistivity. Our approach is further validated on modified D03-CSRO detected in Fe-Ga. The proposed strategy can be generally employed to investigate short/medium/long-range ordering phenomena in different materials and help design future high-performance materials.

Machine learning-enabled high-entropy alloy discovery.

High-entropy alloys are solid solutions of multiple principal elements that are capable of reaching composition and property regimes inaccessible for dilute materials. Discovering those with valuable properties, however, too often relies on serendipity, because thermodynamic alloy design rules alone often fail in high-dimensional composition spaces. We propose an active learning strategy to accelerate the design of high-entropy Invar alloys in a practically infinite compositional space based on very sparse data. Our approach works as a closed-loop, integrating machine learning with density-functional theory, thermodynamic calculations, and experiments. After processing and characterizing 17 new alloys out of millions of possible compositions, we identified two high-entropy Invar alloys with extremely low thermal expansion coefficients around 2 × 10-6 per degree kelvin at 300 kelvin. We believe this to be a suitable pathway for the fast and automated discovery of high-entropy alloys with optimal thermal, magnetic, and electrical properties.

Massive interstitial solid solution alloys achieve near-theoretical strength.

Interstitials, e.g., C, N, and O, are attractive alloying elements as small atoms on interstitial sites create strong lattice distortions and hence substantially strengthen metals. However, brittle ceramics such as oxides and carbides usually form, instead of solid solutions, when the interstitial content exceeds a critical yet low value (e.g., 2 at.%). Here we introduce a class of massive interstitial solid solution (MISS) alloys by using a highly distorted substitutional host lattice, which enables solution of massive amounts of interstitials as an additional principal element class, without forming ceramic phases. For a TiNbZr-O-C-N MISS model system, the content of interstitial O reaches 12 at.%, with no oxides formed. The alloy reveals an ultrahigh compressive yield strength of 4.2 GPa, approaching the theoretical limit, and large deformability (65% strain) at ambient temperature, without localized shear deformation. The MISS concept thus offers a new avenue in the development of metallic materials with excellent mechanical properties.

Ultrastrong and Ductile Soft Magnetic High-Entropy Alloys via Coherent Ordered Nanoprecipitates.

The lack of strength and damage tolerance can limit the applications of conventional soft magnetic materials (SMMs), particularly in mechanically loaded functional devices. Therefore, strengthening and toughening of SMMs is critically important. However, conventional strengthening concepts usually significantly deteriorate soft magnetic properties, due to Bloch wall interactions with the defects used for hardening. Here a novel concept to overcome this dilemma is proposed, by developing bulk SMMs with excellent mechanical and attractive soft magnetic properties through coherent and ordered nanoprecipitates (<15 nm) dispersed homogeneously within a face-centered cubic matrix of a non-equiatomic CoFeNiTaAl high-entropy alloy (HEA). Compared to the alloy in precipitate-free state, the alloy variant with a large volume fraction (>42%) of nanoprecipitates achieves significantly enhanced strength (≈1526 MPa) at good ductility (≈15%), while the coercivity is only marginally increased (<10.7 Oe). The ordered nanoprecipitates and the resulting dynamic microband refinement in the matrix significantly strengthen the HEAs, while full coherency between the nanoprecipitates and the matrix leads at the same time to the desired insignificant pinning of the magnetic domain walls. The findings provide guidance for developing new high-performance materials with an excellent combination of mechanical and soft magnetic properties as needed for the electrification of transport and industry.

Comprehensive Evaluation of BIM Calculation Quantity in Domestic Construction Engineering Based on Fuzzy Comprehensive Evaluation.

Nowadays, with the leap-forward development of computer technology and the transformation of information technology management concepts, China's construction industry is quietly entering the era of refined management. Accurate estimation and cost control have become among the key considerations of the construction industry. For the calculation of engineering quantity, there are already many software devices that can be used for the calculation of engineering structure quantity, which means that the incorrect operation of personnel has been reduced to some extent, improving the work efficiency and measurement accuracy. The purpose of this paper is to solve the problems of computational missing, computational errors, inefficiency, data loss, and repetitive system in traditional computing based on the advantages of BIM computing system, which provide a reliable basis for cost forecasting and control. At the same time, using BMI calculation system to solve the problem of steel reinforcement in construction engineering, as well as the use of personnel, the existing calculation software still needs a lot of time and energy. We proposed a comprehensive evaluation study of reinforcement calculation in domestic construction engineering BIM calculation system based on a fuzzy comprehensive evaluation. This paper first summarizes the BIM calculation system of construction engineering, uses fuzzy comprehensive evaluation system as an important evaluation index system in domestic construction engineering BIM calculation system, through the judgment of various factors affecting the actual effect of the calculation system, and uses the fuzzy evaluation system combined with a case to demonstrate the superiority of the proposed research. Therefore, through the above research and experiments, it is concluded that the research method of this paper solves many problems in the process of engineering structure reinforcement calculation and provides a good reference method for the establishment of comprehensive evaluation system of reinforcement calculation, as well as providing an effective validation for the widespread use of BIM technology in the construction industry. Finally, it is also beneficial for users to comprehensively evaluate the BIM calculation system of the construction industry and provide a basic reference condition for different industries to use and choose BIM calculation systems.

Crystal-Glass High-Entropy Nanocomposites with Near Theoretical Compressive Strength and Large Deformability.

High-entropy alloys (HEAs) and metallic glasses (MGs) are two material classes based on the massive mixing of multiple-principal elements. HEAs are single or multiphase crystalline solid solutions with high ductility. MGs with amorphous structure have superior strength but usually poor ductility. Here, the stacking fault energy in the high-entropy nanotwinned crystalline phase and the glass-forming-ability in the MG phase of the same material are controlled, realizing a novel nanocomposite with near theoretical yield strength (G/24, where G is the shear modulus of a material) and homogeneous plastic strain above 45% in compression. The mutually compatible flow behavior of the MG phase and the dislocation flux in the crystals enable homogeneous plastic co-deformation of the two regions. This crystal-glass high-entropy nanocomposite design concept provides a new approach to developing advanced materials with an outstanding combination of strength and ductility.

The value of diffusion-weighted MRI to evaluate the response to radiochemotherapy for cervical cancer.

PURPOSE: To investigate the value of apparent diffusion coefficient (ADC) to predict and monitor the therapy response for cervical cancer patients receiving concurrent radiochemotherapy, and to analyze the influence of different b-value combinations on ADC-based evaluation of treatment response. MATERIAL AND METHODS: Seventy-five cervical cancer patients treated with radiochemotherapy received conventional MRI and DWI prior to therapy, after 2 weeks of therapy, after four weeks of therapy and after therapy completion. Treatment response was classified as complete response (CR, n=35), partial response (PR, n=22) and stable disease (SD, n=18), which was determined according to final tumor size after 6 months of therapy completion. Dynamic changes of apparent diffusion coefficients (ADC) and tumor size in the three tumor groups were observed and compared. All the ADCs were calculated from b=0, 600s/mm(2) and b=0, 1000s/mm(2). RESULTS: The ADC increased percentage was higher in CR group than those in PR and SD groups after two weeks and four weeks of therapy, with significant differences in absolute ADCs between CR and PR, SD groups after therapy completion; the overall discriminatory capability for differentiation of CR and PR, SD groups was higher for high b-value combination (0, 1000s/mm(2)) than for low b-value combination (0, 600s/mm(2)). CONCLUSION: DWI can be used as a predictive and monitoring biomarker of treatment response to radiochemotherapy in patients with cervical cancer. High b-value combination may be more reliable to evaluate the treatment response for cervical cancer.