Wyniki wyszukiwania - BazTech

1

Ograniczenia współczesnych generatorów obrazów opartych o AI w procesie projektowania architektonicznego

Gil Adam

Builder

|

2024

|

R.28, nr 3

28--33

PL

Wraz z wprowadzeniem narzędzi CAD, a potem BIM, architekci zaczęli dostrzegać, że narzędzia, których używają, mają wpływ na ich twórczość. Obecnie architekci zaczynają używać narzędzi AI umożliwiających eksplorację koncepcji projektowych poprzez generowanie obrazów. Te narzędzia, zwane generatorami obrazów AI, generują obrazy na podstawie opisów tekstowych oraz dowolnych wczytanych obrazów, w tym szkiców lub zdjęć. Na rynku dostępne są zarówno narzędzia przeznaczone bezpośrednio do projektowania architektonicznego, jak i narzędzia o charakterze ogólnym, służące do generowania dowolnych obrazów, od fotorealistycznych po malarskie i graficzne. Wszystkie te narzędzia są we wstępnych fazach rozwoju. Artykuł analizuje przydatność i ograniczenia tych narzędzi z perspektywy architekta, biorąc pod uwagę stan technologii na koniec 2023 roku.

EN

With the introduction of CAD and then BIM tools, architects have begun to recognize that the tools they use have an impact on their creations. Today, architects are beginning to use AI tools that enable them to explore design concepts by generating images. These tools, called AI image generators, generate images based on text descriptions and any loaded images, including sketches or photos. There are tools on the market that are designed directly for architectural design, as well as general-purpose tools for generating any image, from photorealistic to painterly and graphic images. All of these tools, are in the preliminary stages of development. This article examines the usefulness and limitations of these tools from an architect's perspective, taking into account the current state of technology at the end of 2023.

2

Analiza metod pracy projektantów korzystających z generatorów obrazów opartych na sztucznej inteligencji w kreatywnych fazach projektowania architektonicznego

Gil Adam

Builder

|

2024

|

R.28, nr 3

22--27

PL

Użycie AI do wizualizacji pomysłów przeszło w ostatnich dwóch latach (2022/2023) z fazy eksperymentalnej do fazy praktycznej. Obserwowana jest ekspansja mediów tworzonych z użyciem AI w obszarach grafiki użytkowej, reklamy, a także w architekturze. Młodzi architekci zainteresowani są użyciem nowych narzędzi opartych o AI, a na rynku pojawiły się różnorodne płatne narzędzia bazujące na szeroko rozumianej AI skierowane do architektów lub szerzej na rynek architektoniczno-budowlany. Architekci podejmują także próby wykorzystywania narzędzi AI ogólnego zastosowania w szczególności generatorów obrazów. Niniejszy artykuł skupia się na omówieniu efektów eksperymentu z użyciem generatorów obrazów AI, który został przeprowadzony przy współudziale grupy studentów II stopnia kierunku architektura, i wyciągnięcia wniosków z tego eksperymentu. Analiza ta pozwala na wskazanie podobieństw i różnic w sposobach pracy z i bez użycia AI oraz na określenie potencjalnego wpływu tego narzędzia fazy twórcze projektowania.

EN

The use of AI to visualize ideas has transitioned from the experimental phase to the practical realm over the past two years (2022/2023). A noticeable uptick in the utilization of AI-generated media has been observed across various domains, including graphic design, illustration, advertising, and related fields. These creative and visually-oriented areas share commonalities with architectural design. They also engage similar tools for idea development, analysis, and presentation. Consequently, architects have shown a keen interest in incorporating new AI-based tools into their workflow. This interest is evident in the emergence of a variety of paid AI tools catering specifically to architects or the broader architecture and construction market. Additionally, architects are exploring the use of general-purpose AI tools (especially AI image generators), a trend particularly embraced by young designers unreserved in adopting novel tools. This article delves into the outcomes of an experiment conducted with a group of masters-degree architecture students, drawing insights from the experiment's results regarding the anticipated impact of such tools on the architectural design process in the near future. The study itself will be briefly discussed, and the visual results obtained during the experiment will be presented to a limited extent.

