			Home About us Contact

Analysis Units (analysis + unit)

Distribution by Scientific Domains

Life Sciences	75%

Selected Abstracts

Covariance-oriented qualitative and quantitative calibration analysis for multistage batch processes

THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, Issue 3 2009
Chunhui Zhao
Abstract An improved stage-specific multivariate calibration scheme is developed for multistage batch processes based on the covariance analysis unit. First, the process duration is automatically and properly divided into different stages, which reveals the changes of quality-related process correlation characteristics. The concept of stage-representative average process behaviour is then introduced, which is comprehensibly realized by averaging all covariance patterns within the same stage based on different weights. In this way, it stacks the cumulative effects of process variations on quality within each stage and meanwhile considers their time-varying characteristics. Subsequently, covariance-oriented OSC and variable selection are effectively combined, which can simplify the calibration model structure and enhance the causal relationship between predictors and quality by excluding the redundant latent factors and input variables. Finally, stage-representative PLS regression models are developed focusing on the critical-to-quality stages for online quality prediction. It shows that a complete multistage calibration solution is readily achieved from an "overall" stage perspective by smartly making use of covariance. The illustration study to injection molding shows the effectiveness of the proposed method for improving process comprehension and quality prediction. On a mis au point un schéma amélioré d'étalonnage multivarié pour des procédés discontinus multi-étapes, spécifique à chaque étape, basé sur l'analyse des covariances sans tenir compte de chaque variable de procédé. En premier lieu, la durée de procédé est automatiquement et correctement divisée en quatre étapes différentes, révélant le changement des caractéristiques de corrélation de procédé lié à la qualité. Le concept de comportement de procédé moyen représentatif d'une étape est alors introduit, ce qui se réalise de façon complète en moyennant tous les profils de covariance d'une étape donnée avec différentes pondérations. De cette façon, on tient compte dans la qualité des effets cumulatifs des variations de procédé spécifiques à chaque étape et considère de ce fait leurs caractéristiques évolutives. Par conséquent, l'OSC orienté vers la covariance et la technique de sélection des variables sont combinés efficacement afin de simplifier la structure du modèle d'étalonnage et améliorer la relation causale en excluant les facteurs latents redondants ou les variables d'entrée. Enfin, des modèles de régression de type PLS représentatifs des étapes sont mis au point, mettant l'accent sur les étapes critiques en terme de qualité pour la prédiction de la qualité en ligne. Cela montre qu'une solution complète d'étalonnage multi-étapes peut se réaliser parfaitement en utilisant bien la covariance. L'illustration dans le cas du moulage par injection montre l'efficacité de la méthode proposée pour la compréhension du procédé et l'amélioration des prédictions. [source]

Expanding the Global Network of Protected Areas to Save the Imperiled Mediterranean Biome

CONSERVATION BIOLOGY, Issue 1 2009
EMMA C. UNDERWOOD
análisis de disparidad; áreas protegidas; biodiversidad; ecosistemas Mediterráneos; pérdida de hábitat Abstract:,Global goals established by the Convention on Biological Diversity stipulate that 10% of the world's ecological regions must be effectively conserved by 2010. To meet that goal for the mediterranean biome, at least 5% more land must be formally protected over the next few years. Although global assessments identify the mediterranean biome as a priority, without biologically meaningful analysis units, finer-resolution data, and corresponding prioritization analysis, future conservation investments could lead to more area being protected without increasing the representation of unique mediterranean ecosystems. We used standardized analysis units and six potential natural vegetation types stratified by 3 elevation zones in a global gap analysis that systematically explored conservation priorities across the mediterranean biome. The highest levels of protection were in Australia, South Africa, and California-Baja California (from 9,11%), and the lowest levels of protection were in Chile and the mediterranean Basin (<1%). Protection was skewed to montane elevations in three out of five regions. Across the biome only one of the six vegetation types,mediterranean shrubland,exceeded 10% protection. The remaining vegetation types,grassland, scrub, succulent dominated, woodland, and forest,each had <3% protection. To guard against biases in future protection efforts and ensure the protection of species characteristic of the mediterranean biome, we identified biodiversity assemblages with <10% protection and subject to >30% conversion and suggest that these assemblages be elevated to high-priority status in future conservation efforts. Resumen:,Las metas globales establecidas por la Convención sobre Diversidad Biológica estipulan que 10% de las regiones ecológicas del mundo deberán estar conservadas efectivamente en 2010. Para alcanzar esa meta en el bioma mediterráneo, por lo menos 5% más de superficie debe estar protegida formalmente en los próximos años. Aunque las evaluaciones globales identifican al bioma mediterráneo como una prioridad, sin unidades de análisis biológicamente significativas, datos de resolución más fina y los correspondientes análisis de priorización, las inversiones futuras en conservación pudieran conducir a la protección de más superficie sin incrementar la representación de los ecosistemas mediterráneos únicos. Utilizamos unidades de análisis estandarizadas y seis tipos potenciales de vegetación natural estratificados en tres zonas de elevación en un análisis global de disparidad que exploró sistemáticamente las prioridades de conservación en el bioma mediterráneo. Los niveles de protección más altos se localizaron en Australia, África del Sur y California-Baja California (de 9,11%) y los niveles de protección más bajos se localizaron en Chile y la Cuenca del mediterráneo (<1%). La protección estaba sesgada hacia elevaciones altas en tres de las cinco regiones. En todo el bioma, solo uno de los seis tipos de vegetación,matorral mediterráneo,excedió 10% de protección. Los tipos de vegetación restantes,pastizal, matorral, dominio de suculentas, y bosques,tenían <3% de protección cada uno. Para evitar sesgos en futuros esfuerzos de protección y asegurar la protección de especies características del bioma mediterráneo, identificamos ensambles de biodiversidad con <10% de protección y sujetos a >30% de conversión y sugerimos que estos ensambles sean elevados a un estatus de alta prioridad en esfuerzos de conservación en el futuro. [source]

Scale dependence of spatial patterns and cartography on the detection of landscape change: relationships with species' perception

ECOGRAPHY, Issue 4 2002
Susana Suárez-Seoane
This paper analyses how landscape pattern detection changes when different spatial and temporal scales and several levels of detail of the cartography are used to describe a landscape affected by land abandonment in northern Spain. In order to integrate landscape composition and structure at different temporal and spatial scales in the same framework, a multiple correspondence factorial analysis was ran for each typology of landscape units. Annual rates of change and scale dependencies were calculated for each typology from the Euclidean distances in the factorial space. Finally, the potential assessment of habitat utilisation by species with different landscape perception and movement capacity was modelled for the range of typologies. The amount of variance explained by the factorial analysis decreased with the complexity of the typology. Annual rates of change appeared different according to the time span and the detail of the landscape unit typology used. For all typologies, changes were faster during 1983,95, a period characterised by massive land abandonment. However, when the whole period (1956,95) was considered, annual changes were much lower, showing differences between typologies. As a general trend, the variance of the mean annual change decreased with the size of the analysis units. In response to land abandonment, different scale dependencies were found for different levels of detail of the cartography. Coarser typologies are suitable when analysing highly mobile species. However, species with small movement capacity or with a preference for homogeneous habitats perceive more detail in landscape. In this case, a detailed typology is more appropriate. [source]