3

The unipore and bidisperse diffusion models for methane in hard coal solid structures related to the conditions in the Upper Silesian Coal Basin

Karbownik Marcin, Krawczyk Jerzy, Schlieter Tomasz

Archives of Mining Sciences

|

2020

|

Vol. 65, no. 3

591--603

EN

The safety of mining operations in hard coal mines must be constantly developed and improved. There is ongoing multi-directional research focused at best recognition of the phenomenon associated with the properties of the coal-gas system and its connections with mining and geological conditions. This article presents the results of sorption experiments on coals from the Upper Silesian Coal Basin, which are characterized by varying degrees of coalification. One of the parameters that describes the kinetics of methane sorption, determining and providing valuable information about gas hazard and in particular the risk of gas and rock outbursts, is the effective diffusion coefficient De. It is derived from the solution of Fick’s second law using many simplifying assumptions. Among them is the assumption that the carbon matrix consists of only one type of pore-micropores. In fact, there are quite often at least two different mechanisms, which are connected to each other, related to the diffusion of methane from the microporous matrix and flows occurring in voids and macropores. This article presents both the unipore and bidisperse models and a set of comparisons which fit them to experimental curves for selected coals. For some samples the more complex bidisperse model gave much better results than the classic unipore one. The supremacy of the bidisperse model could be associated with the differences in the coal structure related to the coalification degree. Initial results justify further analyses on a wider set of coals using the methodology developed in this paper.

4

Modelling of mass transport in watercourses at unsteady states

Bielski A.

Environment Protection Engineering

|

2012

|

Vol. 38, nr 3

159--173

EN

In terms of quality particularly difficult to describe are processes of mass exchange between different phases (e.g., atmospheric air-water, water-river sediment, water-algae, etc.). Whitman's model is most often used to describe the mass transport processes through the phase boundary. Theoretical analysis of the mass transfer process through the phase boundary showed that in unsteady states, the calculation results obtained from Whitman's model differ from the results obtained using the accurate diffusion model. These differences are due to the fact that concentration profiles in the direction of diffusion process change in time. Assumptions for Whitman's model do not include changes in the concentration distribution over time. Therefore, the correction factor was introduced to Whitman's model. The correction factor is defined as a parameter that multiplies a concentration derivative over time in the mass transport model. The correction factor can be used to estimate the effective diffusion coefficient of the substance that permeates from the aqueous phase to the sediment. The correction factor improves the degree of fit of the mass transport model to the measurement data. It can be used to estimate the effective turbulent diffusion coefficient from water phase to the sediment phase.

5

Modelling of mass transport in watercourses considering mass transfer between phases in unsteady states. Part I. Mass transfer process for periodic and aperiodic changes of concentration

Bielski A.

Environment Protection Engineering

|

2011

|

Vol. 37, nr 2

35-51

EN

A model most often used for the description of the processes of mass transport through phase boundaries is the model of Whitman. Results of calculations obtained using this model may occasionally considerably differ from the results obtained using diffusion models. Thus an attempt has been made to correct the model proposed by Whitman. The dynamics of the processes of mass transport from a liquid phase (river water) to a solid phase (layer of material in the river bottom) has been analysed. Several equations have been derived describing the rate of absorption with a chemical reaction and periodical changes of the concentration of the analysed substance. An attempt has been made to determine the relation between the concentration gradient and concentration at the phase boundary. In dynamic conditions, the concentration gradient at the phase boundary can be approximated by means of time dependence of a linear combination of concentration, delayed concentration, and concentration derivative at the phase boundary. Analysis of the dynamics of the absorption process with the chemical reaction enabled one to derive an equation describing the stream of the substance penetrating to the inside of the solid phase. Such equations may be used to determine the error generated by the film model of Whitman for the process of mass penetration.

6

Fotoporządkowanie : alternatywna technika dla szybkich przetworników ciekłokrystalicznych

Chrzanowski M. M., Zieliński J., Nowinowski-Kruszelnicki E., Olifierczuk M.

Biuletyn Wojskowej Akademii Technicznej

|

2010

|

Vol. 59, nr 3

375-394

PL

W artykule podjęto temat wyboru techniki fotoporządkowania ciekłego kryształu. Technologia ta jest alternatywą dla najczęściej stosowanego rubbingu, czyli jednostronnego polerowania warstw poliimidowych i poliamidowych. Zarówno w przypadku rubbingu (buffingu), który ma naturę kontaktową, jak i bezkontaktowego fotoporządkowania efektem pracy jest wytworzenie anizotropii warstwy, która następnie ma za zadanie porządkować leżącą bezpośrednio na niej warstwę ciekłego kryształu (LC). Rubbing jest metodą stosunkowo prostą i tanią i jest zwyczajowo używany przy produkcji wyświetlaczy ciekłokrystalicznych. W trakcie polerowania mogą jednak powstawać uszkodzenia struktury matrycy aktywnej oraz tworzyć się pyły, które są szkodliwe zwłaszcza dla bardzo cienkich warstw ciekłego kryształu. W związku z tym poszukuje się metod alternatywnych porządkowania. Z fotoporządkowaniem mamy do czynienia wtedy, gdy anizotropia warstwy porządkującej wywołana jest przez ekspozycję materiału fotoczułego na działanie światła ultrafioletowego spolaryzowanego liniowo. Jest to proces bezkontaktowy, więc eliminuje on większość wymienionych wcześniej wad rubbingu. Ponadto umożliwia pracę na plastycznych oraz zakrzywionych podłożach. Istnieją cztery różne drogi realizacji procesu fotoporządkowania warstwy światłoczułych materiałów organicznych: 1) fotodegradacja w materiałach poliimidowych, 2) izomeryzacja cis-trans azozwiązków, 3) czysta reorientacja związków azowych w polu potencjału światła UV oraz 4) sieciowanie w strukturze polimerowej. W pracy przedstawiono krytyczny przegląd wyżej wymienionych metod. Zastosowanie fotoporządkowania badane jest pod kątem przydatności w komórkach działających w oparciu między innymi o efekty: FLC (ang. ferroelectric liquid crystal, ciekły kryształ ferroelektryczny), TN (ang. twisted nematic, skręcony nematyk), BTN (ang. bistable twisted nematic, bistabilny skręcony nematyk), VAN (ang. vertical alignment nematic, pionowa deformacja fazy nematycznej), IPS (ang. in plane switching, przełączanie w płaszczyźnie) i inne. Celem poniższej pracy jest porównanie i określenie przydatności różnych metod fotoporządkowania dla technologii przetworników ciekłokrystalicznych o dużej jasności zobrazowania.

EN

In this study we consider the topic of the photoalignment of liquid crystal material. It is an alternative method to the rubbing of polyimide and polyamide layers. Both techniques provide the anisotropy in organic layers. This anisotropy aligns overlaying liquid crystal. The rubbing is a well known and cheap technique and it's ordinarily used in liquid crystal displays manufacturing. However, it has several disadvantages, for example it produces dust particles and charges the layer. Furthermore it may damage the surface electronic. These drawbacks can be critical factors in working with thin liquid crystal layers (about 2-5 µm) for example ferroelectric and antiferroelectric Lcs. Heterogeneity at the microscopic level can be noxious especially in the area of some types of liquid crystals and in manufacturing high-resolution liquid crystal displays (over 40”) and state-of-art projector display systems. In order to omit these problems, very important for the future work is to develop a new, universal and high-performance, technique for liquid crystal aligning. One of alternatives is photoalignment. It eliminates many of drawbacks mentioned above. This technique uses light (generally linearly polarized ultraviolet light, LPUV) to produce anisotropy in photodephiniable materials. In addition this technique allows us to obtain layers on plastic or curve surfaces. We can distinguish four main photoalignment mechanisms (including photo-chemical reactions and photo-physical changes): (1) cis-trans isomerization in azo-structures: pure dyes films, polymers containing azo-groups and monolayers of azodyes, (2) pure reorientation of photo-chemically stable azo-dyes molecules, (3) photodegradation in polyimide materials and (4) crosslinking in cinnamoyl and coumarin side-chain polymers. There are some modifications among these procedures, such as polymerizable azo-dyes or Langmuir-Blodgett films. Photoalignment is investigated in diverse liquid crystal modes, such as: twisted nematic (TN), vertical alignment (VAN), in plane switching (IPS) and many others. In this paper we describe all of the photoalignment mechanisms mentioned above and compare their ability to align the liquid crystal material in the field of application in various modes of LC cells.

7

Modeling of articular cartilage replacement materials

Stoffel M., Weichert D., Muller-Rath R.

Archives of Mechanics

|

2009

|

Vol. 61, nr 1

69-87

EN

The development of replacement material for human articular cartilage exhibiting similar mechanical properties as the native tissue is a problem of high actuality in biomeclicine. In the present work a new condensed collagen material is investigated. The study aims at developing a mechanical model especially adapted to this particular collagen material. For this purpose, a viscoelastic-diffusion (VED) model is proposed, accounting for two different diffusion evolutions assumed. Moreover, the need for a gradient material description is discussed in order to cover fabrication influences leading to a variable Young's modulus for the material. On this background, a phe-nornenological law is presented to predict deformation-dependent diffusion behavior and internal reaction forces. Furthermore, the present approach allows a practible identification of diffusion parameters. The theoretical model is implemented into a finite element code and parameters are identified by tension tests. The simulation results are validated experimentally.

8

Model transportu i dyfuzji w aspekcie zarządzania środowiskiem naturalnym

Przybycin Z.

Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska

|

2006

|

z. 35

149-157

PL

W artykule podjęto próbę zastosowania tez modelu transportu i dyfuzji do sterowania jakością powietrza atmosferycznego. Proponowany model umożliwia identyfikację sprawców zanieczyszczeń powietrza, a tym samym przypisanie im ekologicznych negatywnych skutków ich działalności. Zagadnienie sterowania jakością powietrza atmosferycznego sprowadzono do zagadnienia programowania liniowego.

EN

The paper presents the attemption to seploy so-named transportation and diffusion model to control of quality of the atmospheric air. The proposed model offers the possibility to identify pollution originators and to attribute them their activity results. The atmospheric air control problems comes to linear programming issues.

9

Modeling of light transmittance measurement in a finite layer of whole blood - a collimated transmittance problem in Monte carlo simulation and diffusion model

Mroczka J., Szczepanowski R.

Optica Applicata

|

2005

|

Vol. 35, nr 2

311-331

EN

The paper presents the modeling of transmittance measurement in a finite layer of whole blood. To describe light propagation in whole blood medium, a Monte Carlo simulation was used. The propagation of light in whole blood medium in the model required the assumption of photon transport approximation in highly scattering media. We have analyzed collimated-diffuse transmittance, which depended on the technique of free path length simulation. The Monte Carlo simulation was compared with the diffusion model designed for a finite incident light beam and available measurement data of whole blood optical density. The research revealed that discrepancy between the models discussed may be attributed to inaccuracy of the diffusion model due to an increase of anisotropic radiance under the thin sample conditions. Moreover, comparison of the Monte Carlo simulation versus measurement data showed that adding off-sets enabled agreement between them for hematocrit up to 60-70%, which is sufficient for many applications in oximetry. In fact, discrepancy in the Monte Carlo simulation, requiring off-sets to fit measurement data, most likely originates from measurement problem, such as divergence of light source or perturbation of light beam.

10

Nutrient flux fuels the summer primary productivity in the oligotrophic waters of the Gulf of Aqaba, Red Sea

Badran M. I., Rasheed M., Manasrah R., Al-Najjar T.

Oceanologia

|

2005

|

No. 47 (1)

47-60

EN

The thermohaline characteristics of the Gulf of Aqaba, Red Sea, depict a well-defined seasonal pattern of winter mixing from December to April and summer stratification from May to November. This thermohaline structure is a major controlling factor of the nutrient, chlorophyll a and primary productivity seasonal cycles. The nitrate and chlorophyll a concentration records generated down to 200 m at a vertical resolution of 25 m - weekly during 1994, 1995 and every two weeks from April 1997 through to December 2000 - are employed to assess the nitrogen flux across the summer thermocline of the Gulf of Aqaba. The flux calculations are based on a simple diffusion model that incorporates the physical stress eddy diffusivity factor Kz and a biological stress factor k. Both Kz and k are calculated using the Michaelis-Menten equation and the nitrate concentration gradient. The total nitrate flux of the Gulf of Aqaba during the seven summer months (May-November) is estimated at 0.52 mole N m-2. In relation to established primary productivity values (75.5 g C m-2 (May November)-1) and the generated chlorophyll a records, this yields an f fraction of new to total primary production of 0.50. This relatively high f value is discussed with respect to the geophysical characteristics of the Gulf of Aqaba and similar oceanic basins. The remaining 50% is accounted for by cross-sectional flow from the relatively nutrient-rich coral reef coastal habitat and rapid recycling, triggered by high irradiance and water temperature.

11

Double diffusivity from the distributed source on surface of a layer

Chaplia Y., Chernukha O.

Journal of Applied Computer Science

|

2004

|

Vol. 12, nr 2

45-64

EN

The solution has been obtained for the space problem of admixture diffusion by two ways allowing for mass transitions between the states under action of a constant circular source on a layer surface. This case has been compared with the solutions of the particular diffusion problems in a body with traps and diffusion in a medium with effective characteristics.

12

Dependence of adsorption/diffusion processes in porous media on bulk and surface permeabilities

Albers B.

Archives of Mechanics

|

2001

|

Vol. 53, nr 4-5

289-306

EN

The paper contains a macroscopic continuum model for adsorption in porous materials (B. Albers [1, 2]) which is an extension of the model for porous bodies by K. Wilmański [7] on mass exchange processes. We consider the flow of a fluid/adsorbate mixture through channels of a solid component. The fluid serves as carrier for an adsorbate whose mass balance equation contains a source term. Due to low adsorbate concentration we deal with a physical adsorption process which means that particles of the adsorbate stick to the skeleton due to weak van der Waals forces. The model contains two different permeability parameters whose nature is completely different: The first one, the usual bulk permeability coefficient, describes the resistance of the skeleton to the flow of the fluid/adsorbate mixture. The second one describes the surface resistance to the outflow of the mixture from the solid. This work shows within a simple example the range of these parameters and the dependence of adsorption/diffusion on them.

13

Porównanie spiekania ceramiki konwencjonalnej i nanostrukturalnej.

Rzeszotarska J., Ranachowski J.

Szkło i Ceramika

|

2000

|

R. 51, nr 3

7-13